The study shows AI could act like a virtual assistant for physicians. The model tested in B.C. predicts cancer patient survival more accurately and with more readily available data than previous tools. It uses natural language processing to analyze oncologist notes following a patient’s initial consultation visit. By identifying characteristics unique to each patient, the model was shown to predict six-month, 36-month and 60-month survival with greater than 80% accuracy. The technology's applicable to all cancers, whereas previous models were limited to certain cancer types. The model's trained on B.C. data but researchers say the technology could be applied in cancer clinics across Canada and around the world.
Related:
AI model can detect AD with over 90% accuracy, research finds - CTV News
UBC AI model predicts cancer patient survival by reading doctors’ notes
Note : Ce résumé a été produit à partir de la nouvelle sous Lire l'article ci-dessous. Si vous avez des questions, prière de les acheminer à cette source.
mars 02, 2023