Using machine learning approaches to predict tumour recurrence and progression to provide personalized and compassionate care for non-muscle invasive bladder cancer patients

Surgical Oncologist, Division of Urology at Trillium Health Partners

Bladder cancer is the most expensive cancer to treat due to high recurrence rates and a need for lifelong disease surveillance. Current monitoring strategies are costly, patient unfriendly, and lack supporting evidence. Existing predictive models perform poorly and do not capture changes in patients’ cancer course after initial diagnosis. Andrew’s study aims to integrate complete cancer timelines of each patient with machine learning and compassionate care approaches to improve prediction for tumour recurrence and progression.