Abstract
Artificial intelligence (AI) is emerging as a transformative tool in the diagnosis of bladder cancer, offering the potential to enhance accuracy, consistency, and early detection. This narrative review aimed to summarize and critically appraise recent developments in AI applications across diagnostic modalities, based on studies identified through PubMed, Scopus, and Google Scholar up to April 2025. Evidence shows that deep learning algorithms applied to cystoscopy improve lesion detection, including subtle flat and early-stage tumors that may be overlooked by conventional assessment. In histopathology and cytology, AI systems contribute to grading, classification, and identification of malignant features with accuracy comparable to expert pathologists. Integration of AI with urinary biomarkers and genomic data further supports personalized risk prediction and molecular characterization. Additionally, AI-driven clinical decision support tools assist in prognosis, treatment planning, and post-therapy surveillance. While current findings underscore AI's promising role in improving diagnostic precision and workflow efficiency, its clinical adoption requires addressing issues of data quality, algorithm transparency, and ethical governance. Future research should focus on developing explainable and validated models through multicenter collaborations between clinicians and data scientists to facilitate safe and reliable integration of AI into routine bladder cancer diagnosis.