Study of bladder cancer detection in standard white light versus AI-supported endoscopy-01 (RAISE-01) – Development and validation of an AI-based support too

Comprehensive Summary

Hjort et al present a study in which an artificial intelligence (AI) tool called CystoAID was developed and validated to improve detection of bladder cancer during cystoscopy, the standard diagnostic procedure that uses white-light endoscopy. The model, based on a convolutional neural network, was trained on video recordings from cystoscopies and tumor resections and then tested on an external dataset reflecting real clinical practice. The AI system achieved very high sensitivity (100%) and strong precision (~88%), outperforming typical white-light cystoscopy in detecting bladder lesions while operating fast enough for real-time use. Overall, the study shows that AI-supported endoscopy can reliably identify bladder tumors and reduce the risk of missed lesions, which are a major cause of cancer recurrence.

Outcomes and Implications

This research is important because bladder cancer diagnosis currently depends on cystoscopy, which can miss small or flat tumors and contributes to high recurrence rates. The AI tool demonstrated strong potential to assist clinicians in real time, improving detection accuracy and consistency regardless of operator experience. Clinically, this could lead to earlier diagnosis, more complete tumor removal, and better patient outcomes, but the authors emphasize that prospective trials are still needed before widespread implementation in routine urologic practice.