Advertisements
Feeds:
Posts
Comments

Archive for the ‘Artificial Intelligence in Medicine – Applications in Therapeutics’ Category


Tweets, Pictures and Retweets at 18th Annual Cancer Research Symposium – Machine Learning and Cancer, June 14, 2019, MIT by @pharma_BI and @AVIVA1950 for #KIsymposium PharmaceuticalIntelligence.com and Social Media

 

Pictures taken in Real Time

 

Notification from Twitter.com on June 14, 2019 and in the 24 hours following the symposium

 

     liked your Tweet

    3 hours ago

  1.  Retweeted your Tweet

    5 hours ago

    1 other Retweet

  2.  liked your Tweets

    11 hours ago

    2 other likes

     and  liked a Tweet you were mentioned in

    11 hours ago

     liked your reply

    12 hours ago

  3. Replying to 

    It was an incredibly touching and “metzamrer” surprise to meet you at MIT

  4.  liked your Tweets

    13 hours ago

    3 other likes

     liked your reply

    15 hours ago

    Amazing event @avivregev @reginabarzilay 2pharma_BI Breakthrough in

     and  liked a Tweet you were mentioned in

    17 hours ago

  5. ‘s machine learning tool characterizes proteins, which are biomarkers of disease development and progression. Scientists can know more about their relationship to specific diseases and can interview earlier and precisely. ,

  6. learning and are undergoing dramatic changes and hold great promise for cancer research, diagnostics, and therapeutics. @KIinstitute by

     liked your Tweet

    Jun 16

     Retweeted your Retweet

    Jun 16

     liked your Retweet

    Jun 15

     Retweeted your Tweet

    Jun 15

     Retweeted your Tweet

    Jun 15

     Retweeted your Retweet

    Jun 15

     and 3 others liked your reply

    Jun 15

     and  Retweeted your Tweet

    Jun 14

     and  liked your Tweet

    Jun 14

     and  Retweeted your Tweet

    Jun 14

     liked your Tweet

    Jun 14

  7.  liked your Tweets

    Jun 14

    2 other likes

     liked your Tweet

    Jun 14

     Retweeted your Retweet

    Jun 14

     liked your Tweet

    Jun 14

     and  Retweeted your Tweet

    Jun 14

     liked your Tweet

    Jun 14

     liked your Tweet

    Jun 14

  8. identification in the will depend on highly

  9.  liked your Tweets

    Jun 14

    2 other likes

     Retweeted your Tweet

    Jun 14

     liked your Tweet

    Jun 14

     and 3 others liked your reply

    Jun 14

     liked your Retweet

    Jun 14

  10. this needed to be done a long time ago

     Retweeted your Tweet

    Jun 14

     and  Retweeted your reply

    Jun 14

     liked your Tweet

    Jun 14

     liked your reply

    Jun 14

     Retweeted your reply

    Jun 14

 

Tweets by @pharma_BI and by @AVIVA1950

&

Retweets and replies by @pharma_BI and @AVIVA1950

eProceedings 18th Symposium 2019 covered in Amazing event, Keynote best talks @avivregev ’er @reginabarzelay

  1. Top lectures by @reginabarzilay @avivaregev

  2. eProceeding 2019 Koch Institute Symposium – 18th Annual Cancer Research Symposium – Machine Learning and Cancer, June 14, 2019, 8:00 AM-5:00 PMET MIT Kresge Auditorium, 48 Massachusetts Ave, Cambridge, MA via

  1.   Retweeted

    eProceedings 18th Symposium 2019 covered in Amazing event, Keynote best talks @avivregev ’er @reginabarzelay

  2. Top lectures by @reginabarzilay @avivaregev

  3. eProceeding 2019 Koch Institute Symposium – 18th Annual Cancer Research Symposium – Machine Learning and Cancer, June 14, 2019, 8:00 AM-5:00 PMET MIT Kresge Auditorium, 48 Massachusetts Ave, Cambridge, MA via

  4. eProceedings & eProceeding 2019 Koch Institute Symposium – 18th Annual Cancer Research Symposium – Machine Learning and Cancer, June 14, 2019, 8:00 AM-5:00 PMET MIT Kresge Auditorium, 48 Massachusetts Ave, Cambridge, MA via

  5.   Retweeted
  6.   Retweeted

    Einstein, Curie, Bohr, Planck, Heisenberg, Schrödinger… was this the greatest meeting of minds, ever? Some of the world’s most notable physicists participated in the 1927 Solvay Conference. In fact, 17 of the 29 scientists attending were or became Laureates.

  7.   Retweeted

    identification in the will depend on highly

  8. eProceeding 2019 Koch Institute Symposium – 18th Annual Cancer Research Symposium – Machine Learning and Cancer, June 14, 2019, 8:00 AM-5:00 PMET MIT Kresge Auditorium, Cambridge, MA via

 

Advertisements

Read Full Post »


Reported by Dror Nir, PhD

Deep Learning–Assisted Diagnosis of Cerebral Aneurysms Using the HeadXNet Model

Allison Park, BA1Chris Chute, BS1Pranav Rajpurkar, MS1;  et al, Original Investigation, Health Informatics, June 7, 2019, JAMA Netw Open. 2019;2(6):e195600. doi:10.1001/jamanetworkopen.2019.5600

Key Points

Question  How does augmentation with a deep learning segmentation model influence the performance of clinicians in identifying intracranial aneurysms from computed tomographic angiography examinations?

Findings  In this diagnostic study of intracranial aneurysms, a test set of 115 examinations was reviewed once with model augmentation and once without in a randomized order by 8 clinicians. The clinicians showed significant increases in sensitivity, accuracy, and interrater agreement when augmented with neural network model–generated segmentations.

Meaning  This study suggests that the performance of clinicians in the detection of intracranial aneurysms can be improved by augmentation using deep learning segmentation models.

 

Abstract

Importance  Deep learning has the potential to augment clinician performance in medical imaging interpretation and reduce time to diagnosis through automated segmentation. Few studies to date have explored this topic.

Objective  To develop and apply a neural network segmentation model (the HeadXNet model) capable of generating precise voxel-by-voxel predictions of intracranial aneurysms on head computed tomographic angiography (CTA) imaging to augment clinicians’ intracranial aneurysm diagnostic performance.

Design, Setting, and Participants  In this diagnostic study, a 3-dimensional convolutional neural network architecture was developed using a training set of 611 head CTA examinations to generate aneurysm segmentations. Segmentation outputs from this support model on a test set of 115 examinations were provided to clinicians. Between August 13, 2018, and October 4, 2018, 8 clinicians diagnosed the presence of aneurysm on the test set, both with and without model augmentation, in a crossover design using randomized order and a 14-day washout period. Head and neck examinations performed between January 3, 2003, and May 31, 2017, at a single academic medical center were used to train, validate, and test the model. Examinations positive for aneurysm had at least 1 clinically significant, nonruptured intracranial aneurysm. Examinations with hemorrhage, ruptured aneurysm, posttraumatic or infectious pseudoaneurysm, arteriovenous malformation, surgical clips, coils, catheters, or other surgical hardware were excluded. All other CTA examinations were considered controls.

Main Outcomes and Measures  Sensitivity, specificity, accuracy, time, and interrater agreement were measured. Metrics for clinician performance with and without model augmentation were compared.

Results  The data set contained 818 examinations from 662 unique patients with 328 CTA examinations (40.1%) containing at least 1 intracranial aneurysm and 490 examinations (59.9%) without intracranial aneurysms. The 8 clinicians reading the test set ranged in experience from 2 to 12 years. Augmenting clinicians with artificial intelligence–produced segmentation predictions resulted in clinicians achieving statistically significant improvements in sensitivity, accuracy, and interrater agreement when compared with no augmentation. The clinicians’ mean sensitivity increased by 0.059 (95% CI, 0.028-0.091; adjusted P = .01), mean accuracy increased by 0.038 (95% CI, 0.014-0.062; adjusted P = .02), and mean interrater agreement (Fleiss κ) increased by 0.060, from 0.799 to 0.859 (adjusted P = .05). There was no statistically significant change in mean specificity (0.016; 95% CI, −0.010 to 0.041; adjusted P = .16) and time to diagnosis (5.71 seconds; 95% CI, 7.22-18.63 seconds; adjusted P = .19).

Conclusions and Relevance  The deep learning model developed successfully detected clinically significant intracranial aneurysms on CTA. This suggests that integration of an artificial intelligence–assisted diagnostic model may augment clinician performance with dependable and accurate predictions and thereby optimize patient care.

Introduction

Diagnosis of unruptured aneurysms is a critically important clinical task: intracranial aneurysms occur in 1% to 3% of the population and account for more than 80% of nontraumatic life-threatening subarachnoid hemorrhages.1 Computed tomographic angiography (CTA) is the primary, minimally invasive imaging modality currently used for diagnosis, surveillance, and presurgical planning of intracranial aneurysms,2,3but interpretation is time consuming even for subspecialty-trained neuroradiologists. Low interrater agreement poses an additional challenge for reliable diagnosis.47

Deep learning has recently shown significant potential in accurately performing diagnostic tasks on medical imaging.8 Specifically, convolutional neural networks (CNNs) have demonstrated excellent performance on a range of visual tasks, including medical image analysis.9 Moreover, the ability of deep learning systems to augment clinician workflow remains relatively unexplored.10 The development of an accurate deep learning model to help clinicians reliably identify clinically significant aneurysms in CTA has the potential to provide radiologists, neurosurgeons, and other clinicians an easily accessible and immediately applicable diagnostic support tool.

In this study, a deep learning model to automatically detect intracranial aneurysms on CTA and produce segmentations specifying regions of interest was developed to assist clinicians in the interpretation of CTA examinations for the diagnosis of intracranial aneurysms. Sensitivity, specificity, accuracy, time to diagnosis, and interrater agreement for clinicians with and without model augmentation were compared.

Methods

The Stanford University institutional review board approved this study. Owing to the retrospective nature of the study, patient consent or assent was waived. The Standards for Reporting of Diagnostic Accuracy (STARD) reporting guideline was used for the reporting of this study.

Data

A total of 9455 consecutive CTA examination reports of the head or head and neck performed between January 3, 2003, and May 31, 2017, at Stanford University Medical Center were retrospectively reviewed. Examinations with parenchymal hemorrhage, subarachnoid hemorrhage, posttraumatic or infectious pseudoaneurysm, arteriovenous malformation, ischemic stroke, nonspecific or chronic vascular findings such as intracranial atherosclerosis or other vasculopathies, surgical clips, coils, catheters, or other surgical hardware were excluded. Examinations of injuries that resulted from trauma or contained images degraded by motion were also excluded on visual review by a board-certified neuroradiologist with 12 years of experience. Examinations with nonruptured clinically significant aneurysms (>3 mm) were included.11

Radiologist Annotations

The reference standard for all examinations in the test set was determined by a board-certified neuroradiologist at a large academic practice with 12 years of experience who determined the presence of aneurysm by review of the original radiology report, double review of the CTA examination, and further confirmation of the aneurysm by diagnostic cerebral angiograms, if available. The neuroradiologist had access to all of the Digital Imaging and Communications in Medicine (DICOM) series, original reports, and clinical histories, as well as previous and follow-up examinations during interpretation to establish the best possible reference standard for the labels. For each of the aneurysm examinations, the radiologist also identified the location of each of the aneurysms. Using the open-source annotation software ITK-SNAP,12 the identified aneurysms were manually segmented on each slice.

Model Development

In this study, we developed a 3-dimensional (3-D) CNN called HeadXNet for segmentation of intracranial aneurysms from CT scans. Neural networks are functions with parameters structured as a sequence of layers to learn different levels of abstraction. Convolutional neural networks are a type of neural network designed to process image data, and 3-D CNNs are particularly well suited to handle sequences of images, or volumes.

HeadXNet is a CNN with an encoder-decoder structure (eFigure 1 in the Supplement), where the encoder maps a volume to an abstract low-resolution encoding, and the decoder expands this encoding to a full-resolution segmentation volume. The segmentation volume is of the same size as the corresponding study and specifies the probability of aneurysm for each voxel, which is the atomic unit of a 3-D volume, analogous to a pixel in a 2-D image. The encoder is adapted from a 50-layer SE-ResNeXt network,1315and the decoder is a sequence of 3 × 3 transposed convolutions. Similar to UNet,16 skip connections are used in 3 layers of the encoder to transmit outputs directly to the decoder. The encoder was pretrained on the Kinetics-600 data set,17 a large collection of YouTube videos labeled with human actions; after pretraining the encoder, the final 3 convolutional blocks and the 600-way softmax output layer were removed. In their place, an atrous spatial pyramid pooling18 layer and the decoder were added.

Training Procedure

Subvolumes of 16 slices were randomly sampled from volumes during training. The data set was preprocessed to find contours of the skull, and each volume was cropped around the skull in the axial plane before resizing each slice to 208 × 208 pixels. The slices were then cropped to 192 × 192 pixels (using random crops during training and centered crops during testing), resulting in a final input of size 16 × 192 × 192 per example; the same transformations were applied to the segmentation label. The segmentation output was trained to optimize a weighted combination of the voxelwise binary cross-entropy and Dice losses.19

Before reaching the model, inputs were clipped to [−300, 700] Hounsfield units, normalized to [−1, 1], and zero-centered. The model was trained on 3 Titan Xp graphical processing units (GPUs) (NVIDIA) using a minibatch of 2 examples per GPU. The parameters of the model were optimized using a stochastic gradient descent optimizer with momentum of 0.9 and a peak learning rate of 0.1 for randomly initialized weights and 0.01 for pretrained weights. The learning rate was scheduled with a linear warm-up from 0 to the peak learning rate for 10 000 iterations, followed by cosine annealing20 over 300 000 iterations. Additionally, the learning rate was fixed at 0 for the first 10 000 iterations for the pretrained encoder. For regularization, L2 weight decay of 0.001 was added to the loss for all trainable parameters and stochastic depth dropout21 was used in the encoder blocks. Standard dropout was not used.

To control for class imbalance, 3 methods were used. First, an auxiliary loss was added after the encoder and focal loss was used to encourage larger parameter updates on misclassified positive examples. Second, abnormal training examples were sampled more frequently than normal examples such that abnormal examples made up 30% of training iterations. Third, parameters of the decoder were not updated on training iterations where the segmentation label consisted of purely background (normal) voxels.

To produce a segmentation prediction for the entire volume, the segmentation outputs for sequential 16-slice subvolumes were simply concatenated. If the number of slices was not divisible by 16, the last input volume was padded with 0s and the corresponding output volume was truncated back to the original size.

Study Design

We performed a diagnostic accuracy study comparing performance metrics of clinicians with and without model augmentation. Each of the 8 clinicians participating in the study diagnosed a test set of 115 examinations, once with and once without assistance of the model. The clinicians were blinded to the original reports, clinical histories, and follow-up imaging examinations. Using a crossover design, the clinicians were randomly and equally divided into 2 groups. Within each group, examinations were sorted in a fixed random order for half of the group and sorted in reverse order for the other half. Group 1 first read the examinations without model augmentation, and group 2 first read the examinations with model augmentation. After a washout period of 14 days, the augmentation arrangement was reversed such that group 1 performed reads with model augmentation and group 2 read the examinations without model augmentation (Figure 1A).

Clinicians were instructed to assign a binary label for the presence or absence of at least 1 clinically significant aneurysm, defined as having a diameter greater than 3 mm. Clinicians read alone in a diagnostic reading room, all using the same high-definition monitor (3840 × 2160 pixels) displaying CTA examinations on a standard open-source DICOM viewer (Horos).22 Clinicians entered their labels into a data entry software application that automatically logged the time difference between labeling of the previous examination and the current examination.

When reading with model augmentation, clinicians were provided the model’s predictions in the form of region of interest (ROI) segmentations directly overlaid on top of CTA examinations. To ensure an image display interface that was familiar to all clinicians, the model’s predictions were presented as ROIs in a standard DICOM viewing software. At every voxel where the model predicted a probability greater than 0.5, readers saw a semiopaque red overlay on the axial, sagittal, and coronal series (Figure 1C). Readers had access to the ROIs immediately on loading the examinations, and the ROIs could be toggled off to reveal the unaltered CTA images (Figure 1B). The red overlays were the only indication that was given whether a particular CTA examination had been predicted by the model to contain an aneurysm. Given these model results, readers had the option to take it into consideration or disregard it based on clinical judgment. When readers performed diagnoses without augmentation, no ROIs were present on any of the examinations. Otherwise, the diagnostic tools were identical for augmented and nonaugmented reads.

 

Statistical Analysis

On the binary task of determining whether an examination contained an aneurysm, sensitivity, specificity, and accuracy were used to assess the performance of clinicians with and without model augmentation. Sensitivity denotes the number of true-positive results over total aneurysm-positive cases, specificity denotes the number of true-negative results over total aneurysm-negative cases, and accuracy denotes the number of true-positive and true-negative results over all test cases. The microaverage of these statistics across all clinicians was also computed by measuring each statistic pertaining to the total number of true-positive, false-negative, and false-positive results. In addition, to convert the models’ segmentation output of the model into a binary prediction, a prediction was considered positive if the model predicted at least 1 voxel as belonging to an aneurysm and negative otherwise. The 95% Wilson score confidence intervals were used to assess the variability in the estimates for sensitivity, specificity, and accuracy.23

To assess whether the clinicians achieved significant increases in performance with model augmentation, a 1-tailed t test was performed on the differences in sensitivity, specificity, and accuracy across all 8 clinicians. To determine the robustness of the findings and whether results were due to inclusion of the resident radiologist and neurosurgeon, we performed a sensitivity analysis: we computed the t test on the differences in sensitivity, specificity, and accuracy across board-certified radiologists only.

The average time to diagnosis for the clinicians with and without augmentation was computed as the difference between the mean entry times into the spreadsheet of consecutive diagnoses; 95% t score confidence intervals were used to assess the variability in the estimates. To account for interruptions in the clinical read or time logging errors, the 5 longest and 5 shortest time to diagnosis for each clinician in each reading were excluded. To assess whether model augmentation significantly decreased the time to diagnosis, a 1-tailed t test was performed on the difference in average time with and without augmentation across all 8 clinicians.

The interrater agreement of clinicians and for the radiologist subset was computed using the exact Fleiss κ.24 To assess whether model augmentation increased interrater agreement, a 1-tailed permutation test was performed on the difference between the interrater agreement of clinicians on the test set with and without augmentation. The permutation procedure consisted of randomly swapping clinician annotations with and without augmentation so that a random subset of the test set that had previously been labeled as read with augmentation was now labeled as being read without augmentation, and vice versa; the exact Fleiss κ values (and the difference) were computed on the test set with permuted labels. This permutation procedure was repeated 10 000 times to generate the null distribution of the Fleiss κ difference (the interrater agreement of clinician annotations with augmentation is not higher than without augmentation) and the unadjusted value calculated as the proportion of Fleiss κ differences that were higher than the observed Fleiss κ difference.

To control the familywise error rate, the Benjamini-Hochberg correction was applied to account for multiple hypothesis testing; a Benjamini-Hochberg–adjusted P ≤ .05 indicated statistical significance. All tests were 1-tailed.25

Results

The data set contained 818 examinations from 662 unique patients with 328 CTA examinations (40.1%) containing at least 1 intracranial aneurysm and 490 examinations (59.9%) without intracranial aneurysms (Figure 2). Of the 328 aneurysm cases, 20 cases from 15 unique patients contained 2 or more aneurysms. One hundred forty-eight aneurysm cases contained aneurysms between 3 mm and 7 mm, 108 cases had aneurysms between 7 mm and 12 mm, 61 cases had aneurysms between 12 mm and 24 mm, and 11 cases had aneurysms 24 mm or greater. The location of the aneurysms varied according to the following distribution: 99 were located in the internal carotid artery, 78 were in the middle cerebral artery, 50 were cavernous internal carotid artery aneurysms, 44 were basilar tip aneurysms, 41 were in the anterior communicating artery, 18 were in the posterior communicating artery, 16 were in the vertebrobasilar system, and 12 were in the anterior cerebral artery. All examinations were performed either on a GE Discovery, GE LightSpeed, GE Revolution, Siemens Definition, Siemens Sensation, or a Siemens Force scanner, with slice thicknesses of 1.0 mm or 1.25 mm, using standard clinical protocols for head angiogram or head/neck angiogram. There was no difference between the protocols or slice thicknesses between the aneurysm and nonaneurysm examinations. For this study, axial series were extracted from each examination and a segmentation label was produced on every axial slice containing an aneurysm. The number of images per examination ranged from 113 to 802 (mean [SD], 373 [157]).

The examinations were split into a training set of 611 examinations (494 patients; mean [SD] age, 55.8 [18.1] years; 372 [60.9%] female) used to train the model, a development set of 92 examinations (86 patients; mean [SD] age, 61.6 [16.7] years; 59 [64.1%] female) used for model selection, and a test set of 115 examinations (82 patients; mean [SD] age, 57.8 [18.3] years; 74 [64.4%] female) to evaluate the performance of the clinicians when augmented with the model (Figure 2).

Using stratified random sampling, the development and test sets were formed to include 50% aneurysm examinations and 50% normal examinations; the remaining examinations composed the training set, of which 36.5% were aneurysm examinations. Forty-three patients had multiple examinations in the data set due to examinations performed for follow-up of the aneurysm. To account for these repeat patients, examinations were split so that there was no patient overlap between the different sets. Figure 2 contains pathology and patient demographic characteristics for each set.

A total of 8 clinicians, including 6 board-certified practicing radiologists, 1 practicing neurosurgeon, and 1 radiology resident, participated as readers in the study. The radiologists’ years of experience ranged from 3 to 12 years, the neurosurgeon had 2 years of experience as attending, and the resident was in the second year of training at Stanford University Medical Center. Groups 1 and 2 consisted of 3 radiologists each; the resident and neurosurgeon were both in group 1. None of the clinicians were involved in establishing the reference standard for the examinations.

Without augmentation, clinicians achieved a microaveraged sensitivity of 0.831 (95% CI, 0.794-0.862), specificity of 0.960 (95% CI, 0.937-0.974), and an accuracy of 0.893 (95% CI, 0.872-0.912). With augmentation, the clinicians achieved a microaveraged sensitivity of 0.890 (95% CI, 0.858-0.915), specificity of 0.975 (95% CI, 0.957-0.986), and an accuracy of 0.932 (95% CI, 0.913-0.946). The underlying model had a sensitivity of 0.949 (95% CI, 0.861-0.983), specificity of 0.661 (95% CI, 0.530-0.771), and accuracy of 0.809 (95% CI, 0.727-0.870). The performances of the model, individual clinicians, and their microaverages are reported in eTable 1 in the Supplement.

 

With augmentation, there was a statistically significant increase in the mean sensitivity (0.059; 95% CI, 0.028-0.091; adjusted P = .01) and mean accuracy (0.038; 95% CI, 0.014-0.062; adjusted P = .02) of the clinicians as a group. There was no statistically significant change in mean specificity (0.016; 95% CI, −0.010 to 0.041; adjusted P = .16). Performance improvements across clinicians are detailed in the Table, and individual clinician improvement in Figure 3.

Individual performances with and without model augmentation are shown in eTable 1 in the Supplement. The sensitivity analysis confirmed that even among board-certified radiologists, there was a statistically significant increase in mean sensitivity (0.059; 95% CI, 0.013-0.105; adjusted P = .04) and accuracy (0.036; 95% CI, 0.001-0.072; adjusted P = .05). Performance improvements of board-certified radiologists as a group are shown in eTable 2 in the Supplement.

 

The mean diagnosis time per examination without augmentation microaveraged across clinicians was 57.04 seconds (95% CI, 54.58-59.50 seconds). The times for individual clinicians are detailed in eTable 3 in the Supplement, and individual time changes are shown in eFigure 2 in the Supplement.

 

With augmentation, there was no statistically significant decrease in mean diagnosis time (5.71 seconds; 95% CI, −7.22 to 18.63 seconds; adjusted P = .19). The model took a mean of 7.58 seconds (95% CI, 6.92-8.25 seconds) to process an examination and output its segmentation map.Confusion matrices, which are tables reporting true- and false-positive results and true- and false-negative results of each clinician with and without model augmentation, are shown in eTable 4 in the Supplement.

There was a statistically significant increase of 0.060 (adjusted P = .05) in the interrater agreement among the clinicians, with an exact Fleiss κ of 0.799 without augmentation and 0.859 with augmentation. For the board-certified radiologists, there was an increase of 0.063 in their interrater agreement, with an exact Fleiss κ of 0.783 without augmentation and 0.847 with augmentation.

Discussion

In this study, the ability of a deep learning model to augment clinician performance in detecting cerebral aneurysms using CTA was investigated with a crossover study design. With model augmentation, clinicians’ sensitivity, accuracy, and interrater agreement significantly increased. There was no statistical change in specificity and time to diagnosis.Given the potential catastrophic outcome of a missed aneurysm at risk of rupture, an automated detection tool that reliably detects and enhances clinicians’ performance is highly desirable. Aneurysm rupture is fatal in 40% of patients and leads to irreversible neurological disability in two-thirds of those who survive; therefore, an accurate and timely detection is of paramount importance. In addition to significantly improving accuracy across clinicians while interpreting CTA examinations, an automated aneurysm detection tool, such as the one presented in this study, could also be used to prioritize workflow so that those examinations more likely to be positive could receive timely expert review, potentially leading to a shorter time to treatment and more favorable outcomes.The significant variability among clinicians in the diagnosis of aneurysms has been well documented and is typically attributed to lack of experience or subspecialty neuroradiology training, complex neurovascular anatomy, or the labor-intensive nature of identifying aneurysms. Studies have shown that interrater agreement of CTA-based aneurysm detection is highly variable, with interrater reliability metrics ranging from 0.37 to 0.85,6,7,2628 and performance levels that vary depending on aneurysm size and individual radiologist experience.4,6 In addition to significantly increasing sensitivity and accuracy, augmenting clinicians with the model also significantly improved interrater reliability from 0.799 to 0.859. This implies that augmenting clinicians with varying levels of experience and specialties with models could lead to more accurate and more consistent radiological interpretations. Currently, tools to improve clinician aneurysm detection on CTA include bone subtraction,29 as well as 3-D rendering of intracranial vasculature,3032 which rely on application of contrast threshold settings to better delineate cerebral vasculature and create a 3-D–rendered reconstruction to assist aneurysm detection. However, using these tools is labor- and time-intensive for clinicians; in some institutions, this process is outsourced to a 3-D lab at additional costs. The tool developed in this study, integrated directly in a standard DICOM viewer, produces a segmentation map on a new examination in only a few seconds. If integrated into the standard workflow, this diagnostic tool could substantially decrease both cost and time to diagnosis, potentially leading to more efficient treatment and more favorable patient outcomes.Deep learning has recently shown success in various clinical image-based recognition tasks. In particular, studies have shown strong performance of 2-D CNNs in detecting intracranial hemorrhage and other acute brain findings, such as mass effect or skull fractures, on CT head examinations.3336 Recently, one study10 examined the potential role for deep learning in magnetic resonance angiogram–based detection of cerebral aneurysms, and another study37 showed that providing deep learning model predictions to clinicians when interpreting knee magnetic resonance studies increased specificity in detecting anterior cruciate ligament tears. To our knowledge, prior to this study, deep learning had not been applied to CTA, which is the first-line imaging modality for detecting cerebral aneurysms. Our results demonstrate that deep learning segmentation models may produce dependable and interpretable predictions that augment clinicians and improve their diagnostic performance. The model implemented and tested in this study significantly increased sensitivity, accuracy, and interrater reliability of clinicians with varied experience and specialties in detecting cerebral aneurysms using CTA.

Limitations

This study has limitations. First, because the study focused only on nonruptured aneurysms, model performance on aneurysm detection after aneurysm rupture, lesion recurrence after coil or surgical clipping, or aneurysms associated with arteriovenous malformations has not been investigated. Second, since examinations containing surgical hardware or devices were excluded, model performance in their presence is unknown. In a clinical environment, CTA is typically used to evaluate for many types of vascular diseases, not just for aneurysm detection. Therefore, the high prevalence of aneurysm in the test set and the clinician’s binary task could have introduced bias in interpretation. Also, this study was performed on data from a single tertiary care academic institution and may not reflect performance when applied to data from other institutions with different scanners and imaging protocols, such as different slice thicknesses.

Conclusions

A deep learning model was developed to automatically detect clinically significant intracranial aneurysms on CTA. We found that the augmentation significantly improved clinicians’ sensitivity, accuracy, and interrater reliability. Future work should investigate the performance of this model prospectively and in application of data from other institutions and hospitals.

Article Information:

Accepted for Publication: April 23, 2019.

Published: June 7, 2019. doi:10.1001/jamanetworkopen.2019.5600

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Park A et al. JAMA Network Open.

Corresponding Author: Kristen W. Yeom, MD, School of Medicine, Department of Radiology, Stanford University, 725 Welch Rd, Ste G516, Palo Alto, CA 94304 (kyeom@stanford.edu).

Author Contributions: Ms Park and Dr Yeom had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Ms Park and Messrs Chute and Rajpurkar are co–first authors. Drs Ng and Yeom are co–senior authors.

Concept and design: Park, Chute, Rajpurkar, Lou, Shpanskaya, Ni, Basu, Lungren, Ng, Yeom.

Acquisition, analysis, or interpretation of data: Park, Chute, Rajpurkar, Lou, Ball, Shpanskaya, Jabarkheel, Kim, McKenna, Tseng, Ni, Wishah, Wittber, Hong, Wilson, Halabi, Patel, Lungren, Yeom.

Drafting of the manuscript: Park, Chute, Rajpurkar, Lou, Ball, Jabarkheel, Kim, McKenna, Hong, Halabi, Lungren, Yeom.

Critical revision of the manuscript for important intellectual content: Park, Chute, Rajpurkar, Ball, Shpanskaya, Jabarkheel, Kim, Tseng, Ni, Wishah, Wittber, Wilson, Basu, Patel, Lungren, Ng, Yeom.

Statistical analysis: Park, Chute, Rajpurkar, Lou, Ball, Lungren.

Administrative, technical, or material support: Park, Chute, Shpanskaya, Jabarkheel, Kim, McKenna, Tseng, Wittber, Hong, Wilson, Lungren, Ng, Yeom.

Supervision: Park, Ball, Tseng, Halabi, Basu, Lungren, Ng, Yeom.

Conflict of Interest Disclosures: Drs Wishah and Patel reported grants from GE and Siemens outside the submitted work. Dr Patel reported participation in the speakers bureau for GE. Dr Lungren reported personal fees from Nines Inc outside the submitted work. Dr Yeom reported grants from Philips outside the submitted work. No other disclosures were reported.

Funding/Support: This work was supported by National Institutes of Health National Center for Advancing Translational Science Clinical and Translational Science Award UL1TR001085.

Role of the Funder/Sponsor: The National Institutes of Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

References

1.Jaja  BN, Cusimano  MD, Etminan  N,  et al.  Clinical prediction models for aneurysmal subarachnoid hemorrhage: a systematic review.  Neurocrit Care. 2013;18(1):143-153. doi:10.1007/s12028-012-9792-zPubMedGoogle ScholarCrossref
2.Turan  N, Heider  RA, Roy  AK,  et al.  Current perspectives in imaging modalities for the assessment of unruptured intracranial aneurysms: a comparative analysis and review.  World Neurosurg. 2018;113:280-292. doi:10.1016/j.wneu.2018.01.054PubMedGoogle ScholarCrossref
3.Yoon  NK, McNally  S, Taussky  P, Park  MS.  Imaging of cerebral aneurysms: a clinical perspective.  Neurovasc Imaging. 2016;2(1):6. doi:10.1186/s40809-016-0016-3Google ScholarCrossref
4.Jayaraman  MV, Mayo-Smith  WW, Tung  GA,  et al.  Detection of intracranial aneurysms: multi-detector row CT angiography compared with DSA.  Radiology. 2004;230(2):510-518. doi:10.1148/radiol.2302021465PubMedGoogle ScholarCrossref
5.Bharatha  A, Yeung  R, Durant  D,  et al.  Comparison of computed tomography angiography with digital subtraction angiography in the assessment of clipped intracranial aneurysms.  J Comput Assist Tomogr. 2010;34(3):440-445. doi:10.1097/RCT.0b013e3181d27393PubMedGoogle ScholarCrossref
6.Lubicz  B, Levivier  M, François  O,  et al.  Sixty-four-row multisection CT angiography for detection and evaluation of ruptured intracranial aneurysms: interobserver and intertechnique reproducibility.  AJNR Am J Neuroradiol. 2007;28(10):1949-1955. doi:10.3174/ajnr.A0699PubMedGoogle ScholarCrossref
7.White  PM, Teasdale  EM, Wardlaw  JM, Easton  V.  Intracranial aneurysms: CT angiography and MR angiography for detection prospective blinded comparison in a large patient cohort.  Radiology. 2001;219(3):739-749. doi:10.1148/radiology.219.3.r01ma16739PubMedGoogle ScholarCrossref
8.Suzuki  K.  Overview of deep learning in medical imaging.  Radiol Phys Technol. 2017;10(3):257-273. doi:10.1007/s12194-017-0406-5PubMedGoogle ScholarCrossref
9.Rajpurkar  P, Irvin  J, Ball  RL,  et al.  Deep learning for chest radiograph diagnosis: a retrospective comparison of the CheXNeXt algorithm to practicing radiologists.  PLoS Med. 2018;15(11):e1002686. doi:10.1371/journal.pmed.1002686PubMedGoogle ScholarCrossref
10.Bien  N, Rajpurkar  P, Ball  RL,  et al.  Deep-learning-assisted diagnosis for knee magnetic resonance imaging: development and retrospective validation of MRNet.  PLoS Med. 2018;15(11):e1002699. doi:10.1371/journal.pmed.1002699PubMedGoogle ScholarCrossref
11.Morita  A, Kirino  T, Hashi  K,  et al; UCAS Japan Investigators.  The natural course of unruptured cerebral aneurysms in a Japanese cohort.  N Engl J Med. 2012;366(26):2474-2482. doi:10.1056/NEJMoa1113260PubMedGoogle ScholarCrossref
12.Yushkevich  PA, Piven  J, Hazlett  HC,  et al.  User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability.  Neuroimage. 2006;31(3):1116-1128. doi:10.1016/j.neuroimage.2006.01.015PubMedGoogle ScholarCrossref
13.He  K, Zhang  X, Ren  S, Sun  J. Deep residual learning for image recognition. Paper presented at: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition; June 27, 2016; Las Vegas, NV.
14.Xie  S, Girshick  R, Dollár  P, Tu  Z, He  K. Aggregated residual transformations for deep neural networks. Paper presented at: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); July 25, 2017; Honolulu, HI.
15.Hu  J, Shen  L, Sun  G. Squeeze-and-excitation networks. Paper presented at: 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); June 21, 2018; Salt Lake City, Utah.
16.Ronneberger  O, Fischer  P, Brox  T. U-net: Convolutional networks for biomedical image segmentation. International Conference on Medical Image Computing and Computer-Assisted Intervention. Basel, Switzerland: Springer International; 2015:234–241.
17.Carreira  J, Zisserman  A. Quo vadis, action recognition? a new model and the kinetics dataset. Paper presented at: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); July 25, 2017; Honolulu, HI.
18.Chen  L-C, Papandreou  G, Schroff  F, Adam  H. Rethinking atrous convolution for semantic image segmentation. https://arxiv.org/abs/1706.05587. Published June 17, 2017. Accessed May 7, 2019.
19.Milletari  F, Navab  N, Ahmadi  S-A. V-net: Fully convolutional neural networks for volumetric medical image segmentation. Paper presented at: 2016 IEEE Fourth International Conference on 3D Vision (3DV); October 26-28, 2016; Stanford, CA.
20.Loshchilov  I, Hutter  F. Sgdr: Stochastic gradient descent with warm restarts. Paper presented at: 2017 Fifth International Conference on Learning Representations; April 24-26, 2017; Toulon, France.
21.Huang  G, Sun  Y, Liu  Z, Sedra  D, Weinberger  KQ. Deep networks with stochastic depth. European Conference on Computer Vision. Basel, Switzerland: Springer International; 2016:646–661.
22.Horos. https://horosproject.org. Accessed May 1, 2019.
23.Wilson  EB.  Probable inference, the law of succession, and statistical inference.  J Am Stat Assoc. 1927;22(158):209-212. doi:10.1080/01621459.1927.10502953Google ScholarCrossref
24.Fleiss  JL, Cohen  J.  The equivalence of weighted kappa and the intraclass correlation coefficient as measures of reliability.  Educ Psychol Meas. 1973;33(3):613-619. doi:10.1177/001316447303300309Google ScholarCrossref
25.Benjamini  Y, Hochberg  Y.  Controlling the false discovery rate: a practical and powerful approach to multiple testing.  J R Stat Soc Series B Stat Methodol. 1995;57(1):289-300.Google Scholar
26.Maldaner  N, Stienen  MN, Bijlenga  P,  et al.  Interrater agreement in the radiologic characterization of ruptured intracranial aneurysms based on computed tomography angiography.  World Neurosurg. 2017;103:876-882.e1. doi:10.1016/j.wneu.2017.04.131PubMedGoogle ScholarCrossref
27.Wang  Y, Gao  X, Lu  A,  et al.  Residual aneurysm after metal coils treatment detected by spectral CT.  Quant Imaging Med Surg. 2012;2(2):137-138.PubMedGoogle Scholar
28.Yoon  YW, Park  S, Lee  SH,  et al.  Post-traumatic myocardial infarction complicated with left ventricular aneurysm and pericardial effusion.  J Trauma. 2007;63(3):E73-E75. doi:10.1097/01.ta.0000246896.89156.70PubMedGoogle ScholarCrossref
29.Tomandl  BF, Hammen  T, Klotz  E, Ditt  H, Stemper  B, Lell  M.  Bone-subtraction CT angiography for the evaluation of intracranial aneurysms.  AJNR Am J Neuroradiol. 2006;27(1):55-59.PubMedGoogle Scholar
30.Shi  W-Y, Li  Y-D, Li  M-H,  et al.  3D rotational angiography with volume rendering: the utility in the detection of intracranial aneurysms.  Neurol India. 2010;58(6):908-913. doi:10.4103/0028-3886.73743PubMedGoogle ScholarCrossref
31.Lin  N, Ho  A, Gross  BA,  et al.  Differences in simple morphological variables in ruptured and unruptured middle cerebral artery aneurysms.  J Neurosurg. 2012;117(5):913-919. doi:10.3171/2012.7.JNS111766PubMedGoogle ScholarCrossref
32.Villablanca  JP, Jahan  R, Hooshi  P,  et al.  Detection and characterization of very small cerebral aneurysms by using 2D and 3D helical CT angiography.  AJNR Am J Neuroradiol. 2002;23(7):1187-1198.PubMedGoogle Scholar
33.Chang  PD, Kuoy  E, Grinband  J,  et al.  Hybrid 3D/2D convolutional neural network for hemorrhage evaluation on head CT.  AJNR Am J Neuroradiol. 2018;39(9):1609-1616. doi:10.3174/ajnr.A5742PubMedGoogle ScholarCrossref
34.Chilamkurthy  S, Ghosh  R, Tanamala  S,  et al.  Deep learning algorithms for detection of critical findings in head CT scans: a retrospective study.  Lancet. 2018;392(10162):2388-2396. doi:10.1016/S0140-6736(18)31645-3PubMedGoogle ScholarCrossref
35.Jnawali  K, Arbabshirani  MR, Rao  N, Patel  AA. Deep 3D convolution neural network for CT brain hemorrhage classification. Paper presented at: Medical Imaging 2018: Computer-Aided Diagnosis. February 27, 2018; Houston, TX. doi:10.1117/12.2293725
36.Titano  JJ, Badgeley  M, Schefflein  J,  et al.  Automated deep-neural-network surveillance of cranial images for acute neurologic events.  Nat Med. 2018;24(9):1337-1341. doi:10.1038/s41591-018-0147-yPubMedGoogle ScholarCrossref
37.Ueda  D, Yamamoto  A, Nishimori  M,  et al.  Deep learning for MR angiography: automated detection of cerebral aneurysms.  Radiology. 2019;290(1):187-194.PubMedGoogle ScholarCrossref

Read Full Post »


Real Time Coverage @BIOConvention #BIO2019: Machine Learning and Artificial Intelligence: Realizing Precision Medicine One Patient at a Time

Reporter: Stephen J Williams, PhD @StephenJWillia2

The impact of Machine Learning (ML) and Artificial Intelligence (AI) during the last decade has been tremendous. With the rise of infobesity, ML/AI is evolving to an essential capability to help mine the sheer volume of patient genomics, omics, sensor/wearables and real-world data, and unravel the knot of healthcare’s most complex questions.

Despite the advancements in technology, organizations struggle to prioritize and implement ML/AI to achieve the anticipated value, whilst managing the disruption that comes with it. In this session, panelists will discuss ML/AI implementation and adoption strategies that work. Panelists will draw upon their experiences as they share their success stories, discuss how to implement digital diagnostics, track disease progression and treatment, and increase commercial value and ROI compared against traditional approaches.

  • most of trials which are done are still in training AI/ML algorithms with training data sets.  The best results however have been about 80% accuracy in training sets.  Needs to improve
  • All data sets can be biased.  For example a professor was looking at heartrate using a IR detector on a wearable but it wound up that different types of skin would generate a different signal to the detector so training sets maybe population biases (you are getting data from one group)
  • clinical grade equipment actually haven’t been trained on a large set like commercial versions of wearables, Commercial grade is tested on a larger study population.  This can affect the AI/ML algorithms.
  • Regulations:  The regulatory bodies responsible is up to debate.  Whether FDA or FTC is responsible for AI/ML in healtcare and healthcare tech and IT is not fully decided yet.  We don’t have the guidances for these new technologies
  • some rules: never use your own encryption always use industry standards especially when getting personal data from wearables.  One hospital corrupted their system because their computer system was not up to date and could not protect against a virus transmitted by a wearable.
  • pharma companies understand they need to increase value of their products so very interested in how AI/ML can be used.

Please follow LIVE on TWITTER using the following @ handles and # hashtags:

@Handles

@pharma_BI

@AVIVA1950

@BIOConvention

# Hashtags

#BIO2019 (official meeting hashtag)

Read Full Post »


AI in Psychiatric Treatment – Using Machine Learning to Increase Treatment Efficacy in Mental Health

Reporter: Aviva Lev- Ari, PhD, RN

Featuring Start Up: aifred

www.aifredhealth.com

About Us

The inability to predict any given individual’s unique response to psychiatric treatment is a huge bottleneck to recovery from mental health conditions.
To address this challenge, we are creating a deep-learning based clinical decision tool for physicians to bring personalized medicine to psychiatry.
Initially, we will be focusing on treatments for depression, but we plan to scale Aifred to encompass all mental health conditions in order to amplify clinical utility. At its core, aifred is leveraging the collective intelligence of the scientific and medical community to bring better healthcare to all.
We are a proud official IBM Watson AI XPrize team, headquartered in Montreal, Canada.

Read more about us:

Deep Learning


Something unique to every machine learning company is the precise nature of their hyperparameter optimization and goals of their model. We will optimize aifred with the help of a distributed network of domain experts in psychiatry — a collaboration unique to aifred health. We are implementing attention networks responsible for removing the “black-box” nature of neural networks. As well, we are analyzing the quality of model predictions, allowing both for greater interpretability of model decisions and the generation of new basic research questions, which are going to be unique to the data-set and optimization techniques we develop in-house. By training aifred on reliable datasets, we are able to ensure quality input to our model. De-identified patient outcomes will feed back into our neural networks to continuously improve aifred’s predictive power. Feature engineering is an important part of determining which inputs go into a network and varies how it’s done for every team- once again, this will be undertaken with the support of diverse group of experts we are recruiting.

Our Product


Treatment Prediction

The aifred solution makes use of innovative and powerful machine learning techniques predict treatment efficacy based on an array of patient characteristics.

Interpretability

Forget the blackbox! Our system will provide a report highlighting the most significant features that led to a treatment prediction.

Patient Data Tracking

Track patient symptoms and test results to monitor outcomes or make new predictions. Banks of standardized questionnaires, data visualization, scheduling software — all of it modular and capable of being tailored to clinicians’ needs.

Electronic Patient Record

Keep all important patient information in one place, and get insights using our analytics.

 

In the News:

Montreal Gazette article written about our startup:

https://montrealgazette.com/news/local-news/a-software-tool-to-improve-treatment-of-depression-was-developed-in-montreal

Press about us winning first place globally in the IBM Watson AI XPrize milestone competition

http://www.concordia.ca/news/stories/2018/12/07/Aifred-Health-and-Nectar-take-home-top-honours-at-the-ibm-Watson-ai-xprize-milestone-competition.html

Forbes article that features our CTO, Robert Fratila:

https://www.forbes.com/sites/insights-intelai/2018/11/29/5-entrepreneurs-on-the-rise-in-ai/

Post about our graduation from the prestigious creative destruction lab program:

https://medium.com/@aifred/aifred-health-graduates-from-the-creative-destruction-lab-500e4b2a83c?fbclid=IwAR2qz9iQf8-4B07ljB1ZP3GAUCGZcK-CyxgG9cu1jtN8moEDSexvZeEcN7c

McGill University article featuring us:

https://www.mcgill.ca/giving/why-giving-matters/2019/02/05/taking-depression-using-ai?fbclid=IwAR1NN_ID04IJMto97cT-28fDfVxg1rbp7c7arbGf48MrzL4_Q4EaEzmegj8

 

REFERENCE

The Incredible Ways Artificial Intelligence Is Now Used In Mental Health

Bernard Marr 12:23 am

https://www.forbes.com/sites/bernardmarr/2019/05/03/the-incredible-ways-artificial-intelligence-is-now-used-in-mental-health/amp/?__twitter_impression=true

4 Benefits of using AI to help solve the mental health crisis

There are several reasons why AI could be a powerful tool to help us solve the mental health crisis. Here are four benefits:

  1.      Support mental health professionals

As it does for many industries, AI can help support mental health professionals in doing their jobs. Algorithms can analyze data much faster than humans, can suggest possible treatments, monitor a patient’s progress and alert the human professional to any concerns. In many cases, AI and a human clinician would work together.

  1.      24/7 access

Due to the lack of human mental health professionals, it can take months to get an appointment. If patients live in an area without enough mental health professionals, their wait will be even longer. AI provides a tool that an individual can access all the time, 24/7 without waiting for an appointment.

  1.      Not expensive

The cost of care prohibits some individuals from seeking help. Artificial intelligent tools could offer a more accessible solution.

  1.      Comfort talking to a bot

While it might take some people time to feel comfortable talking to a bot, the anonymity of an AI algorithm can be positive. What might be difficult to share with a therapist in person is easier for some to disclose to a bot.

Other related articles published in this Open Access Online Scientific Journal include the following:

Resources on Artificial Intelligence in Health Care and in Medicine:

Articles of Note at PharmaceuticalIntelligence.com @AVIVA1950 @pharma_BI

Curator: Aviva Lev-Ari, PhD, RN

https://www.linkedin.com/pulse/resources-artificial-intelligence-health-care-note-lev-ari-phd-rn/

R&D for Artificial Intelligence Tools & Applications: Google’s Research Efforts in 2018

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2019/01/16/rd-for-artificial-intelligence-tools-applications-googles-research-efforts-in-2018/

 

McKinsey Top Ten Articles on Artificial Intelligence: 2018’s most popular articles – An executive’s guide to AI

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2019/01/21/mckinsey-top-ten-articles-on-artificial-intelligence-2018s-most-popular-articles-an-executives-guide-to-ai/

 

LIVE Day Three – World Medical Innovation Forum ARTIFICIAL INTELLIGENCE, Boston, MA USA, Monday, April 10, 2019

https://pharmaceuticalintelligence.com/2019/04/10/live-day-three-world-medical-innovation-forum-artificial-intelligence-boston-ma-usa-monday-april-10-2019/

 

LIVE Day Two – World Medical Innovation Forum ARTIFICIAL INTELLIGENCE, Boston, MA USA, Monday, April 9, 2019

https://pharmaceuticalintelligence.com/2019/04/09/live-day-two-world-medical-innovation-forum-artificial-intelligence-boston-ma-usa-monday-april-9-2019/

 

LIVE Day One – World Medical Innovation Forum ARTIFICIAL INTELLIGENCE, Boston, MA USA, Monday, April 8, 2019

https://pharmaceuticalintelligence.com/2019/04/08/live-day-one-world-medical-innovation-forum-artificial-intelligence-westin-copley-place-boston-ma-usa-monday-april-8-2019/

The Regulatory challenge in adopting AI

Author and Curator: Dror Nir, PhD

https://pharmaceuticalintelligence.com/2019/04/07/the-regulatory-challenge-in-adopting-ai/

 

VIDEOS: Artificial Intelligence Applications for Cardiology

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2019/03/11/videos-artificial-intelligence-applications-for-cardiology/

 

Artificial Intelligence in Health Care and in Medicine: Diagnosis & Therapeutics

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2019/01/21/artificial-intelligence-in-health-care-and-in-medicine-diagnosis-therapeutics/

 

World Medical Innovation Forum, Partners Innovations, ARTIFICIAL INTELLIGENCE | APRIL 8–10, 2019 | Westin, BOSTON

https://worldmedicalinnovation.org/agenda/

https://pharmaceuticalintelligence.com/2019/02/14/world-medical-innovation-forum-partners-innovations-artificial-intelligence-april-8-10-2019-westin-boston/

 

Digital Therapeutics: A Threat or Opportunity to Pharmaceuticals

Reporter and Curator: Dr. Sudipta Saha, Ph.D.

https://pharmaceuticalintelligence.com/2019/03/18/digital-therapeutics-a-threat-or-opportunity-to-pharmaceuticals/

 

The 3rd STATONC Annual Symposium, April 25-27, 2019, Hilton Hartford, CT, 315 Trumbull St., Hartford, CT 06103

Reporter: Stephen J. Williams, Ph.D.

https://pharmaceuticalintelligence.com/2019/02/26/the-3rd-stat4onc-annual-symposium-april-25-27-2019-hilton-hartford-connecticut/

 

2019 Biotechnology Sector and Artificial Intelligence in Healthcare

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2019/05/10/2019-biotechnology-sector-and-artificial-intelligence-in-healthcare/

 

The Journey of Antibiotic Discovery

Reporter and Curator: Dr. Sudipta Saha, Ph.D.

https://pharmaceuticalintelligence.com/2019/05/19/the-journey-of-antibiotic-discovery/

 

Artificial intelligence can be a useful tool to predict Alzheimer

Reporter: Irina Robu, PhD

https://pharmaceuticalintelligence.com/2019/01/26/artificial-intelligence-can-be-a-useful-tool-to-predict-alzheimer/

 

HealthCare focused AI Startups from the 100 Companies Leading the Way in A.I. Globally

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2018/01/18/healthcare-focused-ai-startups-from-the-100-companies-leading-the-way-in-a-i-globally/

 

2018 Annual World Medical Innovation Forum Artificial Intelligence April 23–25, 2018 Boston, Massachusetts | Westin Copley Place

https://worldmedicalinnovation.org/

https://pharmaceuticalintelligence.com/2018/01/18/2018-annual-world-medical-innovation-forum-artificial-intelligence-april-23-25-2018-boston-massachusetts-westin-copley-place/

 

MedCity Converge 2018 Philadelphia: Live Coverage @pharma_BI

Reporter: Stephen J. Williams

https://pharmaceuticalintelligence.com/2018/07/11/medcity-converge-2018-philadelphia-live-coverage-pharma_bi/

 

IBM’s Watson Health division – How will the Future look like?

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2019/04/24/ibms-watson-health-division-how-will-the-future-look-like/

 

Live Coverage: MedCity Converge 2018 Philadelphia: AI in Cancer and Keynote Address

Reporter: Stephen J. Williams, PhD

https://pharmaceuticalintelligence.com/2018/07/11/live-coverage-medcity-converge-2018-philadelphia-ai-in-cancer-and-keynote-address/

 

HUBweek 2018, October 8-14, 2018, Greater Boston – “We The Future” – coming together, of breaking down barriers, of convening across disciplinary lines to shape our future

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2018/10/08/hubweek-2018-october-8-14-2018-greater-boston-we-the-future-coming-together-of-breaking-down-barriers-of-convening-across-disciplinary-lines-to-shape-our-future/

 

Role of Informatics in Precision Medicine: Notes from Boston Healthcare Webinar: Can It Drive the Next Cost Efficiencies in Oncology Care?

Reporter: Stephen J. Williams, Ph.D.

https://pharmaceuticalintelligence.com/2019/01/03/role-of-informatics-in-precision-medicine-can-it-drive-the-next-cost-efficiencies-in-oncology-care/

 

Gene Editing with CRISPR gets Crisper

Curators: Larry H. Bernstein, MD, FCAP and Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2016/05/03/gene-editing-with-crispr-gets-crisper/

 

Disease related changes in proteomics, protein folding, protein-protein interaction

Curator: Larry H. Bernstein, MD, FCAP

https://pharmaceuticalintelligence.com/2016/05/13/disease-related-changes-in-proteomics-protein-folding-protein-protein-interaction/

 

Can Blockchain Technology and Artificial Intelligence Cure What Ails Biomedical Research and Healthcare

Curator: Stephen J. Williams, Ph.D.

https://pharmaceuticalintelligence.com/2018/12/10/can-blockchain-technology-and-artificial-intelligence-cure-what-ails-biomedical-research-and-healthcare/

 

N3xt generation carbon nanotubes

Curator: Larry H. Bernstein, MD, FCAP

https://pharmaceuticalintelligence.com/2015/12/14/n3xt-generation-carbon-nanotubes/

 

Healthcare conglomeration to access Big Data and lower costs

Curator: Larry H. Bernstein, MD, FCAP

https://pharmaceuticalintelligence.com/2016/01/13/healthcare-conglomeration-to-access-big-data-and-lower-costs/

 

Mindful Discoveries

Curator: Larry H. Bernstein, MD, FCAP

https://pharmaceuticalintelligence.com/2016/01/28/mindful-discoveries/

 

Synopsis Days 1,2,3: 2018 Annual World Medical Innovation Forum Artificial Intelligence April 23–25, 2018 Boston, Massachusetts | Westin Copley Place

Curator: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2018/04/26/synopsis-days-123-2018-annual-world-medical-innovation-forum-artificial-intelligence-april-23-25-2018-boston-massachusetts-westin-copley-place/

 

Unlocking the Microbiome

Curator: Larry H. Bernstein, MD, FCAP

https://pharmaceuticalintelligence.com/2016/02/07/unlocking-the-microbiome/

 

Linguamatics announces the official launch of its AI self-service text-mining solution for researchers.

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2018/05/10/linguamatics-announces-the-official-launch-of-its-ai-self-service-text-mining-solution-for-researchers/

 

Novel Discoveries in Molecular Biology and Biomedical Science

Curator: Larry H. Bernstein, MD, FCAP

https://pharmaceuticalintelligence.com/2016/05/30/novel-discoveries-in-molecular-biology-and-biomedical-science/

 

Biomarker Development

Curator: Larry H. Bernstein, MD, FCAP

https://pharmaceuticalintelligence.com/2015/11/16/biomarker-development/

 

Imaging of Cancer Cells

Curator: Larry H. Bernstein, MD, FCAP

https://pharmaceuticalintelligence.com/2016/04/20/imaging-of-cancer-cells/

 

Future of Big Data for Societal Transformation

Curator: Larry H. Bernstein, MD, FCAP

https://pharmaceuticalintelligence.com/2015/12/14/future-of-big-data-for-societal-transformation/

 

mRNA Data Survival Analysis

Curators: Larry H. Bernstein, MD, FCAP and Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2016/06/18/mrna-data-survival-analysis/

 

Applying AI to Improve Interpretation of Medical Imaging

Author and Curator: Dror Nir, PhD

https://pharmaceuticalintelligence.com/2019/05/28/applying-ai-to-improve-interpretation-of-medical-imaging/

Read Full Post »


LIVE Day Three – World Medical Innovation Forum ARTIFICIAL INTELLIGENCE, Boston, MA USA, Monday, April 10, 2019

 

www.worldmedicalinnovation.org

 

The Forum will focus on patient interactions across care settings, and the role technology and data can play in advancing knowledge discovery and care delivery. The agenda can be found here.

https://worldmedicalinnovation.org/agenda/

Leaders in Pharmaceutical Business Intelligence (LPBI) Group

represented by Founder & Director, Aviva Lev-Ari, PhD, RN will cover this event in REAL TIME using Social Media

@pharma_BI

@AVIVA1950

@PHSInnovation

#WMIF19 

Wednesday, April 10, 2019

7:00 am – 12:00 pm
7:30 am – 9:30 am
Bayer Ballroom

Innovation Discovery Grant Awardee Presentations

Eleven clinical teams selected to receive highly competitive Innovation Discovery Grants present their work illustrating how AI can be used to improve patient health and health care delivery. This session is designed for investors, entrepreneurs, investigators, and others who are interested in commercializing AI opportunities that are currently in development with support from the Innovation Office.

To view speakers and topics, click here.

Where AI Meets Clinical Care

Twelve clinical AI teams culled through the Innovation Discovery Grant program present their work illustrating how AI can be used to improve patient health and healthcare delivery. This session is designed for investors, entrepreneurs, investigators, and others who are interested in commercializing AI opportunities that are currently in development with support from the Innovation Office.

IDG logo

Peter Dunn, MD

Vice President, Perioperative Services and Healthcare System Engineering, MGH; Assistant Professor, Anesthesia, HMS

Using Deep Learning to Optimize Hospital Capacity Management

  • collaboration with @MIT @MGH
  • deploy mobile app across all Partners institutions

 

Kevin Elias, MD

Director, Gynecologic Oncology Research Laboratory, BH; Assistant Professor, HMS

Screening for Cancer Using Serum miRNA Neural Networks

  • cancer screening fragmented process – tests not efficient No screening for many common cancer type
  • Cervical, Breast, Colon, Ovarian Uterus Cancer
  • Serum miRNA multiple cancer types

 

Alexandra Golby, MD

Director, Image-Guided Neurosurgery, BH; Professor, Neurosurgery and Radiology, HMS

Using Machine Learning to Optimize Optical Image Guidance for Brain Tumor Surgery

  • optical visualization in Neurosurgery – to improve Brain Cancer surgery Tumor removal complete resection could cause neurological deficits
  • BWH original research on Neuronavigations, intraops MRI
  • New Tool Real Time: Color code tumors using light diagnostics with machine learning
  • GUIDING Brain surgery, applicable for Breast Cancer
  • iP filling prototype creation, testing, pre-clinical testing, clinical protocol established academic-industrial partnerships
  • AI based – World 1st guided neurosurgery

 

Jayashree Kalpathy-Cramer, PhD

Director, QTIM Lab, MGH; Associate Professor, Radiology, HMS

DeepROP: Point-of-Care System for Diagnosis of Plus Disease in Retinopathy of Prematurity

  • Prematurity 1250 gr <31 weeks f gestation
  • ROP – Retinopathy of prematurity (ROP)
  • Images annotated Plus/not plus – algorithm for rating images “normal” or “plus”
  • DeepROP Applicationsinto Camera for data acquisition, iPhone

 

Jochen Lennerz, MD, PhD

Associate Director, Center for Integrated Diagnostics, MGH; Assistant Professor, HMS

Predicting Unnecessary Surgeries in High-Risk Breast Lesions

  • 10% reduction of high risk lesion equivalent to $1.4Billion in cost savings
  • Funding for Production line

Bruno Madore, PhD

Associate Professor, Radiology, BH, HMS

Sensor Technology for Enhanced Medical Imaging

  • ML Ultrasound – Organ configuration Motion (OCM) sensor
  • Hybrid MRI-ultrasound acquisitions
  • Long term vision – collaboration with Duke for a wireless device

 

Jinsong Ouyang, PhD

Physicist, MGH; Associate Professor, HMS

Training a Neural Network to Detect Lesions

  • Approach – train a NN using artificially inserted lesions

APPLICATIONS:

  • Build unlimitted number of training sets using small 15-50 human data sets generated
  • bone lession detection using SPECT
  • cardiac detect myocardial perfusion SPECT
  • Tumor detection PET
  • Volume detection/locatization of artificial Spinal Lesions (L1-L5)

 

David Papke, MD, PhD

Resident, Surgical Pathology, BH; Clinical Fellow, HMS

Augmented Digital Microscopy for Diagnosis of Endometrial Neoplasia

See tweet

 

Martin Teicher, MD, PhD

Director, Developmental Biopsychiatry Research Program, McLean; Associate Professor, Psychiatry, HMS

Poly-Exposure Risk Scores for Psychiatric Disorders

  • MACE Scale – psychopathology development – collinearity
  • Identifying sensitivity period predictors of major depression
  • predicting risk in adolescence – dataset with high collinearity
  • Onset of depression age 10-15
  • 50% assessment exposure to adversity – based on neuroimaging
  • Analytics and AI longitudinal studies

 

 

Christian Webb, PhD

Director, Treatment and Etiology of Depression, Youth Lab, McLean; Assistant Professor, Psychiatry, HMS

Leveraging Machine Learning to Match Depressed Patients to the Optimal Treatment

  • 4-8 wks of treatment till psychotropic drugs work
  • Data driven approaches: ML can match better patients to antidepressant treatments (Zoloft vs Placebo responder /non responder)?
  • Large number of variables prediction, prognosis calculator, good vs poor outcome
  • Better on Zoloft vs Placebo

 

Brandon Westover, MD, PhD

Executive Director, Clinical Data Animation Center, MGH; Associate Professor, Neurology, HMS
  • seizure, prediction of next attack
  • EEG readings – accurate diagnosis on epilepsy
  • 50 million World wide
  • automated epilepsy detection
  • @MGH – 1,063 EEGs 88,000 spikes 7 experts scored – not all agreed
  • How well can experts identify spikes?
  • Super spike detector is better than Experts – False positive 60% 87% Sensitivity vs 10% and 87% by AI
Moderator: David Louis, MD
  • Pathologist-in-Chief, MGH; Benjamin Castleman Professor of Pathology, HMS
Moderator: Clare Tempany, MD
  • Vice-Chair, Radiology Research, BH; Ferenc Jolesz MD Professor of Radiology, HMS
9:30 am – 10:00 am
10:00 am – 10:30 am
Bayer Ballroom

1:1 Fireside Chat: Stefan Oelrich, Member of the Board of Management; President, Pharmaceutical, Bayer AG

Introduction by: John Fish
  • CEO, Suffolk; Chairman of Board Trustees, Brigham Health
Moderator: Betsy Nabel, MD
  • President, Brigham Health; Professor of Medicine, HMS
  • Member of the Board of Management, Bayer AG; President, Pharmaceutical, Bayer AG

Chief Digital Officers

  • Leaders at the top needs to understand AI
  • Millennials needs to fill Baby boomer retiring
  • Boston – funding Research by NIH by private investment technology transfer to commercialization
  • Career advice: Academia is the first step for credibility move to Big Pharma, create own company
  • America economic strength built on innovation in Healthcare to invest
  • Leadership at Bayer: “Culture eat strategy for Breakfast”
  • AI overcoming barriers – AI improving what we know Medical imaging human vs machine – AI is the new norm – platforms Imaging AI device to detect Hypertension more accurately development of Bayer and Merck – Bayer leader in Radiology
  • Clinical research End point to reach compare
  • Future billion end point which therapeutic pathway is best for which patient
  • Incentives for risky strategy
  • Motivation to collaborate in Boston: Cardiology with broad Institute
  • BWH data and algorithms to increase knowledge
  • Pricing medicine around the World
  • US system in-transparent – patients do not understand Price of meds Rebates to Payers
  • Medical Part B – no pass to Rebates price tied to value
  • As industry – innovations in Pharma reduce healthcare costs Germany 15% of HealthCare on Drugs, generics, “Patented medicine 4%” of all Best in Europe
  • beak silos
  • In US training physicians to lead innovations
10:30 am – 11:00 am
Bayer Ballroom

1:1 Fireside Chat: Deepak Chopra, MD, Founder, The Chopra Foundation

Moderator: Rudolph Tanzi, PhD
  • Vice-Chair, Neurology, Director, Genetics and Aging Research Unit, MGH; Joseph P. and Rose F. Kennedy Professor of Neurology, HMS
  • IMAGING of Brains of Women in Meditation – enlongate telemeres
  • inflammation decrease – Sleep health interactions exsercise learning new things diet
  • flashing from brain wastes – amaloydosis AD – 35 genes variance leading to disease
  • Founder, The Chopra Foundation – Body-Mind Connection
  • AI – re-invest our bodies Telemeres, transferdomics,
  • Nutrition, sleep, excercise, BP, HR, sympathetic vs non sympatheric nervous system breathing pattern, – microbiome subjective experience with Vitals emotional well being
  • emersive augmented
  • longer Telemerese – anti aging correlation
  • biomarkers vs states of energy
  • wisdom best knowledge for self awareness – highest intelligence – NOT artificial
  • Thoughts on being aware
11:00 am – 11:50 am
Bayer Ballroom

Using AI to Predict and Monitor Human Performance and Neurological Disease

In the quest for effective treatments aimed at devastating neurological diseases like Alzheimer’s and ALS, there is a critical need for robust methods to predict and monitor disease progression. AI-based approaches offer promise in this important area. Panelists will discuss efforts to map movement-related disorders and use machine learning to predict the path of disease with imaging and biomarkers.

  • Chief of Neurology, Co-Director, Neurological Clinical Research Institute, MGH; Julieanne Dorn Professor of Neurology, HMS
  • Chief Scientist, Dolby Laboratories Stanford & Adobe – measuring experience
  • convergence of skills
  • internal wellness measured in the ear, motions
  • Stimulate Vagal nerve through the ear for depression treatment
  • Legislation in CA contribution to spaces
  • Global Therapeutic Head, Neuroscience Janssen Research & Development
  • Disease starts earlier Biogen contributions in the field
  • measurement surrogate indicators for outcome given interventions
  • Autism-spectrum not one disease
  • AI will enhance the human competence for measurement
  • UK based efforts to share dat and launch programs for Dementia
  • Conditions of Brain & Mind – declining cognitive
  • Democratization of discovery
  • AI benefit iterative process in changing and improving Algorithms — FDA approved algorithm needs several versions in the future
  • Complexity of CNS Polygenic gene scores
  • Dynamics of AI
  • EVP and CMO, Biogen
  • MS – follow patients, patient reporting in 10 centers , vision cognitions –
  • Obtain measurement even on normal people for early detection – FDA introduced Stage 1,2,3 Biomarker based
  • Newborn Kit of screening teat early helps
  • Home monitoring at Home for onset of AD

Dr. Isaac Galatzer-Levy – NYU & AiCure

  • All CNS diseases are heterogeneous
  • ML requires collaboration
  • AiCure – Medication adherence monitoring from Voice of patients
  • Sampling populations – cell phone
  • Re-investigate studies that have failed with new AI tools
11:50 am – 12:50 pm
Bayer Ballroom

Disruptive Dozen: 12 Technologies that will reinvent AI in the Next 12 Months

The Disruptive Dozen identifies and ranks the AI technologies that Partners faculty feel will break through over the next year to significantly improve health care.

  • innovations, technologies close to make to market

#12 David Ahern – Mental Health in US closing the Gap

#11 David Ting – Voice first

#10 Bharti Khurana – Partners Violence

#9 Gilberto Gonzales – Acute Stroke care

#8 James Hefferman – Burden og Health care ADM

#7 Samuel Aronson – FHIR Health information exchange

#6 Joan Miller – AI for eye health

#5 Brsndon Westover – A window to the Brain

#4 Rochelle Walensky – Automated detection of Malaria

#3 Annette Kim – Streamlining Diagnosis 

  #2 Thomas McCoy – Better Prediction of Suicide risk

  #1 Alexandra Golby – Reimagining Medical Imaging 

 

Moderator: Jeffrey Golden, MD
  • Chair, Department of Pathology, BH; Ramzi S. Cotran Professor of Pathology, HMS
  • Associate Chief, Infection Control Unit, MGH; Assistant Professor, Medicine, HMS
1:00 pm – 1:10 pm
Bayer Ballroom

Read Full Post »


LIVE Day Two – World Medical Innovation Forum ARTIFICIAL INTELLIGENCE, Boston, MA USA, Monday, April 9, 2019

 

www.worldmedicalinnovation.org

 

The Forum will focus on patient interactions across care settings, and the role technology and data can play in advancing knowledge discovery and care delivery. The agenda can be found here.

https://worldmedicalinnovation.org/agenda/

Leaders in Pharmaceutical Business Intelligence (LPBI) Group

represented by Founder & Director, Aviva Lev-Ari, PhD, RN will cover this event in REAL TIME using Social Media

@pharma_BI

@AVIVA1950

@PHSInnovation

#WMIF19 

Tuesday, April 9, 2019

7:00 am – 8:00 am
7:00 am – 5:00 pm
7:40 am – 7:50 am
Bayer Ballroom

Opening Remarks

  • Chief Innovation Officer, PHS; President, Partners HealthCare International
7:50 am – 8:40 am
Bayer Ballroom

Implementing AI in Cancer Care

With AI-enabled care strategies and digital technologies, clinicians and patients are embracing new approaches to improve the lives of cancer patients through enhanced diagnosis and treatment. These include AI-guided tools for more precise methods of predicting risk, more effective screening strategies, patient data driven insights  and more personalized treatments. Panelists will engage on how these and other innovations are enabling a new era of cancer care.

  • Chief, Breast Imaging Division, MGH; Professor of Radiology, HMS
  • FDA
  • President and Co-Founder, LunaDNA
  • Patients contribute personal data get share in the company
  • democratization by AI use
  • unrepresented population in research
  • education on technology
  • Retrospective and longitudinal studies
  • Bid Trust engaging responsively
  • Delta Electronics Professor, Electrical Engineering and Computer Science Department, MIT
  • developper of AI based applications @MGH Cancer Center
  • Training AI on 3% of population vs randomized that has its bias of patient selection
  • no standards of publishing AI in medicine
  • AI to help women
  • Integration of systems to help patients
  • Director, Cancer Genome Analysis, Broad Institute; Professor, Pathology, HMS
  • AI for early detection
  • big data analysis – noise vs point of signals
  • drug resistance using genomics
  • AI – regulate the type information reviewed by doctors
  • data acquisition and monitoring along the life of the product not only till FDA approve it
  • Reporting adverse events
  • Data cost of sequencing is dropping, biomarkers,
  • regulatory needed to adopt AI and reimbursement starts at academic center followed by the entire country
  • CEO, insitro
  • AI for drug discovery
  • epigenetic effect on lesions
  • Physician are over promised on Genomics, asking them to use complex data from multiple source need be curated before it gets to Physicians
  • Reversed clinical trial vs randomized 30 years follow up
  • Data is anonymized used in research contributors get back own diagnosis genomics understanding

 

8:40 am – 9:30 am
Bayer Ballroom

Imagining Medicine in the Year 2054

In 1984 Isaac Asimov was asked to predict what life in 2019 would be like. Using the same aperture, we as what will constitute health care 35 years from now? Current trends suggest that there will be significant gains in immunotherapy, gene therapy, and breakthrough treatments for neurologic, cardiovascular and oncologic diseases. Panelists will draw on their visionary perspective and will reflect on what to expect and why.

Moderator: Keith Flaherty, MD
  • Director, Clinical Research, Cancer Center, MGH; Professor of Medicine, HMS
  • CEO, Flagship Pioneering
  • Vice Chair for Scientific Innovation, Department of Medicine, BH; Associate Professor of Medicine, HMS
  • Director, Cellular Immunotherapy Program, Cancer Center, MGH; Assistant Professor, Medicine, HMS
  • Vice-Chair, Neurology, Director, Genetics and Aging Research Unit, MGH; Joseph P. and Rose F. Kennedy Professor of Neurology, HMS
9:30 am – 9:50 am
9:50 am – 10:15 am
Bayer Ballroom

1:1 Fireside Chat: Ash Carter, U.S. Secretary of Defense (2015 – 2017)

Moderator: Gregg Meyer, MD
  • Chief Clinical Officer, PHS; Professor of Medicine, HMS; 2019 Forum Co-Chair
  • U.S. Secretary of Defense (2015–2017)
10:15 am – 10:40 am
Bayer Ballroom

1:1 Fireside Chat: Honorable Alex Azar II, Secretary of Health and Human Services

Moderator: Gregg Meyer, MD
  • Chief Clinical Officer, PHS; Professor of Medicine, HMS; 2019 Forum Co-Chair
  • 24th Secretary of Health and Human Services
  • quality cate means outcomes
  • Pricing Transparency by HMOs and Hospitals
  • Plan D – instant electronic to Drug Pricing information
  • Medicare moves away from Procedure based payment
  • Data on services, drugs and procedures in a Patient-centered system
  • Big data, pricing information, CMS
  • AI inspector General – Claims – AI – do get yield
  • AI in procurement
  • AI for services to Medicare – prescription Tools for advising Patients on best drug to use based on medcial information
  • Patient HC information is owned by Pations and is portable
  • Blue Data 2.0 – access record by patients @CMS
10:40 am – 11:30 am
Bayer Ballroom

CEO Roundtable

Chief executives share perspectives on the impact of AI on their respective companies and industry segments. Panelists will discuss their views of AI, how AI figures into their organizations’ current product and investment strategies, and how they are measuring return on existing AI investments. The panel will also address opportunities and challenges surrounding AI, ranging from workforce needs to managing bias in AI development.

Moderator: Anne Klibanski, MD
  • Interim President and CEO, Chief Academic Officer, PHS; Laurie Carrol Guthart Professor of Medicine, HMS; 2019 Forum Co-Chair
  • Partnerships between companies like : GE, Phillips, Siemens
  • CEO, Philips
  • efficiencies and outcomes
  • adaptive intelligence to be integrated AI 1.8Billion Euro invested 600 scientists
  • collaboration with Dana Farber
  • Design thinking – work with clinicians to get insights on experience with technologies
  • system change for delivery of care
  • Open API – federated data architecture EMR companies will also need to adapt
  • Phillips builds centers in Pittsburgh, Cambridge, Amsterdam, Paris
  • EVP, Head, Pharmaceuticals Research and Development, Bayer AG
  • AI – R&D efficiency
  • Disruptive approaches optimization of synthesis of chemical reactions productivity and selection of molecules
  • In house data science expertise vs image pattern recognition of HTN collaboration with Merck
  • Collaboration with MIT on clinical Trials
  • changing provides vs longitudinal care
  • Access to talent – Data scientists Amazon is a competitor on talent for AI SKILLS DOMAIN EXPRET TOPIC
  • R&D AT BAYER – DATA SCIENCE IN each division
  • CEO, Siemens Healthineers
  • 400 research collaborations
  • “analog” way innovations generations
  • CEO, GE Healthcare
  • HC – Clinical command center in Hospitals collaboration with Partners
  • Investment is in platforms vs applications – Edison platform tool kits – Radiologist will develop their own on top of PLATFORMS from GE
  • Clinicians productivity will change with AI
  • Data scientist new identity – bigger developers of systems
11:30 am – 11:35 am
Bayer Ballroom
11:35 am – 11:45 am
11:45 am – 1:00 pm

Discovery Cafe Sessions

Lunch with Experts: Intensive sessions addressing cutting-edge artificial intelligence topics.

Provider Back Office of the Future

The application of AI-based technologies to the business side of health care — including functions such as billing, payment, and insurance claims management — could lead to significant improvements in health care operations and efficiency, with billions of dollars in savings each year. Panelists will discuss emerging tools and technologies as well as the opportunities and pitfalls of using AI to innovate and automate back office functions.

Moderator: Peter Markell, EVP, Administration and Finance, CFO and Treasurer, PHS

Inge Harrison, CNO/VP of Strategic Advisory Services, Verge Health

Kent Ivanoff, CEO, VisitPay

Mary Beth Remorenko, VP, Revenue Cycle Operations, PHS

Brian Robertson, CEO, VisiQuate

 

Chief Digital Strategy Officer Roundtable

With the advent of AI-enabled technologies, this session brings together leading chief digital health officers. The discussion will address tradeoffs in sequencing technology across academic medical centers; what technologies are being prioritized; and consumer expectations.

Moderator: Alistair Erskine, MD, Chief Digital Health Officer, PHS

Michael Anderes, Chief Innovation and Digital Health Officer, Froedtert Health; President, Inception Health

Adam Landman, MD, VP and CIO, BH; Associate Professor of Emergency Medicine, HMS

Aimee Quirk, CEO, innovationOchsner

Richard Zane, MD, Chief Innovation Officer, UCHealth; Professor and Chair,Department of Emergency Medicine, University of Colorado School of Medicine

 

Innovation Fellows: A New Model of Collaboration

The Innovation Fellows Program provides experiential career development opportunities for future leaders in health care. It facilitates personnel exchanges between Harvard Medical School staff from Partners’ hospitals and participating biopharmaceutical, device, venture capital, digital health, payor and consulting firms. Fellows and Hosts learn from each other as they collaborate on projects ranging from clinical development to digital health and artificial intelligence. Learn how this new model of collaboration can deliver value and lead to broader relationships between industry and academia.

Moderator: Seema Basu, PhD, Market Sector Leader, Innovation, PHS

Nathalie Agar, PhD, Research Scientist, Neurosurgery, BH; Associate Professor, Neurosurgery, Radiology, HMS

Paul Anderson, MD, PhD, Chief Academic Officer, BH; SVP, Research, BH; K. Frank Austen Professor of Medicine, HMS

Laurie Braun, MD, Partners Innovation Fellow, MGH and Boston Pharmaceuticals; Instructor in Pediatrics, HMS

David Chiang, MD, PhD, Research Fellow, BH; Innovation Fellow, Boston Scientific

David Feygin, PhD, Chief Digital Health Officer, Boston Scientific

Peter Ho, MD, PhD, CMO, Boston Pharmaceuticals

Harry Orf, PhD, SVP, Research, MGH; Principal Associate, HMS

 

Last Mile: Fully Implementing AI in Healthcare

This session will focus on how radiology and pathology specialties are currently applying AI in the clinic. Where will it be built out first? What are the barriers and how will these challenges be overcome?

Moderator: Keith Dreyer, DO, PhD, Chief Data Science Officer, PHS; Vice Chairman, Radiology, MGH; Associate Professor, Radiology, HMS

Katherine Andriole, PhD, Director of Research Strategy and Operations, MGH & BWH CCDS; Associate Professor, Radiology, HMS

Samuel Aronson, Executive Director, IT, Personalized Medicine, PHS

Peter Durlach, SVP, Healthcare Strategy & New Business Development, Nuance

Seth Hain, VP of R&D, Epic

Jonathan Teich, MD, PhD, Chief Medical Information Officer, InterSystems; Emergency Medicine, BH

 

Reimagining Disease Management

The management of disease has become vastly more challenging, both for patients and providers. AI-based technologies promise to improve and streamline patient care through a variety of approaches. This session will feature a discussion of these new tools and how they can enhance patient engagement and optimize care management.

Moderator: Sree Chaguturu, MD, Chief Population Health Officer, PHS; Assistant Professor, Medicine, HMS

Murray Brozinsky, Chief Strategy Officer, Conversa

Jean Drouin, MD, CEO, Clarify Health Solutions

Julian Harris, MD, President, CareAllies

Erika Pabo, MD, Chief Health Officer, Humana Edge; Associate Faculty, Ariadne Labs; Associate Physician, BH; Instructor, HMS

 

Standards and Regulation: The Emerging AI Framework

As the health care industry faces an explosion of AI-based tools, the FDA’s approach to these technologies is evolving. This session will focus on the agency’s approach to AI-based products, how to calculate the risk profile of these new technologies, and the challenges of securing adequate data rights.

Moderator: Brent Henry, Member, Mintz Levin

Bethany Hills, Member/ Chair, FDA Practice, Mintz Levin

Michelle McMurry-Heath, MD, PhD, VP, Global Regulatory Affairs and International Clinical Evidence, Johnson & Johnson Medical Devices

Bakul Patel, Associate Director, Digital Health, FDA

Michael Spadafore, Managing Director, Sandbox Industries

 

From Startup to Impact (Provider Solutions)

This session will introduce you to five leading startup companies who will each share their respective impact in delivery provider solutions in ten-minute pitches.

Moderator: Meredith Fisher, PhD, Partner, Partners Innovation Fund, PHS

Moderator: James Stanford, Managing Director, Fitzroy Health

William Grambley, COO, AllazoHealth

Gal Salomon, CEO, CLEW

Siddarth Satish, CEO, Gauss Surgical

Pelu Tran, CEO, Ferrum Health

Ed Zecchini, CIO, Remedy Partners

1:00 pm – 1:10 pm
1:10 pm – 2:00 pm
Bayer Ballroom

China: AI Enabled Healthcare Leadership

China’s health care system faces major challenges — and its population is aging more rapidly than nearly every other country. To help address these problems, the Chinese health technology sector is strongly embracing AI. What are the most exciting applications? What lessons does China’s early forays into AI-enabled patient care hold for other health care systems?

Moderator: James Bradner, MD
  • President, Novartis Institutes for BioMedical Research
  • Chief Innovation Officer, GE Healthcare
  • Analytics allowing higher throughput in China in Rural areas
  • Sepsis – detection is too late
  • data exhaust for facial recognition – anticipatory diagnosis
  • oncology tumor algorithm
  • CEO, Infervision
  • Medical imaging – four years to mature nodule detection
  • AI – no resale of data
  • Chairman and Co-Founder, Yidu Cloud
  • Medical records
  • Data privacy is personal consent if identification Passport level:
  • Doctor looking on Medical record need consent
  • Administration – clearance for access
  • Managing Partner, Qiming Venture Partners
  • AI HC companies execution to build companies
  • Valuation of all AI not only HC, dropped 30%
  • Real Doctor – 14 licensing for Internet medicine 90,000 patients a day are seen
  • Consumer EMR – Alibaba invested in
  • Investment in CRISPR
  • Invest in drug discovery in China
  • In China 150 programs of drug development of PD-1
  • Government  – 90% of patients go to Public Hospital which guard the data
  • Challenges AI in China — US – China Trade issue
  • CEO, Real Doctor Corporation Limited
  • Medical imaging 12 disease found from pictures build models to other 100 hospitals
  • small nodules detection
  • China-FDA no regulation established yet Learn from US FDA
2:00 pm – 2:30 pm
Bayer Ballroom

1:1 Fireside Chat: Mark Benjamin, CEO, Nuance

Moderator: Peter Slavin, MD
  • President, MGH; Professor, Health Care Policy, HMS
  • CEO, Nuance Communications
  • System produce NOTES from conversation, clinical language, notes read interactively by looking at other chart – LIVE EXAM more that an invoicing tool
  • patient case management made efficient
  • Documentation and Clinical notes embedded into the EHR enhance intelligence at Point-of-Care

 

2:30 pm – 3:00 pm
3:00 pm – 3:50 pm
Bayer Ballroom

Getting to the AI Investment Decision

The billions invested worldwide in AI-based health care technologies underscore the enthusiasm of global investors. But where are the greatest opportunities and what is the timeline to meaningful impact? In this panel, venture, private equity investors, and buy side analysts will discuss investment priorities, timelines, and key areas of interest

  • Partner, Partners Innovation Fund, PHS
  • When is the time right and when there is only a promise
  • VP, Venture and Managing Partner, Partners Innovation Fund, PHS
  • Looks like therapeutics but it is AI
  • Managing Director, Bain Capital Life Sciences
  • companies leveraging competencies
  •  Capital put to work what is it coming to do – specific value creation
  • Is the problem HC or an Academic Medical Center, i.e., MGH problem to solve
  • If no one at PHS willing to pay — let’s think again
  • Managing Partner, Polaris Partners
  • Data in Pharma companies are ready for AI application
  • algorithms and analytics
  • Value proposition
  • Language processing & ML – recognize patterns in consistant datasets – improve decision made in patient care
  • SVP, Strategy, Commercialization and Innovation, Amgen
  • Real data using AI for speeding drug discovery commercial application
  • predictive models for second MI with partner
  • Pilot study vs scaling up
  • Managing Director, Healthcare Group, Goldman Sachs
  • As AI algorithm mature, labor intensity curbed by AI
  • IPO
  • consolidation of big pharma
  • Partner, Google Ventures – started in 2008/9; Instructor in Medicine, BH
  • data quality needed for AI to avoid bias
  • Pharma is interested in Drugs not in Targets
  • Translator between technology and healthcare
  • Teach computer the rules to go then beating its creator unanticipated modes
  • IT is different in various industries more than West Coast vs East Coast
3:50 pm – 4:20 pm
Bayer Ballroom

1:1 Fireside Chat: Robert Bradway, CEO, Amgen

  • Partner, Atlas Venture
  • CEO, Amgen
  • DeCode Genetics acquired by Amgen
  • AI is in the beginning Rapata and Evenity (romosozumab) risk of fractures – review large images archives
  • Migraine only digital health  – this is not a big area for Amgen
  • Transparency
  • Encouraged to role back the Rebate Program the sickest pay to high – policy changes
  • Part 4
  • Rapata – lower LDL reduce risk for stroke MI 600Billion fighting Heart disease – price lowered 60% patients are directed to the more expensive product
  • Investment in Biosimilars and biologics made available free resources
  • risk is Washington, generics may become the rule for biologics
  • no favor innovating products vs Biosimilars
  • ObamaCare create 12 years of data exclusivity for biologics
  • 90% of prescription is generic products
  • cost of CVD in 2019 is a fraction of the cost 15 years ago
  • CURE – is used for Cancer at what price HEP C – is a cure very expansive
  • Meaning of innovations create frameworks for saving live
4:20 pm – 5:10 pm
Bayer Ballroom

Consumer Healthcare and New Models of Care Delivery

Al is powering a revolution in consumer health care, giving patients a deeper role in monitoring their own health and spawning new models of care delivery. Many health care organizations are increasingly focused on creating a digital “front door” for patients – a single gateway to mobile apps and other online services. Panelists will also discuss the role of remote monitoring and virtual care programs as well as the role of Al in care redesign and workflow.

Moderator: Diana Nole
  • CEO, Wolters Kluwer Health
  • President, Global Strategy Group, Samsung; Founder, CareVisor
  • Real time sensing to deliver realtime care plan: Human Avatar
  • AI is hidden
  • communication varies by generations phone vs SMS
  • VP and Global CTO, Sales, Dell EMC
  • IOT – scale
  • social media – peer pressure
  • President, Health Platforms, Verily Life Sciences
  • AI applied in diet management with images of snacks
  • Co-production of Health 50s-60s concept Co-Production health by patients give patients information and they will co-produce their healthier life style
  • VP and Chief Health Officer, IBM Corporation
  • AI continues to improve – actionable insights
  • AI augmented humanity
  • In China a Team of oncologist meet with entire families to discuss plan of care Cancer patients for GrandMa,
  • SVP, Head of Innovation and Health Equity, Microsoft Healthcare
  • AI – sequence T cells
5:15 pm – 5:25 pm
Bayer Ballroom

BioBank Award Announcement

  • Third place MGH – Computational Pathology
  • First Prize – $12,000 UPittsburg – Dept Biomedical Informatics – principal components
  • First Prize – IBM Center for Computational Health – supervised algorithm
5:30 pm – 6:30 pm

 

Read Full Post »


LIVE Day One – World Medical Innovation Forum ARTIFICIAL INTELLIGENCE, Boston, MA USA, Monday, April 8, 2019

 

www.worldmedicalinnovation.org

 

The Forum will focus on patient interactions across care settings, and the role technology and data can play in advancing knowledge discovery and care delivery. The agenda can be found here.

https://worldmedicalinnovation.org/agenda/

Leaders in Pharmaceutical Business Intelligence (LPBI) Group

represented by Founder & Director, Aviva Lev-Ari, PhD, RN will cover this event in REAL TIME using Social Media

@pharma_BI

@AVIVA1950

@PHSInnovation

#WMIF19 

@evanKristel 

Monday, April 8, 2019

7:00 am – 8:00 am
7:00 am – 5:00 pm
8:00 am – 9:40 am
Bayer Ballroom

First Look: Round 1

Nine rapid fire presentations on the applications of AI in Clinical Care

To view speakers and topics, click here.

Henry Chueh, MD

Director, MGH Lab of Computer Science, MGH; Assistant Professor, Medicine, HMS

Dxplain: Expanding diagnostic horizons

 

Synho Do, MD

Director, Laboratory of Medical Imaging and Computation (LMIC), MGH; Assistant Professor, HMS

Leveraging a Deep-Learning Algorithm for the Detection of Acute Intracranial Hemorrhage

 

Laura Germine, PhD

Director, Laboratory for Brain and Cognitive Health Technology, McLean; Assistant Professor, Psychiatry, HMS

The Next Generation of Cognitive and Behavioral Assessment

 

Satrajit Ghosh, PhD

Research Associate, MEE; Principal Research Scientist, MIT; Assistant Professor, Otolaryngology, HMS

Assistive Intelligent Technologies for Brain Health

 

Chris Sidey-Gibbons, PhD

Co-Director, PROVE Center, BH; Member of Faculty, HMS

Three Computational Techniques and One Tool to Bring the Patient Voice into Care

 

Xudong Huang, PhD

Co-Director, Neurochemistry Laboratory; MGH; Associate Professor, Psychiatry, HMS

Leveraging Artificial Intelligence for Brain Drug Discovery

 

Tina Kapur, PhD

Executive Director, Image-Guided Therapy, BH; Assistant Professor, Radiology, HMS

Using AI to Better Visualize Needles in Ultrasound-Guided Liver Biopsies

 

Bharti Khurana, MD

Director, Emergency Musculoskeletal Radiology, BH; Assistant Professor, HMS

 

 

Vesela Kovacheva, MD, PhD

Attending Anesthesiologist, BH; Instructor, Anesthesiology, HMS

Harnessing the Power of Machine Learning to Automate Drug Infusions in the OR and ICU

Constance Lehman, MD, PhD

Chief, Breast Imaging Division, MGH; Professor of Radiology, HMS

AI-Based Care Delivery: A New Paradigm for Curing Cancer

 

Lisa Nickerson, PhD

Director, Applied Neuroimaging Statistics Lab, McLean; Assistant Professor, HMS

Using Digital Phenotyping and Machine Learning to Forecast, Detect, and Prevent Drug Overdose Deaths

 

Federico Parisi, PhD

Research Fellow, Wyss Institute for Biologically Inspired Engineering, SRN

Mobile Health Technologies for Monitoring Motor Fluctuations in Patients with Parkinson’s Disease

 

Stuart Pomerantz, MD

Director, Neuro-CT, Neuroradiology, MGH; Instructor, HMS

AI-Powered Diagnostic Reporting for Spinal MRI of Degenerative Disease

 

Sandro Santagata, MD, PhD

Assistant Professor, Pathology, BH, HMS

 

Joseph Schwab, MD

Chief, Orthopaedic Spine Surgery, MGH; Associate Professor, HMS

Artificial Intelligence for Diagnosis and Management in Spine Surgery

 

Hiroyuki Yoshida, PhD

Director, 3D Imaging Research, MGH; Associate Professor, Radiology, HMS

 

Nazlee Zebardast, MD

Instructor, Ophthalmology, MEE, HMS

 

Li Zhou, MD, PhD

Associate Professor/Lead Investigator, BH; Associate Professor, HMS

 

Machine Learning and NLP to Track Disease Progression and Predict Health Outcomes

Moderator: Giles Boland, MD
  • Chair, Department of Radiology, BH; Philip H. Cook Professor of Radiology, HMS
Moderator: Trung Do
  • VP, Business Development, Innovation, PHS

Henry Chueh, MD

  • wrong diagnosis, leading malpractice claims
  • 1 out of 6 new diagnosis are wrong
  • help clinicians to make 1st diagnosis and every time correct — what need be considered
  • fever, rash, arthrisis (painful swallen joint) – no correct diagnosis
  • Adult Still disease – symptoms trigger condition –
  • DXplain Knowledge base + algorithms curated over 25 yr
  • >1 Million relationships
  • probabilistic inference algorithms
  • Amazon Web Services – micro services on Amazon Web
  • UI widgets for Web apps – mobile prototype
  • 20million hits per month
  • DXplain consumer, clinician, hospitals, payer, malpractice insurer

 

Synho Do, PhD

  • AI and DL for Stroke Patient management detection of acute intracranial haemorrhage from small dat sets
  • 1 of every 10 death is a Stroke caused, 5.8 million people die of Stroke Stroke is a medical emergency, CT Scan
  • Spotting brain bleeding after
  • Deep Learning algorithms – explainable AI  – human mimiking algorithm developed @MGH
  • Explainable AI – Multi-window mixing & multi-slice mixing is in PACS @MGH
  • commercial opportunity: Near stroke detection
  • @MGH Stroke with AI algorithms Patent IP @PartnerInnovation seeking funding for Stroke management

Laura Germine, PhD

  • Next generation of behavior assessment
  • in Psychiatry – neuropsychiatry
  • Problem of measurement of innovation with validity needed – Tools to measure and have outcomes
  • Unreasonable effectiveness of Good Data : Math achievement – visual-spatial attention
  • Looking for partners

Satrajit Ghosh, PhD

  • Mental health 1 in 4 adults 18% of adolescence 13% of children
  • first treatment effective only in 25% of cases
  • Brain structure and Function – using MR – observed behaviors – using Voice, speaking is a very complex activity
  • Talk intent emotions – window into the mind
  • Speech

Xudong Huang, PhD

  • Brain Drug Discovery – leveraging AI
  • Major depressive DIsorder ( MDD) – 16 million in US 210 Billion a year treatment burden
  • Alzheimer’s DIsease  – 5.8 million AS in US – $290 in 2019 a year treatment burden
  • Potential druggable for MDD and AD
  • Tryptophan-Kynurenina pathway
  • Secreted Protein Acidic and Cysteine rich
  • AI-Powered Drug Discovery Platform – AtomNet
  • Preclinical drug discovery and development
  • Screened 10MIllion compounds – 48 inhibitors for tryptophan-catabolizing enzymes in
  • Tryptophan-Kynurenina pathway

Tina Kapur, PhD

  • AI to visualize needles in UltraSound-guided (US) liver biopsy – safer to patient and easier for the physicina
  • mass in liver suspected to be from a metastasis in the pancreas
  • AI to enable the MD to see the needle completely independent of the US technician
  • Benefits if available to all performers of liver biopsy
  • Patients: Benefit from location of tissue biopsy sampling
  • prostate needle in MRI
  • Button labelled Needle, MD turn on/of button
  • navigation systems not in use
  • 95% proceedures done free hand
  • 1 Million US guided liver biopsy/yr, growing @4%
  • manufacturing of US equipment to be interested to embed

Bharti Khurana, MD

  • Home is the most dangerous place for women killing of women hit by husband. ages 25 to 38 – fracture of bone IPV – Intimate Partner Violence – 1 in 4 women and 1 in 9 men IPV is preventable under reporting
  • Tybanny of the Urgent
  • clinical decision support to predict risk probability automate alerts 95% 50% 15% – Probability of IPV – insivible to visible
  • empower healthcare providers
  • reduce ER volume will reduce cost

Vesela Kovacheva, MD, PhD

  • Titrating drug infusions – Personalized for patient safety reduce med error
  • Titrating drug infusions – automation system from anestesia – function automonically
  • local anestatic for Cesearian section – BP drog when spinal administration of anestatic agent
  • calculate every minure – 20 minutes are critical from drug infusion
  • decision to administer vasopressors is taken evey minute on the bP
  • Rural areas one anestosiolog suverviser three OR at the same time
  • 1.25 million C-section
  • 75% develop low BP
  • complications in babies decreased BP – tachepnis in neonatal – NICU 100Million $ per year.
  • develop same algorithms for propofol in sedetion and insulin in ICU
  • other surgeries – knee, hip, spinal

Constance Lehman, Md, PhD

  • Breast Cancer Out of 2 Billion women 2million will be diagnosed with breast cancer
  • screening will prevent development
  • current tools of mamography – no single interpretation and shortage
  • memograph vs Future risk of BC development
  • Deep Learning model; Training model consequitive memograms Risk model developed – AI technology on memograpm 0.71 when other factors added
  • DIverse races – RAce blind AI model
  • AI model of diagnosis in one year after the memogram taken
  • breast density – imager certified, 6% are dense, 85% and every number in between
  • Expertise: MGH, MIT, Prior failure of CAD
  • Patents for commercialization beyond MGH

Lisa Nickerson, PhD

  • 70,000 drug overdose, 50,000 opioids related
  • Death from prescription opioids is on the increase after 2013 – fentanyl – causing overdose
  • prescription opioids overdose Prevention strategies:
  • Targeted Naloxone distribution
  • Medication assisted treatment
  • Fentanyl screening in Tox tests
  • 911 good Samaritan laws
  • Syringe services programs

Federico Parisi, PhD

  • Mobile Health Applications – Monitoring motor fluctuation in Parkinson’s Disease (PD)
  • 7 – 10Million WOrldwide, 1 Million in the US,
  • dopamine-producing neuron
  • main medication in early stage – Levodopa
  • Need an objective and continuous monitoring toool for tacking the symptoms’ dynamics
  • mHealth for monitoring PD – mimiking clinical evaluations mail limitations: Deendency on standardized motor tasks in sufficient time resolution in symptoms severity during ADLs

Stuart Pomerantz, MD

  • DeepSPINE – Challenges of Lumbar Spine Imaging: Lumbar stenosis MR interpretation Suboptimal radiology
  • DeepSPINE – end-to-end processing pipeline for clinical deployment
  • AI-Powered Diagnosis & Reporting Solutions
  • DeepSPINE: Slice Angle Optimization
  • Predict disease severity/interpretation time
  • Route of optimal staffing
  • DeepSpine Data Layer Multi-Format Reporting: Traditional Text vs Tabular Image-Enhanced
  • Portfolio of applicationsWho benefits from MRI
  • Avoid unneccesary imaging – Clinical Decision-aking
  • Better predict who needs surgery

Sandro Santagata, MD, PhD

  • Tissue imaging quant pathology
  • DL for Mass spectrometry – full spectral resolution
  • interoperative paradigm – patient, biopsy, frozen tissue Tissue cyclic immunoflorescence hi Dimensional pathology
  • Human Tumor Atlas Network (HTAN) – phenotype cancers

Joseph Schwab, MD

  • Orthopedic Spin surgery – 1/2 million lumber fusion surgery, 5% complications $1.8 Billion
  • Data science in Spine today – algorithms based on 35,000 patients cases annotated
  • ML algorithm which Pations will need opioids after fusion
  • Predicted Probability – cost-benefit ration – Benefit to patient
  • Cervical stenosis C5-C6 – patient list of current medication – Prediction of a patient probability to need opioids after spinal surgery
  • Spinal metastasis – Survival prediction – is surgery needed if survival is few months?
  • Complications of hip replacement Perspective: Provider or Insurer
  • SORG-AI.com

Chris Sidey-Gibbons, PhD

  • Patient-reported data
  • identification of treatment satisfaction with care, quality of life, mental health,
  • ONE Questionnaire – filled by Patient – used by psychiatry since 1950
  • Clinical meaning, ML, Computer Adaptive Diagnosis (CAT algorithm) , NLP, response burden
  • ML – improve clinical meaning of Patient reported data, train algorithm – likely outcomes
  • Reconstructive surgery following mastectomy – survey of women
  • Plastic surgery Report – to improve CAT algorithm
  • imPROVE
  • InSpire

Hiroyuki Yoshida, PhD

  • Colon screening 150,000 new cases in the US, 55,000 death, 14B spent in the US
  • CT colonography (CTC)  & Colonoscopy
  • @MGH Laxative-free CT colonography: Oral oral contrast  followed by GI CT Scanning
  • GAN – generative adversarial networks: AI virtual bowel cleansing + AI small polyp detection
  • algorithms remove fecal material
  • Sensitivity: AI-latex-free – 96% sensitivity vs. CTC 46% and Laxative 67%

Nazlee Zebardast, MD

  • Deep learning for glaucoma detection – prevent
  • optic nerve disease, irriversible blindness
  • 76 Million 11 Million bilateral blind
  • +50% glacauma not diagnosed in the US – delay progression by screening
  • No reliable out reach programs – USPSTF recommended against screening
  • Deep learning used for Glaucoma detection _ Larger inter-reader interpretation variation
  • Improve reference standard
  • genetic risk of glaucoma
  • intaocular pressure – modifiable factor
  • Diabetic or non diabetic retinopathy
  • Age, gender, smokin SBP, refractive error
  • What the machine pays attention
  • high IOP and high genetic risk
  • commercialize DL based screening tool for glaucoma – 140 Million in the US
  • The market: 120 million age 30 to 40
  • Cost saving S5.8 Billion

Li, Zhou, MD, PhD

  • Palliative care ML and improve value of care
  • end of life care for Dementias: Latent topic modeling and trend analysis using clinical notes
  • reduce anxiety and depression patient more likely to have wishes known
  • Who are the patients that will benefit the most from palliative care
  • determine the right time for this intervention
  • free-text EHR data
  • Physical function status: Nutrition, feeding, swallowing
  • Commercialization – MTERMS Lab – pharmacovigilance, speech recognition, information extraction and decoding data mining

 

9:40 am – 9:55 am
9:55 am – 11:35 am
Bayer Ballroom

First Look: Round 2

Nine rapid fire presentations on the applications of AI in Clinical Care

To view speakers and topics, click here.

11:30 am – 11:45 am
11:45 am – 1:00 pm

Discovery Café Sessions

Lunch with Experts: Intensive sessions addressing cutting-edge artificial intelligence topics.

Applying AI to Save Lives During the Opioid Crisis

The U.S. is in the throes of a devastating epidemic of opioid addiction and overdose — some 130 people die nationally every day from opioids, says the National Institute on Drug Abuse. With a total economic cost of more than $78 billion a year, AI is being harnessed to develop new tools that can help alleviate this national crisis. This session will discuss AI-based strategies that academic and industry teams are leveraging to help clinical and public health officials better predict, identify, and treat opioid addiction, and also data privacy concerns.

Moderator: Thomas Sequist, MD, Chief Quality & Safety Officer, PHS

Bob Burgin, CEO, Amplifire Healthcare Alliance

Carm Huntress, CEO, RxRevu Inc

Sarah Wakeman, MD, Medical Director, Substance Use Disorder Initiative, MGH; Assistant Professor, Medicine, HMS

Scott Weiner, MD, Director, Brigham Comprehensive Opioid Response and Education (B-CORE) Program, BH; Assistant Professor, HMS

 

Community Hospitals: Key Component in Healthcare Transformation

Community hospitals are the largest sources of patient care in the U.S. As such, they represent a frontier in the transformation of health care. How are these organizations using AI and digital technologies to drive transformation? What are the distinctions from academic medical centers? This session will address these and other topics that impact community hospitals.

Moderator: Michael Jaff, DO, President, NWH, PHS, Professor of Medicine, HMS

Fabien Beckers, PhD, CEO, Arterys

Joanna Geisinger, CEO, TORq Interface

John Miller, MD, Director, Retinal Imaging, MEE; Assistant Professor, Ophthalmology, HMS

Lee Schwamm, MD, Director, Center for TeleHealth and Exec Vice Chair, Neurology, MGH; Professor, Neurology, HMS

Tal Wenderow, CEO, Beyond Verbal

 

Digital Management of Diabetes

Across the spectrum of patient care, the management of diabetes has been flooded with new technology and treatment options for both type 1 and type 2 diabetes – there is a range of new devices and software, including automatic insulin infusion systems, glucose sensors, AI-based algorithms and decision support tools, with an artificial pancreas on the horizon. This session will focus on these areas and clinical use cases that highlight the value of AI.

Moderator: Deborah Wexler, MD, Clinical Director, Diabetes Center, MGH; Associate Professor, HMS

Marie McDonnell, MD, Section Chief and Director, Diabetes Program, BH; Lecturer, HMS

Michael Meissner, PhD, CTO and VP, MED, Sanofi

Joshua Riff, MD, CEO, Onduo

Marie Schiller, VP, Connected Care and Insulins Product Development and Site Head, Cambridge Innovation Center, Eli Lilly

 

AI and Its Impact on the Future of Emergency Care

There are over 136 million Emergency Department visits annually in the U.S. providing 24/7 unscheduled treatment for problems from minor illness to life threatening traumatic injuries.  Emergency department care teams provide high quality, safe care in an efficient fashion.  In this session, we consider the future of AI in emergency care from the initial decision to seek emergency care, to diagnostic processes within the ED and final disposition decision..  From chat bots for patient triage, telehealth for patient visits to machine learning outcome prediction, we will consider how these novel technologies will impact emergency care delivery.

Moderator: Adam Landman, MD, VP and CIO, BH; Associate Professor of Emergency Medicine, HMS

Peter Chai, MD, Assistant Professor, Emergency Medicine, BH, HMS

Emily Hayden, MD, Attending Physician, Emergency Medicine, MGH; Instructor, Surgery, HMS

Kohei Hasegawa, MD, Attending Physician, Emergency Medicine, MGH; Associate Professor, Emergency Medicine, HMS

Sean Kelly, MD, CMO, Imprivata; Assistant Professor, Emergency Medicine, HMS

Bijoy Sagar, VP, Chief Digital Technology Officer, Stryker

 

Mental Health, Smartphone Apps and the Promise of AI

Patients can face significant barriers when it comes to accessing high-quality, evidence-based treatment for mental illness. AI-enabled technologies, including smartphone-based tools, that may help close this treatment gap for patients worldwide. This session will focus on efforts to develop smartphone apps and other tools, including those designed to help predict patients’ moods and provide cognitive behavioral therapy.

Moderator: Sabine Wilhelm, PhD, Chief of Psychology; Director, OCD and Related Disorders Program, MGH; Professor, Psychology, HMS

Jennifer Gentile, PsyD, SVP, US Clinical Operations, Ieso Digital Health

Thomas McCoy, MD, Director of Research, Center for Quantitative Health, MGH; Assistant Professor, Psychiatry and Medicine, HMS

Christopher Molaro, CEO, Neuroflow

David Silbersweig, MD, Chairman, Department of Psychiatry, BH; Stanley Cobb Professor of Psychiatry, HMS

Jeremy Sohn, VP, Global Head of Digital Business Development and Licensing , Novartis

 

From Startup to Impact (Pharma and Diagnostics)

This session will introduce you to five leading start-up companies who will each share their respective impact in the pharmaceutical and diagnostic realms in 10-minute pitches.

Moderator: James Brink, MD, Radiologist-in-Chief, MGH; Juan M. Taveras Professor of Radiology, HMS

Moderator: James Nicholls, Managing Director, Fitzroy Health

Sarah Beeby, EVP, GM Lifesciences, Clinithink

Charles Cadieu, PhD, CEO, Bay Labs

JB Michel, PhD, SVP Data Science & GM USA, BenevolentAI

Art Papier, MD, CEO, VisualDx

Alex Zhavoronkov, PhD, CEO, Insilico Medicine, Inc

1:00 pm – 1:15 pm
1:15 pm – 1:30 pm
Bayer Ballroom

Opening Remarks

  • Interim President and CEO, Chief Academic Officer, PHS; Laurie Carrol Guthart Professor of Medicine, HMS; 2019 Forum Co-Chair
1:30 pm – 2:00 pm
Bayer Ballroom

AI Strategy: AI from the Top

As the potential of AI comes into clearer view, many academic medical centers are taking notice and crafting institutional strategies for incorporating AI into clinical practice. But where are the most meaningful opportunities? What are the biggest challenges? And, importantly, will patient care be noticeably different — better, more available, and/or less costly?

  • Board Member, PHS; President Emerita and Professor of Neuroscience, MIT
  • Cross institutional cooperation is advocated
  • AI – what it will deliver in 2 years
  • what is the role of the Top management
  • how we mwasure how we do
  • Ethics and bias  in AI vs non-AI World
  • Chief Data Science Officer, PHS; Vice Chairman, Radiology, MGH; Associate Professor, Radiology, HMS
  • scaling Machine learning focused areas high accuracy, training ground truth, today the humans establish it in the future with AI ground truth will be created by AI
  • how to handle and move the intelligence and discoveries across units
  • Chief Digital Health Officer, PHS
  • Digitization of documentation – recording the session, Nauance – AI does the borden of communication translation
  • Easy button comparison of f patients wwith same ocndition what was the treatment
  • Chief Clinical Officer, PHS; Professor of Medicine, HMS; 2019 Forum Co-Chair
  • Future 5-10 years EHR is dehumanizing at present but with AI EHR will humanize again the relations of Physician and Patients

 

2:00 pm – 2:30 pm
Bayer Ballroom

1:1 Fireside Chat: Jensen Huang, CEO, NVIDIA

Introduction by: Cathy Minehan
  • Managing Director, Arlington Advisory Partners; Chairman, Board of Trustees, MGH
  • Chief Data Science Officer, PHS; Vice Chairman, Radiology, MGH; Associate Professor, Radiology, HMS
  • CEO, NVIDIA, established in 1993 graphics, Genomics analysis
  • storage data validation and
  • AI is reinventing computer graphics taught a NN to produce animation by virtual reality in robotics
  • in next three year: Crypo-currency was not foreseen
  • Data Science ingesting data , processing doing analytics
  • RAPIDS – open source data centers clouds and the edge working together
  • AI needs to be at the edge computing to be create at the edge not in the Cloud
  • self driving cars computation odne at the edge
  • Redundence and diversity – approach is diverse
  • In Radiology – democratization of AI announced today with NVIDIA & Partners
  • Driver intervene, Radiologist will intervene
  • Concept of “Beta” – Cloud application is in Beta
  • SW: data driven algorithm written by AI and know to learn amazing results
  • Conditions for NVIDIA to succeed: Speed, SW defined, pipeline flow data curated validated
  • expertise in the company
  • In 5 years: breakthrough NLP – summarize what was said
  • Curations done by AI
  • One shot learning – AI contextual aware Knowing who goes where, when and what acronyms are
  • AI: is software – yes SW that writes SW AI is automation of Automation

 

2:30 pm – 2:45 pm
Bayer Ballroom

Remarks: The Honorable Charlie Baker

Introduction by: Scott Sperling
  • Co-President, Thomas H. Lee Partners; Chairman of the Board of Directors, PHS
  • Governor of the Commonwealth of Massachusetts
  • AI to assist practitioners in their decisions
  • Information explotions to clinician
  • medical infrastructure needs AI
  • Healthcare is held to a higher standard, people believe in Practitioners – Healthcare is held in very high esteem

 

2:45 pm – 3:35 pm
Bayer Ballroom

Real World Evidence and Trial Optimization in the AI Era

AI is a tool for conducting faster, more efficient clinical trials. Panelists will discuss how AI-enabled methods can further adaptive trial capabilities, trial design and trial management.

Moderator: Thomas Lynch, MD
  • EVP and CSO, R&D, Bristol-Myers Squibb
  • why sharing data is so hard?
  • IBM Watson – PDF can be read by Watson and come out with a Diagnosis
  • Deputy Commissioner, FDA
  • AI assists in recruitment
  • Modernization of clinical trial is acknowledged
  • Data standards for EHR oncology context
  • EVP MA&PV and Bayer CMO, Bayer AG
  • control arms in rare diseases
  • diagnostics in hypertension
  • drug safety – #AI works
  • Chief Architect, Microsoft Healthcare
  • sharing data semantic interoperability is available
  • No clinical data model
  • Which symptoms actual were experienced?
  • Blockchain
  • CEO, My Own Med Inc.
  • Wearable Pharma is adding this dimens
  • Executive Director, Clinical Trials Office, PHS; Associate Professor of Medicine, HMS
  • computation, pattern recognitions to make CT more efficient
  • competitive model among sponsors hinders data sharing
3:35 pm – 4:25 pm
Bayer Ballroom

AI Driven Value-Based Care

As providers embrace value-based approaches, the demands of clinical data collection, assessment, and information-sharing loom large. In this data-driven environment, clinicians must sift through ever-growing pools of information that can exceed the limits of human capability. An assortment of AI-based solutions is now emerging that may offer some relief. Panelists will discuss how these approaches are helping to support better, more personalized care, and the challenges faced by clinicians and managers for effective adoption.

Moderator: Timothy Ferris, MD
  • CEO, MGPO; Professor of Medicine, HMS
  • CEO, American Heart Association
  • guideline on HTN, 1/2 million wake up with HTN a day after guidelines were enacted
  • AI will not be able to replace a clinician encouraging a patient
  • AI to free time of HC professional
  • EVP, President, Network Solutions, Change Healthcare
  • 1 trillion $ is wasted Healthcare is not consumer friendly #AI has opportunities to innovate home-based solutions
  • consumer focus technologies hand held devices
  • Levers
  • CEO, NHS England
  • AI can free time for health professionals
  • diagnostics
  • productivity in Healthcare has impact of the entire econommy US – 3 trillions size of HC sector
  • 2 1/2 million literature new to clinician evry year – AI will assist
  • Clinician explainability is very important
  • AI to benefit Healthcare for all
4:25 pm – 5:15 pm
Bayer Ballroom

Cardiovascular Care: Reinvented Through AI

Cardiovascular diseases remain the leading cause of death worldwide and an expense, making this area ripe for AI-enabled innovations. Teams are pursuing a range of AI-based tools in cardiovascular medicine: including AI-powered drug discovery and diagnostics to automated cardiac image analyses and AI-guided care delivery pathways. Panelists will discuss where AI is having a sizeable impact. The discussion will also include the perspectives of a patient who benefited from AI-enabled cardiovascular care.

  • Vice Chair for Scientific Innovation, Department of Medicine, BH; Associate Professor of Medicine, HMS
  • SVP, Global Head of Digital and Analytics, Sanofi
  • COTY in Copenhagen – AI augment capability of EMTs dispatcher is prompted with questions to decide if this call is Heart arrest caving few minutes for EMT response
  • Patient
  • Independent Recording Engineer Burke Recording
  • President, Bayer Pharma Americas Region, Bayer
  • In-silicon modeling is AI based and shorten cycle of drug discovery
  • Bridge clinical care and with clinical trials
  • Challenge island of dat are disconnected,
  • Chief Cardiovascular Imaging, MGH; Professor, Radiology, HMS

 

  • To see a neurologist you need to have an MRI done already
  • Chest CT, Abdominal CT Chest X-ray — done
  • CVD CT report five pages long, prognostics — AI will tell MD what medication to suggest
  • clinical care more standardized
  • AI in clinical trial is a big premise
  • No more trials if perpatient the cost id more than $5,000
  • AI is a tool to enable lower cost clinical trials
  • imaging data sharing in what ever form
  • ML and AI at all Radiology conferences
  • QA criteria – what is quality data, to inform care
  • EVP/GM, Healthcare and Life Sciences, Persistent Systems
  • How to use AI clinical work flow goal – to be sw driven AI is a component
  • large systems sw automation data and platform dat acapture is very importnat
5:15 pm – 5:45 pm
Bayer Ballroom

1:1 Fireside Chat: Seema Verma, Administrator, Centers for Medicare & Medicaid Services

Moderator: Sree Chaguturu, MD
  • Chief Population Health Officer, PHS; Assistant Professor, Medicine, HMS
  • Administrator, Centers for Medicare and Medicaid Services
  • 2020 20% of all expenses spent will be on Healthcare in the US
  • Gov’t was a barrier to innovations
  • initiative of cutting regulations
  • innovation – how we pay providers for value produced vs regulation that stay in the way
  • gov’t slow to respond: FDA approval and CMS access to treatment and reimbursement
  • Analysis of drug a patient takes, CMS – quality, medical record given to patient across all providers they use and be able to give to a new provides all historical data
  • Data privacy and security
  • Innovators in Colorado – health care cost need be lowered in a major way
5:45 pm – 6:45 pm

Read Full Post »

Older Posts »