Posted on 13 February 2020
A study published in the Lancet has developped an AI-based tool that classifies patients into prognostic groups by scanning tumour tissue sections. 10 neural networks were trained using over 12 million images from patients.
The tool has the potential to guide the selection of adjuvant treatment by avoiding therapy in very low-risk groups, and at the same time, identifying patients who would benefit from more intensive treatment regimes.
Previous methods could result in 80% of patients receiving unclear diagnoses, while the new method could cut this to just 12%.
Deep learning for prediction of colorectal cancer outcome: a discovery and validation study: https://doi.org/10.1016/S0140-6736(19)32998-8
AI-based marker for colon cancer can help improve quality of treatment: https://medicalxpress.com/news/2020-02-ai-based-marker-colon-cancer-quality.html