Posts Tagged ‘Natural Language Processing (NLP)’

WORKFLOW for a Eight-Steps Medical Text Analysis Operation using NLP on LPBI Medical and Life Sciences Content

Author: Aviva Lev-Ari, PhD, RN with inputs related to Wolfram NLP Code by Madison Davis and Yash Choudhary

  • All INTERNS will work 50% on NLP and 50% on Synthetic Biology
  • Training will be offered
  • Protocol will be developed 
  • Software applications will be selected.
  • Mid September we will have an Internal Meeting on Mission #2, LPBI India before the Meeting with Dr. Nir
  • All Interns need to complete at least ½ an e-Book Eight-Step Workflow Protocol for NLP before starting the Synthetic Biology SW Training
  • This is the Eight-Step WORKFLOW Protocol for Medical Text Analysis using NLP at LPBI:
  • https://pharmaceuticalintelligence.com/2021/07/15/workflow-for-a-ten-steps-medical-text-analysis-operation-using-nlp-on-lpbi-medical-and-life-sciences-content/
  • Each INTERN Completing the 50% assignment on NLP will need to submit this table for his/hers NLP Book Assignment with a Check off mark for each article in each Chapter in the Book the intern was assigned for
  • This Table filled in serves as INPUT for QA of the work of the INTERN. Verification is needed for Internship completion for Certification purposes
NLPStep 1Step 2Step 3Step 4Step 5Step 6Step 7Step 8
Chapter 1, Article 1          
Chapter 1, Article n          
Chapter 2 Article 1          
Chapter 2, Article n          
Chapter 3, Article 1          
Chapter 3, Article n          
Chapter 4 Article 1          
Chapter 4, Article n          
Chapter 5, Article 1          
Chapter 5, Article n          
Chapter n Article 1          
Chapter n Article n          

Table Source :

Author: Aviva Lev-Ari, PhD, RN, 7/23/2021

This Table is the supporting evidence for:

LPBI’s WORKFLOW for Medical Text Analysis Operation using NLP on

LPBI Medical and Life Sciences Content


STEP 1: Create Merged .TXT File of Data for Semantic Analysis

  1. Go to Wolfram Cloud and login (use old or make free account)
  2. Make a new folder: click “New” then “Folder” in the corner of your dashboard
  3. Name your folder. This is where you’ll put your text files
  4. Make a new text file: click “New” then “Text” in the corner of your dashboard
  5. Name your text file.
  6. Put this text file in the folder you made.
  7. Copy and paste the text you want to analyze in the text file. The text can be from one article or multiple articles.
  8. Verify that the text has been saved by reopening the file. Sometimes you may need to repeat the process of copying and pasting text so it works.

STEP 2: Create Presentation Slides to Copy and Paste Data, Image Graphs Into

  1. Google Slides Example: Link
  2. Google Slides Template: Link

STEP 3: View this Document for a Guide on How to Make Semantics Graphs:

  1. Link
  2. Add graphs to presentation slides
  3. The link covers the following algorithms:
    1. 25 KeyWord Extraction
    2. Hypergraphs (make one hypergraph per article collection)
    3. Tree Diagrams (make one tree diagram per article collection)
    4. Bar Diagrams

STEP 4:  Use WordItOut.com and .TXT file per article to generated One WordCloud per article

  1. Go to WorditOut.com, as seen here: https://worditout.com/
  2. Click “Create Your Own”
  3. Copy and Paste the text of the article you want into the text box
  4. Click “Generate”
  5. Change the font to “serif”, disable “random on regenerate”
  6. Change the colors based on the following guidelines, although this is just a suggestion.
  7. Change the number of colors to 2, color blending method “Direct”
  8. Save the image
  9. Upload WordClouds to the Media Gallery and record Article title as Legend and Source for the graph, add your name as image producer and date
  10. Insert World Cloud in the Article following the Author/Curator’s name
  11. Place WordCloud in a one PowerPoint Presentation for the entire Article Collection

STEP 5: Transfer Powerpoint to DropBox

STEP 6: Create a .DOCX Interpretation File of Results

  1. Domain Knowledge Expert generates a .DOCX file with his expert interpretation of all the Insights drawn from the visualization artifacts generated by NLP, ML, AI when all the insights are put together for analysis and synthesis. Store the Expert interpretation into the Interpretation Folder.
  2. What are the clinical implications for patient treatment
  3. What are the clinical insights for drug discovery for Big Pharma?
  4. Are there clues for risk adjustment and policy writing tips for health care insurers?

STEP 7:  Transfer copy of Interpretations files for Translation into Foreign Languages: Spanish, Japanese, Russian into Folders with Language Name

STEP 8: Under Construction: Enrichment of the original content with External Repositories

Read Full Post »

%d bloggers like this: