BJU international · 2026
Artificial intelligence in urology: A review of United States Food and Drug Administration-cleared devices.
СтатьяОбзорНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Qian Zhiyu, Brin Pietro, Korn Stephan, Cosenza Andrea, Zurl Hanna, Pohl Klara, Pakpoor Jina, Moore Caroline M, Giganti Francesco, Trinh Quoc-Dien, Cole Alexander P
Аннотация
OBJECTIVE: To systematically characterise United States Food and Drug Administration (FDA) authorised urology-specific artificial intelligence (AI)-enabled medical devices and to evaluate the publicly available evidence supporting their authorisation.
METHODS: We reviewed the FDA AI-Enabled Medical Devices List. Devices cleared by September 2025 were manually screened for urological relevance using FDA clearance documents (e.g., 510(k), De Novo) to confirm clinical application. Eligible devices were categorised by FDA approval pathway, review panel, indication, and year of approval. FDA decision summaries were reviewed to characterise study design, training/testing and external validation features, and reporting of patient demographics. Descriptive statistics were used to summarise distributions and temporal trends.
RESULTS: Of 1358 FDA-cleared AI devices, 30 (2.2%) were urology specific. Approvals increased over time, with 26.7% cleared in 2025 (through September). The predominant FDA medical device category was Radiology (70.0%), with indications led by prostate imaging (43.3%). Information on study design, recruitment, and validation was frequently unavailable. Race and ethnicity were reported in one device (3.3%) for training/testing and four devices (13.3%) for external validation. Among these, aggregated proportions showed that the majority of subjects were White (86.1% for training/testing; 87.6% for external validation).
CONCLUSIONS: The FDA-authorised urological AI devices are dominated by radiology-based, prostate-focused applications and have increased in recent years. However, publicly available evidence supporting their authorisation is often limited, with incomplete reporting of study design and patient demographics. Greater transparency and post-authorisation evaluation are needed to support interpretation of these tools in clinical practice.
METHODS: We reviewed the FDA AI-Enabled Medical Devices List. Devices cleared by September 2025 were manually screened for urological relevance using FDA clearance documents (e.g., 510(k), De Novo) to confirm clinical application. Eligible devices were categorised by FDA approval pathway, review panel, indication, and year of approval. FDA decision summaries were reviewed to characterise study design, training/testing and external validation features, and reporting of patient demographics. Descriptive statistics were used to summarise distributions and temporal trends.
RESULTS: Of 1358 FDA-cleared AI devices, 30 (2.2%) were urology specific. Approvals increased over time, with 26.7% cleared in 2025 (through September). The predominant FDA medical device category was Radiology (70.0%), with indications led by prostate imaging (43.3%). Information on study design, recruitment, and validation was frequently unavailable. Race and ethnicity were reported in one device (3.3%) for training/testing and four devices (13.3%) for external validation. Among these, aggregated proportions showed that the majority of subjects were White (86.1% for training/testing; 87.6% for external validation).
CONCLUSIONS: The FDA-authorised urological AI devices are dominated by radiology-based, prostate-focused applications and have increased in recent years. However, publicly available evidence supporting their authorisation is often limited, with incomplete reporting of study design and patient demographics. Greater transparency and post-authorisation evaluation are needed to support interpretation of these tools in clinical practice.