2-year anniversary of WellAI’s machine learning tool to fight COVID-19

It’s hard to believe it’s been two years since we built our very first machine learning tool – WellAI’s COVID-19 Machine Learning Analytics Tool.

 

To narrow the scope of medical research with the focus on fighting COVID-19, on April 7, 2020, WellAI data scientists released the COVID-19 Machine Learning Tool for Medical Researchers and Physicians available at https://wellai.health/covid

 

We submitted our research related to this tool for publication.  The paper was peer-reviewed and accepted in record terms.  It was published in the IFCC journal on June 2, 2020: https://pubmed.ncbi.nlm.nih.gov/32549878/

 

Two years later, it’s still the only tool that reads and summarizes coronavirus-related peer-reviewed research from all 69 UMLS medical categories. By using synonyms and correlated concepts, WellAI has built a database of 4,224,512 medical concepts, 60,892 of which are used specifically for COVID research.

 

A great number of medical researchers from many fields of medicine, not just infectious disease experts, have been working hard to make sure we understand how COVID-19 and other coronaviruses develop and how they affect the human body, and what the possible side effects and consequences are. This knowledge will help us be better prepared for the next pandemic, should it ever happen.

 

  • Unlike PubMed or Google Scholar, the WellAI NLP search engine is multi-dimensional. It doesn’t just know correlations between medical terms (e.g. genes & diseases). It knows the relationships between every cluster of medical terms in the NIH/Pubmed database.
  • WellAI neural networks summarize, generalize and predict medical relationships.
  • WellAI AI system understands synonyms and correlated concepts. For example, understands that “hypertension” is a synonym for “high blood pressure” and “elevated blood pressure”. This knowledge helps build more accurate relationships between concepts.
  • Every single answer/diagnosis/recommendation/advice is based on a large number of articles.
  • A structured list of concepts with ranked probabilities. This narrows the scope of work and results in greater efficiency. Focus on concepts of interest and exploration of relationships – not only between concepts (e.g., COVID-19 and Diagnostics Radiology) but between clusters of concepts (e.g., COVID-19 + Diagnosis, Clinical + Diagnostic Tests and Diagnostics Radiology)

 

We also offered the software at no cost to primary care practitioners.  While we don’t have precise estimates, the tool may have saved the lives of hundreds, if not thousands, of COVID patients.

 

Since then, the WellAI team, breaking traditional AI boundaries, expanded the baseline tool with additional neural networks enabling the assimilation of over 30 million peer-reviewed medical articles, journals, and case studies, and continue to assimilate new publications as they emerged. This created a multidisciplinary diagnostic solution that addresses practically every known health condition from the common cold to rare diseases. What started as a tool for an academic and research community is now available to everyone, via the WellAI Virtual Triage Assistant and the WellAI for Medical Providers.

 

Thank you for being a loyal WellAI follower.  While our roots are in academia, our mission is to help everyday patients, especially those who are underprivileged, of low income, with no health insurance, or away from a medical office.

 

WellAI Team

wellai.health

Leave a Reply

Your email address will not be published.

%d bloggers like this: