We published the 22 of july a paper in Nature Cancer on MEDomics which is
a data-integrating framework that enables continuous learning from multimodal health
data while simultaneously capturing and integrating longitudinal clinical records.
It uses Natural Language Processing, AI and imaging, across a large and multicentric healthcare system (watch the animation: https://youtu.be/2030Pdgm3_4). Most importantly, after having analysed 15 million medical notes, 200 million
laboratory results, 1.7 million radiology images and 0.86 million pathology reports across different cancer types, we confirmed results of RCT’s on Real World Data (e.g. ICI in lung cancer stage IV) and it allows us to validate several new clinical hypotheses like the importance of cardio-vascular morbidity for the outcome of cancer treatment. We are grateful to our colleagues of the MEDomics consortium. https://www.nature.com/articles/s43018-021-00236-2
Éditorial of J. Deasy (MSKCC, New-York): https://www.nature.com/articles/s43018-021-00239-z
Press release: https://www.prnewswire.com/news-releases/promising-development-in-the-fight-with-cancer-using-artificial-intelligence-with-a-new-privacy-preserving-approach-named-medomics-301339649.html?tc=eml_cleartime MEDomics animation: https://youtu.be/2030Pdgm3_4