Salvador Jaime-Casas, Postdoctoral Research Fellow at City of Hope, shared on X:
“Very excited to share our most recent publication lead by stellar mentors Francesco Del Giudice, Benjamin Chung – a SR of AI models for tumor detection during cysto in pts with suspected Bladder Cancer.
Following PRISMA guidelines:
- 12,840 patients, 13 studies (2018 – 2025) were included.
- 236,925 frames, 36,890 contained lesion
- AI models: CNNs, U-Net variants, GoogLeNet, ResNet, DenseNet, EfficientNet, CystoNet, and CAIDS
- Mod to high accuracy across models
Four AI model clusters:
- CNN-based architectures (8 studies)
- CAIDS architectures (2 studies)
- BSU architecture (1 study)
- non-CNN-based and proprietary models (2 studies)
Overall:
- Promising performance, but context and study design specific
- Better when used as adjunct with human-analyzed frames
- Lack of widespread implementation, mostly limited to academic centers
- Mainly retrospective designs, and single-frames analysis (vs. video)”
Title: Artificial intelligence (AI)-based models for bladder cancer (BC) endoscopic detection: a systematic review
Authors: Salvador Jaime-Casas, Roberta Corvino, Jan Łaszkiewicz, Amir Khan, Benjamin I. Chung, Vincenzo Asero, Valerio Santarelli, Dalila Carino, Sarah Carvalho Ribeiro, Stefano Impaloni, Francesco Del Giudice
Read also “How I Treat Bladder Cancer in 2026: OncoDaily Virtual Summit” on OncoDaily.
