Archivos espanoles de urologia · 2026
Artificial Intelligence and Big Data in Urological Oncology: From Radiomics to Real-World Evidence.
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.
Katsimperis Stamatios, Tzelves Lazaros, Kyriazis Ioannis, Neofytou Panagiotis, Kapsalos-Dedes Sotirios, Feretzakis Georgios, Skolarikos Andreas
Аннотация
BACKGROUND: Artificial intelligence (AI) and big data are transforming urological oncology by enhancing diagnostic precision, prognostic assessment and treatment personalisation for prostate, bladder and kidney cancer.
METHODS: We searched PubMed and MEDLINE up to September 2025 for English-language, peer-reviewed human studies using terms including "artificial intelligence", "deep learning", "radiomics", "real-world evidence" and "urological oncology".
RESULTS: AI-driven radiomics and deep learning models have demonstrated high accuracy in detecting and characterising urological malignancies by using magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET) and histopathology. In prostate, bladder and kidney cancers, AI-driven radiomics and deep learning models have demonstrated high diagnostic performance, with reported area under the curves (AUCs) typically ranging from 0.80 to 0.95 for lesion detection, staging and risk stratification. Sensitivities and specificities in cystoscopic image analysis often exceed 90%, but radiogenomic models for renal cancer achieve mutation prediction accuracies of 85%-95%.
CONCLUSIONS: AI and big data are reshaping urological oncology by integrating diagnostic imaging, pathology and real-world practice. Their continued integration promises a precise, equitable and adaptive model of cancer care. Despite these robust results, most studies rely on retrospective or single-centre datasets with limited external validation, raising concerns about generalisability. Future progress will depend on multicentre standardisation, federated learning frameworks and incorporation of multimodal real-world data to facilitate clinically robust and implementable AI systems.
METHODS: We searched PubMed and MEDLINE up to September 2025 for English-language, peer-reviewed human studies using terms including "artificial intelligence", "deep learning", "radiomics", "real-world evidence" and "urological oncology".
RESULTS: AI-driven radiomics and deep learning models have demonstrated high accuracy in detecting and characterising urological malignancies by using magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET) and histopathology. In prostate, bladder and kidney cancers, AI-driven radiomics and deep learning models have demonstrated high diagnostic performance, with reported area under the curves (AUCs) typically ranging from 0.80 to 0.95 for lesion detection, staging and risk stratification. Sensitivities and specificities in cystoscopic image analysis often exceed 90%, but radiogenomic models for renal cancer achieve mutation prediction accuracies of 85%-95%.
CONCLUSIONS: AI and big data are reshaping urological oncology by integrating diagnostic imaging, pathology and real-world practice. Their continued integration promises a precise, equitable and adaptive model of cancer care. Despite these robust results, most studies rely on retrospective or single-centre datasets with limited external validation, raising concerns about generalisability. Future progress will depend on multicentre standardisation, federated learning frameworks and incorporation of multimodal real-world data to facilitate clinically robust and implementable AI systems.