On this day exactly two years ago, amid the COVID pandemic, WellAI scientists, together with world-renowned medical researchers, published the very first survey paper on AI tools and technologies to fight COVID-19: https://pubmed.ncbi.nlm.nih.gov/32549878/
In particular, as we explain in the paper, the WellAI machine-learning analytics tool for medical researchers available openly at https://wellai.health/covid gave a scientific option for medical researchers from many fields of medicine, not just infectious disease experts, to narrow the scope of their research with the focus on fighting COVID-19, or any other health issue. It’s the only tool that reads and summarizes 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.
So what’s so special about the WellAI engine and the consequent research?
WellAI is the only machine learning engine that knows inter-discipline relationships between every one of roughly 5 million medical terms – symptoms, diseases, drugs, medical tests, genes, proteins, vitamins, etc. – not only 1-to-1 relations, but 2-to-1, 2-to-2, and any combination of medical terms of up to 20-to-20. We are talking about hundreds of trillions of medical relationships!
The uniqueness of the WellAI technology is in the way we are connecting the medical terms. While the NIH groups medicine in 69 UMLS categories, we as data scientists don’t see pre-determined categories. For us, there are no boundaries. Two medical studies from different categories could have a higher correlation than two articles from the same category.
Think about the Principal Component Analysis (PCA). It doesn’t see pre-determined categories or definitions. PCA uses eigenvectors on a given dataset to build principal components to satisfy certain criteria. WellAI system applied a similar approach in NLP, using proprietary word embeddings and convolutional networks.
Most researchers focus their research on a very narrow field. If you are a genome scientist, you could spend your whole career researching 3-4 genes or gene mutations. However, WellAI technology helps expand the boundaries of their research and point researchers in the direction they haven’t thought of focusing on, but that could potentially address the health issue in question much more efficiently.
The timeline was incredible and it went something like this:
➡️ On March 16, 2020, the White House announced a “Call to Action to the Tech Community on New Machine Readable COVID-19 Dataset” (CORD-19).
➡️ On April 7, 2020, WellAI releases its Machine Learning Analytics Tool for Medical Researchers.
➡️ Shortly after a group of international scientists and medical researchers started working on the first paper ever that analyzes AI applications in the fight against COVID-19. By the end of April 2020, the first draft was finished and submitted to the journal of the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC). WellAI’s AI engine was at the heart of the research.
➡️ Realizing the importance and the time sensitivity of the research, the editors and reviewers turned revisions of the paper around in record terms. The study was published on June 2, 2020.
➡️ In January 2021, the Working Group on Artificial Intelligence and Genomic Diagnostics (WG-AIGD) was born. The group is the leader of the global medical community in assessing the role of AI in genomic tests for detecting COVID-19 and developing recommendations/best practices for clinical laboratories validating and evaluating AI-based diagnostic and prognostic methods.
What an Incredible and expedited timeline for such a level of research. The work has involved so many people: data scientists, coders, engineers, molecular diagnostics experts, genome researchers, editors, and reviewers.
Our current research is focused on ethical AI and best practices of AI in healthcare. It’s a super important topic with, once again, our work at WellAI front and center. We’re carefully reviewing academic research and connecting it with the practical implications of the WellAI technology. Stay tuned for the publication… 😊