Feeds:
Posts
Comments

Archive for the ‘Artificial intelligence applications for cardiology’ Category

Science Policy Forum: Should we trust healthcare explanations from AI predictive systems?

Some in industry voice their concerns

Curator: Stephen J. Williams, PhD

Post on AI healthcare and explainable AI

   In a Policy Forum article in ScienceBeware explanations from AI in health care”, Boris Babic, Sara Gerke, Theodoros Evgeniou, and Glenn Cohen discuss the caveats on relying on explainable versus interpretable artificial intelligence (AI) and Machine Learning (ML) algorithms to make complex health decisions.  The FDA has already approved some AI/ML algorithms for analysis of medical images for diagnostic purposes.  These have been discussed in prior posts on this site, as well as issues arising from multi-center trials.  The authors of this perspective article argue that choice of type of algorithm (explainable versus interpretable) algorithms may have far reaching consequences in health care.

Summary

Artificial intelligence and machine learning (AI/ML) algorithms are increasingly developed in health care for diagnosis and treatment of a variety of medical conditions (1). However, despite the technical prowess of such systems, their adoption has been challenging, and whether and how much they will actually improve health care remains to be seen. A central reason for this is that the effectiveness of AI/ML-based medical devices depends largely on the behavioral characteristics of its users, who, for example, are often vulnerable to well-documented biases or algorithmic aversion (2). Many stakeholders increasingly identify the so-called black-box nature of predictive algorithms as the core source of users’ skepticism, lack of trust, and slow uptake (3, 4). As a result, lawmakers have been moving in the direction of requiring the availability of explanations for black-box algorithmic decisions (5). Indeed, a near-consensus is emerging in favor of explainable AI/ML among academics, governments, and civil society groups. Many are drawn to this approach to harness the accuracy benefits of noninterpretable AI/ML such as deep learning or neural nets while also supporting transparency, trust, and adoption. We argue that this consensus, at least as applied to health care, both overstates the benefits and undercounts the drawbacks of requiring black-box algorithms to be explainable.

Source: https://science.sciencemag.org/content/373/6552/284?_ga=2.166262518.995809660.1627762475-1953442883.1627762475

Types of AI/ML Algorithms: Explainable and Interpretable algorithms

  1.  Interpretable AI: A typical AI/ML task requires constructing algorithms from vector inputs and generating an output related to an outcome (like diagnosing a cardiac event from an image).  Generally the algorithm has to be trained on past data with known parameters.  When an algorithm is called interpretable, this means that the algorithm uses a transparent or “white box” function which is easily understandable. Such example might be a linear function to determine relationships where parameters are simple and not complex.  Although they may not be as accurate as the more complex explainable AI/ML algorithms, they are open, transparent, and easily understood by the operators.
  2. Explainable AI/ML:  This type of algorithm depends upon multiple complex parameters and takes a first round of predictions from a “black box” model then uses a second algorithm from an interpretable function to better approximate outputs of the first model.  The first algorithm is trained not with original data but based on predictions resembling multiple iterations of computing.  Therefore this method is more accurate or deemed more reliable in prediction however is very complex and is not easily understandable.  Many medical devices that use an AI/ML algorithm use this type.  An example is deep learning and neural networks.

The purpose of both these methodologies is to deal with problems of opacity, or that AI predictions based from a black box undermines trust in the AI.

For a deeper understanding of these two types of algorithms see here:

https://www.kdnuggets.com/2018/12/machine-learning-explainability-interpretability-ai.html

or https://www.bmc.com/blogs/machine-learning-interpretability-vs-explainability/

(a longer read but great explanation)

From the above blog post of Jonathan Johnson

  • How interpretability is different from explainability
  • Why a model might need to be interpretable and/or explainable
  • Who is working to solve the black box problem—and how

What is interpretability?

Does Chipotle make your stomach hurt? Does loud noise accelerate hearing loss? Are women less aggressive than men? If a machine learning model can create a definition around these relationships, it is interpretable.

All models must start with a hypothesis. Human curiosity propels a being to intuit that one thing relates to another. “Hmm…multiple black people shot by policemen…seemingly out of proportion to other races…something might be systemic?” Explore.

People create internal models to interpret their surroundings. In the field of machine learning, these models can be tested and verified as either accurate or inaccurate representations of the world.

Interpretability means that the cause and effect can be determined.

What is explainability?

ML models are often called black-box models because they allow a pre-set number of empty parameters, or nodes, to be assigned values by the machine learning algorithm. Specifically, the back-propagation step is responsible for updating the weights based on its error function.

To predict when a person might die—the fun gamble one might play when calculating a life insurance premium, and the strange bet a person makes against their own life when purchasing a life insurance package—a model will take in its inputs, and output a percent chance the given person has at living to age 80.

Below is an image of a neural network. The inputs are the yellow; the outputs are the orange. Like a rubric to an overall grade, explainability shows how significant each of the parameters, all the blue nodes, contribute to the final decision.

In this neural network, the hidden layers (the two columns of blue dots) would be the black box.

For example, we have these data inputs:

  • Age
  • BMI score
  • Number of years spent smoking
  • Career category

If this model had high explainability, we’d be able to say, for instance:

  • The career category is about 40% important
  • The number of years spent smoking weighs in at 35% important
  • The age is 15% important
  • The BMI score is 10% important

Explainability: important, not always necessary

Explainability becomes significant in the field of machine learning because, often, it is not apparent. Explainability is often unnecessary. A machine learning engineer can build a model without ever having considered the model’s explainability. It is an extra step in the building process—like wearing a seat belt while driving a car. It is unnecessary for the car to perform, but offers insurance when things crash.

The benefit a deep neural net offers to engineers is it creates a black box of parameters, like fake additional data points, that allow a model to base its decisions against. These fake data points go unknown to the engineer. The black box, or hidden layers, allow a model to make associations among the given data points to predict better results. For example, if we are deciding how long someone might have to live, and we use career data as an input, it is possible the model sorts the careers into high- and low-risk career options all on its own.

Perhaps we inspect a node and see it relates oil rig workers, underwater welders, and boat cooks to each other. It is possible the neural net makes connections between the lifespan of these individuals and puts a placeholder in the deep net to associate these. If we were to examine the individual nodes in the black box, we could note this clustering interprets water careers to be a high-risk job.

In the previous chart, each one of the lines connecting from the yellow dot to the blue dot can represent a signal, weighing the importance of that node in determining the overall score of the output.

  • If that signal is high, that node is significant to the model’s overall performance.
  • If that signal is low, the node is insignificant.

With this understanding, we can define explainability as:

Knowledge of what one node represents and how important it is to the model’s performance.

So how does choice of these two different algorithms make a difference with respect to health care and medical decision making?

The authors argue: 

“Regulators like the FDA should focus on those aspects of the AI/ML system that directly bear on its safety and effectiveness – in particular, how does it perform in the hands of its intended users?”

A suggestion for

  • Enhanced more involved clinical trials
  • Provide individuals added flexibility when interacting with a model, for example inputting their own test data
  • More interaction between user and model generators
  • Determining in which situations call for interpretable AI versus explainable (for instance predicting which patients will require dialysis after kidney damage)

Other articles on AI/ML in medicine and healthcare on this Open Access Journal include

Applying AI to Improve Interpretation of Medical Imaging

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

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

Cardiac MRI Imaging Breakthrough: The First AI-assisted Cardiac MRI Scan Solution, HeartVista Receives FDA 510(k) Clearance for One Click™ Cardiac MRI Package

 

Read Full Post »

Machine Learning Implementation of a Prediction Model for Heart Failure Using Flask and Heroku

Reporter: Aviva Lev-Ari, PhD, RN

Deploying a Heart Failure Prediction Model Using Flask and Heroku

Guest Author: Osasona Ifeoluwa

She Code Africa Cohort 3 Final project.

We published this article as an Educational Example for:

1. Designing a Prediction Model in Cardiology by using Data Created by other Authors

2. Using a Machine Learning Implementation for computation of the prediction values

3. Development of a Web Application to rest in the Public Domain

4. Usage of the Github repository 

5. to be added by Adina Hazan, PhD

6. to be added by Adina Hazan, PhD

7. to be added by Buddhadeb Pradhan, PhD

8. to be added by Buddhadeb Pradhan, PhD

 

Cardiovascular diseases (which often leads to heart failures) are the number 1 cause of death globally, taking an estimated 17.9 million lives each year, which accounts for 31% of global deaths.

Most cardiovascular diseases can be prevented by addressing behavioral risk factors such as tobacco use, unhealthy diet and obesity, physical inactivity, and harmful use of alcohol using population-wide strategies. However, people with cardiovascular disease or who are at high cardiovascular risk (due to the presence of one or more risk factors such as hypertension, diabetes, hyperlipidemia, or already established disease) need early detection and management wherein a machine learning model can be of great help.

This machine learning model could help in predicting mortality caused by heart failure by taking in important features from the dataset and making predictions based on these features.

The dataset consists of 12 variables/features, and 1 output variable/target variable. Let us examine the role of each feature in determining if a person is likely to have heart failure or not:

  1. Age: This is the age of the patient
  2. Anemia: is the decrease in red blood cells or hemoglobin
  3. Creatinine_phosphokinase: is the level of creatine kinase in the blood. This enzyme is important for muscle function.
  4. Diabetes: is a chronic disease that causes high blood sugar
  5. Ejection fraction: is the percentage of blood leaving the heart at each contraction
  6. High blood pressure: is blood pressure that is higher than normal
  7. Platelets: are tiny blood cells that help your body form clots to stop bleeding
  8. Serum creatinine: is the level of serum creatinine in the blood
  9. Serum sodium: is the level of serum sodium in the blood
  10. Sex: gender of the patient
  11. Time: This captures the time of the event
  12. Death event: which is the predictor variable.

Now that we know the function of each feature, Let’s get started

Step 1: Import Libraries

Step 2: Import the Dataset

The Dataset used in building this model was downloaded as a CSV file to my PC from Kaggle.

Step 3: Data Cleaning and EDA

This data was pretty much clean, so I didn’t have to do any more cleaning. However, some important pieces of information can still be explored.

Next, I use Matplotlib to visualize the distribution of the target variable (Death_event)

To check for the relationship between all the features and the target variable, I use a heatmap, which gives a graphical representation of the relationship between the variables.

Note: More analysis was done before Feature Selection, and details can be found on the Jupyter Notebook uploaded to Github.

Step 4: Splitting the Train and Test Data

Step 5: Data Preprocessing

This brings the data to a state that the model can parse easily. For the purpose of this project, the Standard Scaler is used, which standardizes the features by subtracting the mean and then scaling to unit variance.

Step 6: Model Selection

The support vector machine (SVM), a supervised machine learning model that uses classification algorithms for two-group classification problems is used. After giving the SVM model sets of the preprocessed training data for each category, they’re able to categorize new output.

The classification report shows an accuracy of 81%.

Since this model will be deployed, it is saved into a pickle file (model.pkl) created by pickle, and this file will reflect in your project folder.

Pickle is a python module that enables python objects to be written to files on the disk and read back into the python program runtime.

Step 7: Deploying with Flask and Heroku

Deploying a machine learning model means making the model available for end-users to make use of.

Create the Webpage

Here we will create a CSS webpage that has text boxes to take in input from users. The CSS file was named index.html and can be found here.

Several templates for creating a CSS webpage can be found online.

Deploy the model on the webpage using Flask

In deploying this heart failure prediction model into production, a web application framework called Flask is used. Flask makes it easy to write applications, and also gives a variety of choices for developing web applications.

To make use of this web application framework in deploying this model, we install Flask by running the following command:

Next, a Flask environment with an API endpoint that takes in the model and enables it to receive input from users, and return output is setup.

After this, a python file app.py is created, and the required libraries imported

Create the Flask App

Load the pickle

Create an app route to render the HTML template as the home page

Create an API that gets input from the user and computes a predicted value based on the model.

Now, call the run function to start the Flask server.

This should return an output that shows that your app is running. Simply copy the URL and paste it into your browser to test the app.

Deploy the Flask APP to Heroku

Heroku is a multi-language application platform that allows developers to deploy, and manage their applications. It is flexible and easy to use, offering developers the simplest path to getting their apps to market.

The first thing to do in deploying the Flask app to Heroku is to Sign up and Log In to Heroku. After which you can create a Procfile and requirement.txt file, which handles the configuration part in order to deploy the model into the Heroku server.

web: gunicorn is the fixed command for the Procfile.

The requirements file consists of all the libraries that have to get installed in the Heroku environment.

Next, you commit your code to Github and connect Github to Heroku.

After you connect, there are 2 ways to deploy your app. You could either choose automatic deploy or manual deploy. The automatic deployment will take place whenever you commit anything into your Github repository.

By selecting the branch and clicking on deploy, build starts.

After a successful deployment, the app will be created. Click on the view and your app should open. A new URL will also be created and can be shared by users.

Check Out my app via ‘https://heart-failure-prediction-app20.herokuapp.com/

The link to the Github Repository can be found here

Dataset Authors: Davide Chicco, Giuseppe Jurman

Link to Dataset

This was my first machine learning Deployment project, and I hope someone finds this useful🙂.

SOURCE

https://osasonaifeoluwa.medium.com/deploying-a-heart-failure-prediction-model-using-flask-and-heroku-55fdf51ee18e

Read Full Post »

Live Notes, Real Time Conference Coverage AACR 2020 #AACR20: Tuesday June 23, 2020 Noon-2:45 Educational Sessions

Live Notes, Real Time Conference Coverage AACR 2020: Tuesday June 23, 2020 Noon-2:45 Educational Sessions

Reporter: Stephen J. Williams, PhD

Follow Live in Real Time using

#AACR20

@pharma_BI

@AACR

Register for FREE at https://www.aacr.org/

 

Presidential Address

Elaine R Mardis, William N Hait

DETAILS

Welcome and introduction

William N Hait

 

Improving diagnostic yield in pediatric cancer precision medicine

Elaine R Mardis
  • Advent of genomics have revolutionized how we diagnose and treat lung cancer
  • We are currently needing to understand the driver mutations and variants where we can personalize therapy
  • PD-L1 and other checkpoint therapy have not really been used in pediatric cancers even though CAR-T have been successful
  • The incidence rates and mortality rates of pediatric cancers are rising
  • Large scale study of over 700 pediatric cancers show cancers driven by epigenetic drivers or fusion proteins. Need for transcriptomics.  Also study demonstrated that we have underestimated germ line mutations and hereditary factors.
  • They put together a database to nominate patients on their IGM Cancer protocol. Involves genetic counseling and obtaining germ line samples to determine hereditary factors.  RNA and protein are evaluated as well as exome sequencing. RNASeq and Archer Dx test to identify driver fusions
  • PECAN curated database from St. Jude used to determine driver mutations. They use multiple databases and overlap within these databases and knowledge base to determine or weed out false positives
  • They have used these studies to understand the immune infiltrate into recurrent cancers (CytoCure)
  • They found 40 germline cancer predisposition genes, 47 driver somatic fusion proteins, 81 potential actionable targets, 106 CNV, 196 meaningful somatic driver mutations

 

 

Tuesday, June 23

12:00 PM – 12:30 PM EDT

Awards and Lectures

NCI Director’s Address

Norman E Sharpless, Elaine R Mardis

DETAILS

Introduction: Elaine Mardis

 

NCI Director Address: Norman E Sharpless
  • They are functioning well at NCI with respect to grant reviews, research, and general functions in spite of the COVID pandemic and the massive demonstrations on also focusing on the disparities which occur in cancer research field and cancer care
  • There are ongoing efforts at NCI to make a positive difference in racial injustice, diversity in the cancer workforce, and for patients as well
  • Need a diverse workforce across the cancer research and care spectrum
  • Data show that areas where the clinicians are successful in putting African Americans on clinical trials are areas (geographic and site specific) where health disparities are narrowing
  • Grants through NCI new SeroNet for COVID-19 serologic testing funded by two RFAs through NIAD (RFA-CA-30-038 and RFA-CA-20-039) and will close on July 22, 2020

 

Tuesday, June 23

12:45 PM – 1:46 PM EDT

Virtual Educational Session

Immunology, Tumor Biology, Experimental and Molecular Therapeutics, Molecular and Cellular Biology/Genetics

Tumor Immunology and Immunotherapy for Nonimmunologists: Innovation and Discovery in Immune-Oncology

This educational session will update cancer researchers and clinicians about the latest developments in the detailed understanding of the types and roles of immune cells in tumors. It will summarize current knowledge about the types of T cells, natural killer cells, B cells, and myeloid cells in tumors and discuss current knowledge about the roles these cells play in the antitumor immune response. The session will feature some of the most promising up-and-coming cancer immunologists who will inform about their latest strategies to harness the immune system to promote more effective therapies.

Judith A Varner, Yuliya Pylayeva-Gupta

 

Introduction

Judith A Varner
New techniques reveal critical roles of myeloid cells in tumor development and progression
  • Different type of cells are becoming targets for immune checkpoint like myeloid cells
  • In T cell excluded or desert tumors T cells are held at periphery so myeloid cells can infiltrate though so macrophages might be effective in these immune t cell naïve tumors, macrophages are most abundant types of immune cells in tumors
  • CXCLs are potential targets
  • PI3K delta inhibitors,
  • Reduce the infiltrate of myeloid tumor suppressor cells like macrophages
  • When should we give myeloid or T cell therapy is the issue
Judith A Varner
Novel strategies to harness T-cell biology for cancer therapy
Positive and negative roles of B cells in cancer
Yuliya Pylayeva-Gupta
New approaches in cancer immunotherapy: Programming bacteria to induce systemic antitumor immunity

 

 

Tuesday, June 23

12:45 PM – 1:46 PM EDT

Virtual Educational Session

Cancer Chemistry

Chemistry to the Clinic: Part 2: Irreversible Inhibitors as Potential Anticancer Agents

There are numerous examples of highly successful covalent drugs such as aspirin and penicillin that have been in use for a long period of time. Despite historical success, there was a period of reluctance among many to purse covalent drugs based on concerns about toxicity. With advances in understanding features of a well-designed covalent drug, new techniques to discover and characterize covalent inhibitors, and clinical success of new covalent cancer drugs in recent years, there is renewed interest in covalent compounds. This session will provide a broad look at covalent probe compounds and drug development, including a historical perspective, examination of warheads and electrophilic amino acids, the role of chemoproteomics, and case studies.

Benjamin F Cravatt, Richard A. Ward, Sara J Buhrlage

 

Discovering and optimizing covalent small-molecule ligands by chemical proteomics

Benjamin F Cravatt
  • Multiple approaches are being investigated to find new covalent inhibitors such as: 1) cysteine reactivity mapping, 2) mapping cysteine ligandability, 3) and functional screening in phenotypic assays for electrophilic compounds
  • Using fluorescent activity probes in proteomic screens; have broad useability in the proteome but can be specific
  • They screened quiescent versus stimulated T cells to determine reactive cysteines in a phenotypic screen and analyzed by MS proteomics (cysteine reactivity profiling); can quantitate 15000 to 20,000 reactive cysteines
  • Isocitrate dehydrogenase 1 and adapter protein LCP-1 are two examples of changes in reactive cysteines they have seen using this method
  • They use scout molecules to target ligands or proteins with reactive cysteines
  • For phenotypic screens they first use a cytotoxic assay to screen out toxic compounds which just kill cells without causing T cell activation (like IL10 secretion)
  • INTERESTINGLY coupling these MS reactive cysteine screens with phenotypic screens you can find NONCANONICAL mechanisms of many of these target proteins (many of the compounds found targets which were not predicted or known)

Electrophilic warheads and nucleophilic amino acids: A chemical and computational perspective on covalent modifier

The covalent targeting of cysteine residues in drug discovery and its application to the discovery of Osimertinib

Richard A. Ward
  • Cysteine activation: thiolate form of cysteine is a strong nucleophile
  • Thiolate form preferred in polar environment
  • Activation can be assisted by neighboring residues; pKA will have an effect on deprotonation
  • pKas of cysteine vary in EGFR
  • cysteine that are too reactive give toxicity while not reactive enough are ineffective

 

Accelerating drug discovery with lysine-targeted covalent probes

 

Tuesday, June 23

12:45 PM – 2:15 PM EDT

Virtual Educational Session

Molecular and Cellular Biology/Genetics

Virtual Educational Session

Tumor Biology, Immunology

Metabolism and Tumor Microenvironment

This Educational Session aims to guide discussion on the heterogeneous cells and metabolism in the tumor microenvironment. It is now clear that the diversity of cells in tumors each require distinct metabolic programs to survive and proliferate. Tumors, however, are genetically programmed for high rates of metabolism and can present a metabolically hostile environment in which nutrient competition and hypoxia can limit antitumor immunity.

Jeffrey C Rathmell, Lydia Lynch, Mara H Sherman, Greg M Delgoffe

 

T-cell metabolism and metabolic reprogramming antitumor immunity

Jeffrey C Rathmell

Introduction

Jeffrey C Rathmell

Metabolic functions of cancer-associated fibroblasts

Mara H Sherman

Tumor microenvironment metabolism and its effects on antitumor immunity and immunotherapeutic response

Greg M Delgoffe
  • Multiple metabolites, reactive oxygen species within the tumor microenvironment; is there heterogeneity within the TME metabolome which can predict their ability to be immunosensitive
  • Took melanoma cells and looked at metabolism using Seahorse (glycolysis): and there was vast heterogeneity in melanoma tumor cells; some just do oxphos and no glycolytic metabolism (inverse Warburg)
  • As they profiled whole tumors they could separate out the metabolism of each cell type within the tumor and could look at T cells versus stromal CAFs or tumor cells and characterized cells as indolent or metabolic
  • T cells from hyerglycolytic tumors were fine but from high glycolysis the T cells were more indolent
  • When knock down glucose transporter the cells become more glycolytic
  • If patient had high oxidative metabolism had low PDL1 sensitivity
  • Showed this result in head and neck cancer as well
  • Metformin a complex 1 inhibitor which is not as toxic as most mito oxphos inhibitors the T cells have less hypoxia and can remodel the TME and stimulate the immune response
  • Metformin now in clinical trials
  • T cells though seem metabolically restricted; T cells that infiltrate tumors are low mitochondrial phosph cells
  • T cells from tumors have defective mitochondria or little respiratory capacity
  • They have some preliminary findings that metabolic inhibitors may help with CAR-T therapy

Obesity, lipids and suppression of anti-tumor immunity

Lydia Lynch
  • Hypothesis: obesity causes issues with anti tumor immunity
  • Less NK cells in obese people; also produce less IFN gamma
  • RNASeq on NOD mice; granzymes and perforins at top of list of obese downregulated
  • Upregulated genes that were upregulated involved in lipid metabolism
  • All were PPAR target genes
  • NK cells from obese patients takes up palmitate and this reduces their glycolysis but OXPHOS also reduced; they think increased FFA basically overloads mitochondria
  • PPAR alpha gamma activation mimics obesity

 

 

Tuesday, June 23

12:45 PM – 2:45 PM EDT

Virtual Educational Session

Clinical Research Excluding Trials

The Evolving Role of the Pathologist in Cancer Research

Long recognized for their role in cancer diagnosis and prognostication, pathologists are beginning to leverage a variety of digital imaging technologies and computational tools to improve both clinical practice and cancer research. Remarkably, the emergence of artificial intelligence (AI) and machine learning algorithms for analyzing pathology specimens is poised to not only augment the resolution and accuracy of clinical diagnosis, but also fundamentally transform the role of the pathologist in cancer science and precision oncology. This session will discuss what pathologists are currently able to achieve with these new technologies, present their challenges and barriers, and overview their future possibilities in cancer diagnosis and research. The session will also include discussions of what is practical and doable in the clinic for diagnostic and clinical oncology in comparison to technologies and approaches primarily utilized to accelerate cancer research.

 

Jorge S Reis-Filho, Thomas J Fuchs, David L Rimm, Jayanta Debnath

DETAILS

Tuesday, June 23

12:45 PM – 2:45 PM EDT

 

High-dimensional imaging technologies in cancer research

David L Rimm

  • Using old methods and new methods; so cell counting you use to find the cells then phenotype; with quantification like with Aqua use densitometry of positive signal to determine a threshold to determine presence of a cell for counting
  • Hiplex versus multiplex imaging where you have ten channels to measure by cycling of flour on antibody (can get up to 20plex)
  • Hiplex can be coupled with Mass spectrometry (Imaging Mass spectrometry, based on heavy metal tags on mAbs)
  • However it will still take a trained pathologist to define regions of interest or field of desired view

 

Introduction

Jayanta Debnath

Challenges and barriers of implementing AI tools for cancer diagnostics

Jorge S Reis-Filho

Implementing robust digital pathology workflows into clinical practice and cancer research

Jayanta Debnath

Invited Speaker

Thomas J Fuchs
  • Founder of spinout of Memorial Sloan Kettering
  • Separates AI from computational algothimic
  • Dealing with not just machines but integrating human intelligence
  • Making decision for the patients must involve human decision making as well
  • How do we get experts to do these decisions faster
  • AI in pathology: what is difficult? =è sandbox scenarios where machines are great,; curated datasets; human decision support systems or maps; or try to predict nature
  • 1) learn rules made by humans; human to human scenario 2)constrained nature 3)unconstrained nature like images and or behavior 4) predict nature response to nature response to itself
  • In sandbox scenario the rules are set in stone and machines are great like chess playing
  • In second scenario can train computer to predict what a human would predict
  • So third scenario is like driving cars
  • System on constrained nature or constrained dataset will take a long time for commuter to get to decision
  • Fourth category is long term data collection project
  • He is finding it is still finding it is still is difficult to predict nature so going from clinical finding to prognosis still does not have good predictability with AI alone; need for human involvement
  • End to end partnering (EPL) is a new way where humans can get more involved with the algorithm and assist with the problem of constrained data
  • An example of a workflow for pathology would be as follows from Campanella et al 2019 Nature Medicine: obtain digital images (they digitized a million slides), train a massive data set with highthroughput computing (needed a lot of time and big software developing effort), and then train it using input be the best expert pathologists (nature to human and unconstrained because no data curation done)
  • Led to first clinically grade machine learning system (Camelyon16 was the challenge for detecting metastatic cells in lymph tissue; tested on 12,000 patients from 45 countries)
  • The first big hurdle was moving from manually annotated slides (which was a big bottleneck) to automatically extracted data from path reports).
  • Now problem is in prediction: How can we bridge the gap from predicting humans to predicting nature?
  • With an AI system pathologist drastically improved the ability to detect very small lesions

 

Virtual Educational Session

Epidemiology

Cancer Increases in Younger Populations: Where Are They Coming from?

Incidence rates of several cancers (e.g., colorectal, pancreatic, and breast cancers) are rising in younger populations, which contrasts with either declining or more slowly rising incidence in older populations. Early-onset cancers are also more aggressive and have different tumor characteristics than those in older populations. Evidence on risk factors and contributors to early-onset cancers is emerging. In this Educational Session, the trends and burden, potential causes, risk factors, and tumor characteristics of early-onset cancers will be covered. Presenters will focus on colorectal and breast cancer, which are among the most common causes of cancer deaths in younger people. Potential mechanisms of early-onset cancers and racial/ethnic differences will also be discussed.

Stacey A. Fedewa, Xavier Llor, Pepper Jo Schedin, Yin Cao

Cancers that are and are not increasing in younger populations

Stacey A. Fedewa

 

  • Early onset cancers, pediatric cancers and colon cancers are increasing in younger adults
  • Younger people are more likely to be uninsured and these are there most productive years so it is a horrible life event for a young adult to be diagnosed with cancer. They will have more financial hardship and most (70%) of the young adults with cancer have had financial difficulties.  It is very hard for women as they are on their childbearing years so additional stress
  • Types of early onset cancer varies by age as well as geographic locations. For example in 20s thyroid cancer is more common but in 30s it is breast cancer.  Colorectal and testicular most common in US.
  • SCC is decreasing by adenocarcinoma of the cervix is increasing in women’s 40s, potentially due to changing sexual behaviors
  • Breast cancer is increasing in younger women: maybe etiologic distinct like triple negative and larger racial disparities in younger African American women
  • Increased obesity among younger people is becoming a factor in this increasing incidence of early onset cancers

 

 

Other Articles on this Open Access  Online Journal on Cancer Conferences and Conference Coverage in Real Time Include

Press Coverage

Live Notes, Real Time Conference Coverage 2020 AACR Virtual Meeting April 28, 2020 Symposium: New Drugs on the Horizon Part 3 12:30-1:25 PM

Live Notes, Real Time Conference Coverage 2020 AACR Virtual Meeting April 28, 2020 Session on NCI Activities: COVID-19 and Cancer Research 5:20 PM

Live Notes, Real Time Conference Coverage 2020 AACR Virtual Meeting April 28, 2020 Session on Evaluating Cancer Genomics from Normal Tissues Through Metastatic Disease 3:50 PM

Live Notes, Real Time Conference Coverage 2020 AACR Virtual Meeting April 28, 2020 Session on Novel Targets and Therapies 2:35 PM

 

Read Full Post »

2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

Dialogue among principals is a World Forum’s signature. Expert moderators guiding discussion and questions in audience friendly exchanges. No slides – shared perspectives facilitated by Harvard faculty, leading journalists and Mass General Brigham executives.

Jeffrey Golden, MD

Chair, Department of Pathology, BH; Ramzi S. Cotran Professor of Pathology, Harvard Medical School

Hadine Joffe, MD

Vice Chair, Psychiatry, Executive Director, Mary Horrigan Connors Center for Women’s Health and Gender Biology, BH; Paula A. Johnson Professor, Women’s Health, Harvard Medical School

Thomas Sequist, MD

Chief Patient Experience and Equity Officer, Mass General Brigham; Professor of Medicine and Health Care Policy, Harvard Medical School

Erica Shenoy, MD, PhD

Associate Chief, Infection Control Unit, MGH; Assistant Professor, Harvard Medical School

Gregg Meyer, MD

Chief Clinical Officer, Mass General Brigham; Interim President, NWH; Professor, Harvard Medical School

Ravi Thadhani, MD

CAO, Mass General Brigham; Professor and Faculty Dean for Academic Programs, Harvard Medical School

Ann Prestipino

SVP; Incident Commander, MGH

Roger Kitterman

VP, Venture and Managing Partner, Partners Innovation Fund, Mass General Brigham

David Louis, MD

Pathologist-in-Chief, MGH; Benjamin Castleman Professor of Pathology, Harvard Medical School

Janet Wu

Bloomberg

Ron Walls, MD

EVP and Chief Operating Officer, BH; Neskey Family Professor of Emergency Medicine, Harvard Medical School

Alice Park

Senior Writer, TIME

 

Jeffrey Golden, MD

Chair, Department of Pathology, BH; Ramzi S. Cotran Professor of Pathology, Harvard Medical School

Hadine Joffe, MD

Vice Chair, Psychiatry, Executive Director, Mary Horrigan Connors Center for Women’s Health and Gender Biology, BH; Paula A. Johnson Professor, Women’s Health, Harvard Medical School

Thomas Sequist, MD

Chief Patient Experience and Equity Officer, Mass General Brigham; Professor of Medicine and Health Care Policy, Harvard Medical School

Erica Shenoy, MD, PhD

Associate Chief, Infection Control Unit, MGH; Assistant Professor, Harvard Medical School

Gregg Meyer, MD

Chief Clinical Officer, Mass General Brigham; Interim President, NWH; Professor, Harvard Medical School

Ravi Thadhani, MD

CAO, Mass General Brigham; Professor and Faculty Dean for Academic Programs, Harvard Medical School

Ann Prestipino

SVP; Incident Commander, MGH

Roger Kitterman

VP, Venture and Managing Partner, Partners Innovation Fund, Mass General Brigham

David Louis, MD

Pathologist-in-Chief, MGH; Benjamin Castleman Professor of Pathology, Harvard Medical School

Janet Wu

Bloomberg

Ron Walls, MD

EVP and Chief Operating Officer, BH; Neskey Family Professor of Emergency Medicine, Harvard Medical School

Alice Park

Senior Writer, TIME

 

VIEW VIDEOS from the event

https://www.youtube.com/channel/UCauKpbsS_hUqQaPp8EVGYOg

 

From: “Coburn, Christopher Mark” <CMCOBURN@PARTNERS.ORG>

Date: Tuesday, May 12, 2020 at 6:48 AM

To: “Coburn, Christopher Mark” <CMCOBURN@PARTNERS.ORG>

Subject: REGISTRANT RECAP | World Medical Innovation Forum  

 

Dear World Forum Attendee, 

On behalf of Mass General Brigham CEO Anne Klibanski MD and Forum co-Chairs Gregg Meyer MD and Ravi Thadhani MD, many thanks for being among the nearly 11,000 registrants representing 93 countries, 46 states and 3200 organizations yesterday. A community was established around many pressing topics that  will continue long into the future. We hope you have a chance to examine the attached survey results. There are several revealing items that should be the basis for ongoing discussion. We expect to be in touch regularly during the year. Among the plans is a “First Look” video series highlighting top Mass General Brigham Harvard faculty as well as emerging Harvard investigators.  As promised, we  wanted to also share visual Forum session summaries.  You will be able to access the recordings on the Forum’s YouTube page . The first set will go up this morning

We hope you will join us for the 2021 Forum!  

Thanks again, Chris

e-Proceedings 2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

https://pharmaceuticalintelligence.com/2020/04/22/world-medical-innovation-forum-covid-19-ai-and-the-future-of-medicine-featuring-harvard-and-industry-leader-insights-mgh-bwh-virtual-event-monday-may-11-815-a-m-515-p-m-et/

Tweets & Retweets 2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

https://pharmaceuticalintelligence.com/2020/05/11/tweets-retweets-2020-world-medical-innovation-forum-covid-19-ai-and-the-future-of-medicine-featuring-harvard-and-industry-leader-insights-mgh-bwh-virtual-event-mond/

Read Full Post »

Collaborative innovation has never been more important.

2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

Join top leaders guiding the response, technology and people confronting this century’s greatest health challenge.

Priya Abani

CEO, AliveCor

General Keith Alexander

Co-CEO, IronNet; Former NSA Head

Stéphane Bancel

CEO, Moderna

Marc Casper

CEO, Thermo Fisher

Timothy Ferris, MD

CEO, MGPO; Professor, HMS

John Fernandez  

President, MEE; President, Ambulatory Care, Mass General Brigham

 

John Fish

CEO, Suffolk; BH Board Chair

JF Formela, MD

Partner, Atlas Venture

Jan Garfinkle

Manager Partner, Arboretum Ventures; Chair, NVCA

Phillip Gross

Managing Director, Adage Capital Management

Julia Hu

CEO, Lark Health

Anjali Kataria

CEO, Mytonomy

Roger Kitterman

VP, Managing Partner, Mass General Brigahm Fund

Jonathan Kraft

President, Kraft Group; Chair, MGH Board

Brooke LeVasseur

CEO, AristaMD

Mike Mahoney

CEO, Boston Scientific

Bernd Montag, PhD

CEO, Siemens Healthineers

Kieran Murphy

CEO, GE Healthcare

Elizabeth Nabel, MD

President, BH; Professor, HMS

Matt Sause

CEO, Roche Diagnostics

Peter Slavin, MD

President, MGH; Professor, HMS

Scott Sperling

Co-President, TH Lee; Chair, Mass General Brigham Board

Christopher Viehbacher

Managing Partner, Gurnet Point Capital

Michel Vounatsos

CEO, Biogen

Collaborative Innovation

Together we meet the challenge of the coronavirus and share our commitment to the future of medicine.

 

Anne Klibanski, MD

CEO, Mass General Brigham

Amy Abernethy, MD, PhD

Principal Deputy Commissioner and Acting CIO, FDA

PANEL

FDA Role in Managing Crisis and Anticipating the Next

Elizabeth Nabel, MD

President, Brigham Health; Professor of Medicine, HMS

PANEL

Care in the Next 18 Months 

Karen DeSalvo, MD

Chief Health Officer, Google Health

PANEL

Role of AI and Big Data in Fighting COVID-19 

Dawn Sugarman, PhD

Assistant Psychologist, Division of Alcohol, Drugs, and Addiction, McLean; Assistant Professor, Psychiatry, HMS

PANEL

Digital Therapeutics

Ann Prestipino

SVP; Incident Commander, MGH; Teaching Associate, HMS

PANEL

Real Time: Front Line Innovation

Hadine Joffe, MD

Vice Chair, Research, Psychiatry; Executive Director, Mary Horrigan Connors Center for Women’s Health and Gender Biology, BH; Paula Johnson Professor, Women’s Health, HMS

PANEL

Digital Therapeutics

Priya Abani

CEO, AliveCor

PANEL

Digital Therapeutics

Julia Hu

CEO, Lark Health

PANEL

Digital Therapeutics

Jan Garfinkle

Manager Partner, Arboretum Ventures; Chair NVCA

PANEL

Early Stage Investment Environment

Anjali Kataria

CEO, Mytonomy

PANEL

Patient Experience During the Pandemic

Brooke LeVasseur

CEO, AristaMD

PANEL

Digital Health Becomes a Pillar

Julie Lankiewicz

Head, Clinical Affairs & Health Economics Outcomes Research, Bose Health

PANEL

Emergency and Urgent Care

 

VIEW VIDEOS from the event

https://www.youtube.com/channel/UCauKpbsS_hUqQaPp8EVGYOg

 

From: “Coburn, Christopher Mark” <CMCOBURN@PARTNERS.ORG>

Date: Tuesday, May 12, 2020 at 6:48 AM

To: “Coburn, Christopher Mark” <CMCOBURN@PARTNERS.ORG>

Subject: REGISTRANT RECAP | World Medical Innovation Forum  

 

Dear World Forum Attendee, 

On behalf of Mass General Brigham CEO Anne Klibanski MD and Forum co-Chairs Gregg Meyer MD and Ravi Thadhani MD, many thanks for being among the nearly 11,000 registrants representing 93 countries, 46 states and 3200 organizations yesterday. A community was established around many pressing topics that  will continue long into the future. We hope you have a chance to examine the attached survey results. There are several revealing items that should be the basis for ongoing discussion. We expect to be in touch regularly during the year. Among the plans is a “First Look” video series highlighting top Mass General Brigham Harvard faculty as well as emerging Harvard investigators.  As promised, we  wanted to also share visual Forum session summaries.  You will be able to access the recordings on the Forum’s YouTube page . The first set will go up this morning

We hope you will join us for the 2021 Forum!  

Thanks again, Chris

e-Proceedings 2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

https://pharmaceuticalintelligence.com/2020/04/22/world-medical-innovation-forum-covid-19-ai-and-the-future-of-medicine-featuring-harvard-and-industry-leader-insights-mgh-bwh-virtual-event-monday-may-11-815-a-m-515-p-m-et/

Tweets & Retweets 2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

https://pharmaceuticalintelligence.com/2020/05/11/tweets-retweets-2020-world-medical-innovation-forum-covid-19-ai-and-the-future-of-medicine-featuring-harvard-and-industry-leader-insights-mgh-bwh-virtual-event-mond/

Read Full Post »

2020 World Medical Innovation Forum – COVID-19, AI  – Life Science and Digital Health Investments, MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

 

 

 

Life science and digital health investments have continued at a strong pace during the COVID-19 crisis. Senior investment leaders discuss what to expect. Will:

  • social distancing affect deal making?
  • key asset categories remain strong – venture, private equity, public offerings, acquisitions?
  • valuations hold up in some categories while others fall?

Moderator: Roger Kitterman, VP, Venture and Managing Partner, Partners Innovation Fund, Mass General Brigham


Jan Garfinkle
, Founder & Manager Partner, Arboretum Ventures, Chair NVCA

Phillip Gross, Managing Director, Adage Capital Management

Christopher Viehbacher, Managing Partner, Gurnet Point Capital

 

VIEW VIDEOS from the event

https://www.youtube.com/channel/UCauKpbsS_hUqQaPp8EVGYOg

From: “Coburn, Christopher Mark” <CMCOBURN@PARTNERS.ORG>

Date: Tuesday, May 12, 2020 at 6:48 AM

To: “Coburn, Christopher Mark” <CMCOBURN@PARTNERS.ORG>

Subject: REGISTRANT RECAP | World Medical Innovation Forum  

 

Dear World Forum Attendee, 

On behalf of Mass General Brigham CEO Anne Klibanski MD and Forum co-Chairs Gregg Meyer MD and Ravi Thadhani MD, many thanks for being among the nearly 11,000 registrants representing 93 countries, 46 states and 3200 organizations yesterday. A community was established around many pressing topics that  will continue long into the future. We hope you have a chance to examine the attached survey results. There are several revealing items that should be the basis for ongoing discussion. We expect to be in touch regularly during the year. Among the plans is a “First Look” video series highlighting top Mass General Brigham Harvard faculty as well as emerging Harvard investigators.  As promised, we  wanted to also share visual Forum session summaries.  You will be able to access the recordings on the Forum’s YouTube page . The first set will go up this morning

We hope you will join us for the 2021 Forum!  

Thanks again, Chris

e-Proceedings 2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

https://pharmaceuticalintelligence.com/2020/04/22/world-medical-innovation-forum-covid-19-ai-and-the-future-of-medicine-featuring-harvard-and-industry-leader-insights-mgh-bwh-virtual-event-monday-may-11-815-a-m-515-p-m-et/

Tweets & Retweets 2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

https://pharmaceuticalintelligence.com/2020/05/11/tweets-retweets-2020-world-medical-innovation-forum-covid-19-ai-and-the-future-of-medicine-featuring-harvard-and-industry-leader-insights-mgh-bwh-virtual-event-mond/

Read Full Post »

Tweets & Retweets 2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

From: “Partners Innovation (via Twitter)” <notify@twitter.com>

Date: Tuesday, May 12, 2020 at 2:24 PM

To: Aviva Lev-Ari <AvivaLev-Ari@alum.berkeley.edu>

Subject: Partners Innovation (@PHSInnovation) has sent you a Direct Message on Twitter!

 

Thanks for tweeting about the live event Aviva! We appreciate the support!

 

e-Proceedings 2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

https://pharmaceuticalintelligence.com/2020/04/22/world-medical-innovation-forum-covid-19-ai-and-the-future-of-medicine-featuring-harvard-and-industry-leader-insights-mgh-bwh-virtual-event-monday-may-11-815-a-m-515-p-m-et/

VIEW ALL VIDEOS

https://www.youtube.com/channel/UCauKpbsS_hUqQaPp8EVGYOg

 

Aviva Lev-Ari
@AVIVA1950

#WMIF2020

Michel Vounatsos, CEO, Biogen Venture community supportive to be on the safe side  employees tested every evenings to prevent rebound of the pandemic Pandemic is acceleration progress technologies new drugs Biogen will lead new model

Notifications

#WMIF2020 @PHSInnovation @pharma_BI @AVIVA1950 Digital Therapeutics Hadine Joffe, MD @BH; Paula A. Johnson Professor, Women’s Health, HMS Priya Abani, CEO, AliveCor Julia Hu, CEO, Lark Health Dawn Sugarman, PhD @McLeanHospital

liked your Tweet

#WMIF2020 @PHSInnovation @pharma_BI @AVIVA1950 Joerg Moeller, MD, PhD, Head of Research @BayerPharmaAG led team of 9 products Unprecedented is COVID-19: effect on work, travel, lifevAnti-Malaria vs COVID-19: In China testing early chloroquine approved for RA and anti Malaria

Retweeted 78 of your Tweets

#WMIF2020 @PHSInnovation @pharma_BI @AVIVA1950 Michael Mina, MD, PhD @BH Antigen test for home administration consumerization of the Testing  Walmart can be positioned for blood tests Not only Physicians can order tests @Microsoft @Amazon can interpretation of Test using Alexa

Show all

liked 85 of your Tweets

#WMIF2020 @PHSInnovation @pharma_BI @AVIVA1950 Michael Mina, MD, PhD @BH Antigen test for home administration consumerization of the Testing  Walmart can be positioned for blood tests Not only Physicians can order tests @Microsoft @Amazon can interpretation of Test using Alexa

Show all

Stephen J Williams
@StephenJWillia2

Quote Tweet
Aviva Lev-Ari
@AVIVA1950
·
#WMIF2020 @PHSInnovation @pharma_BI @AVIVA1950 Ross Zafonte, DO, SVP, Research Education and Medical Affairs, SRN; Earle P. and Ida S. Charlton Professor of Physical Medicine and Rehabilitation, HMS @MGH is family, the unattainable is attainable

Stephen J Williams
@StephenJWillia2

#WMIF2020 #Telemedicine so important for #COVID19 pandemic. Platforms developed years ago. Who would have known?

Quote Tweet
Aviva Lev-Ari
@AVIVA1950
·
#WMIF2020 @PHSInnovation @pharma_BI @AVIVA1950 Jan Garfinkle, Founder & Manager Partner, Arboretum Ventures Can you close a deal with out meeting management team Known funds will prevail vs new funds Parma adjacencies vs medical devices Telehealth is of interest GI Cardiovascular

Stephen J Williams
@StephenJWillia2

Quote Tweet
Aviva Lev-Ari
@AVIVA1950
·
#WMIF2020 @PHSInnovation @pharma_BI @AVIVA1950 Ravi Thadhani, MD, CAO, Mass General Brigham; Professor of Medicine and Faculty Dean for Academic Programs, HMS Great Broadcasting services, expertise on the top Management of the Event 100% no room to improve Recovery COVID Patients

Stephen J Williams
@StephenJWillia2

2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – #MGH & #BWH Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET #WMIF

#bostonbio pharmaceuticalintelligence.com/2020/04/22/wor via

2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard…
2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. –…
pharmaceuticalintelligence.com
1
1
News for you
Fears of homophobic backlash sparked as Korean media focus on gay club COVID-19 case

Retweeted your Tweet

#WMIF20 @pharma_BI @AVIVA1950 covering event in #realtime +9,500 Global Attendees for lnkd.in/ePwTDxm about worldmedicalinnovation.org/2020-disruptiv 2020 #Virtual #World #Medical #Innovation #Forum#COVID-19 #AI #Future #Medicine @MGH & @BWH, Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

liked your Tweet

#WMIF20 @pharma_BI @AVIVA1950 covering event in #realtime +9,500 Global Attendees for lnkd.in/ePwTDxm about worldmedicalinnovation.org/2020-disruptiv 2020 #Virtual #World #Medical #Innovation #Forum#COVID-19 #AI #Future #Medicine @MGH & @BWH, Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

All Tweets by @AVIVA1950 on May 11, 2020

Read Full Post »

e-Proceedings 2020 World Medical Innovation Forum – COVID-19, AI and the Future of Medicine, Featuring Harvard and Industry Leader Insights – MGH & BWH, Virtual Event: Monday, May 11, 8:15 a.m. – 5:15 p.m. ET

 

Featuring Clinical, Scientific, Tech, AI and Venture Experts

https://worldmedicalinnovation.org/

7:50NOW PLAYING

2020 WMIF | Welcome

34 views1 hour ago

5:31NOW PLAYING

2020 WMIF | Disruptive Dozen #1

122 views1 day ago

3:27NOW PLAYING