Researchers at Weill Cornell Medicine have created a new predictive model using Artificial Intelligence and machine learning to better forecast how patients with muscle-invasive bladder cancer will respond to chemotherapy. This model stands out by integrating comprehensive data, including whole-slide tumor imaging and gene expression analyses, surpassing previous models that relied on a single data type.
Published in npj Digital Medicine, the study highlights the model’s capacity to identify key genes and tumor characteristics, vital for predicting treatment success. Such precision could enable medical professionals to personalize treatments, potentially sparing patients from invasive procedures like bladder removal.
The model’s development was a collaboration led by Fei Wang and Dr. Bishoy Morris Faltas, integrating data from clinical trials to craft a model that combines both imaging and gene expression data for superior prediction accuracy. Early results show a significant improvement in prediction capability over earlier models, promising a shift towards more personalized and effective cancer care strategies.
