UBC and BC Cancer researchers have created an AI model that can predict cancer patient survival with greater than 80% accuracy. The model analyzes oncologist notes after a patient's initial consultation visit using natural language processing (NLP). The model can predict six-month, 36-month, and 60-month survival rates more accurately than previous tools by identifying unique characteristics of each patient.
Cancer survival rates have traditionally been calculated retrospectively and classified by only a few generic factors. The new model, on the other hand, can detect unique clues within a patient's initial consultation document to provide a more nuanced assessment. This personalized approach could be used to optimize the immediate care a patient receives, ensuring the best possible outcome.
Unlike chart reviews by human research assistants, the AI approach has the added benefit of maintaining complete confidentiality of patient records. The researchers trained and tested the model using data from 47,625 patients across all six BC Cancer sites. Previous models were limited to specific cancer types, whereas the AI model can be applied to all cancers. This advancement in cancer care may result in earlier referrals to support services or more aggressive treatment options.
Source: Techouver (link)
Images: MidJourney, Prompts by Lewis Farrell
Interesting story? Please click on the ? button below!