Teaching AI to Spot Bladder Cancer During Blue Light Exams

This study is collecting video and image data from bladder cancer screening procedures called blue light cystoscopy—a procedure where doctors use a special camera and fluorescent dye to look inside the bladder for cancerous lesions. The goal is to gather this information to train an artificial intelligence (AI) computer system to automatically detect bladder cancer during these exams, which could help doctors catch more cancers and reduce human error.

Currently, doctors visually examine the bladder during these procedures, which can be subjective and varies depending on the physician's experience. This medication helps highlight cancer cells under blue light, but the decision of what to biopsy still depends on the doctor's judgment. By training AI on many real-world exam recordings, researchers hope to create a system that can automatically spot suspicious lesions in real time and help physicians make better decisions.