AI scans tumor images to predict cancer drug response, bypassing genetic tests

Researchers demonstrated that machine learning can analyze colorectal cancer scans to identify which tumors will respond to specific immunotherapies—potentially eliminating costly genetic testing. The finding could streamline cancer treatment decisions and reduce diagnosis time for thousands of patients annually.

Originaltitel: Systematic review of machine learning-based radiomics approach for predicting microsatellite instability status in colorectal cancer