Current cancer drug targets · 2026
Integrative Deep Learning from H&E Images Reveals Prognostically Distinct Pathology-Based Subtypes in Bladder Cancer.
СтатьяНазвание и аннотация: Перевод в очереди
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.
Li Huanhui, Dai Fazhong, Zhang Yongqiang, Li Xiaoyang, Zhong Biling, Fang Jia, Wang Zhenwei, He Yanyan, Wang Mancun, Qiu Xiaofu, Zheng Zongtai
Аннотация
BACKGROUND: Molecular subtyping guides bladder cancer (BCa) care but typically requires RNA profiling. This study aimed to develop pathology-based subtypes of BCa using pathology deep learning features derived from routinely obtained hematoxylin and eosin (H&E)-stained whole-slide images (WSIs).
METHODS: We developed a pathology-based subtype of BCa based on deep learning features extracted from H&E-stained WSIs. A modified Resnet50 model was trained to distinguish between tumor and normal regions and extract patch-level deep learning features. These features were aggregated at the WSI level, followed by weighted gene co-expression network analysis (WGCNA), Cox regression, and unsupervised K-means clustering to define pathology-based subtypes. External validation was performed using WSIs from four independent centers and transcriptomic data from IMvigor210 and GSE32894 cohorts. Interpretability used Grad-CAM on tumor patches.
RESULTS AND DISCUSSION: The Resnet50 model achieved high and consistent performance in distinguishing tumor and normal patches. The analysis identified four BCa subtypes, with distinct clinical outcomes. Clusters 2/3 were predominantly luminal-like (cluster 3 was enriched for FGFR3 and had the lowest TMB), whereas clusters 0/1 exhibited mixed luminal/ basal features with higher immune/stromal scores; regulon and pathway profiles diverged (e.g., EGFR/FOXM1/STAT3 in cluster 1 vs. FGFR3/FOXA1/PPARG in cluster 3). Grad-CAM highlighted distinct nuclear morphologies supporting interpretability.
CONCLUSION: We present a pathology-based subtype of BCa that is portable, prognostically informative, and mechanistically concordant with RNA-defined biology. The approach offers a cost-effective and accessible alternative to traditional molecular profiling methods, with potential applications in personalized treatment and improved patient stratification.
METHODS: We developed a pathology-based subtype of BCa based on deep learning features extracted from H&E-stained WSIs. A modified Resnet50 model was trained to distinguish between tumor and normal regions and extract patch-level deep learning features. These features were aggregated at the WSI level, followed by weighted gene co-expression network analysis (WGCNA), Cox regression, and unsupervised K-means clustering to define pathology-based subtypes. External validation was performed using WSIs from four independent centers and transcriptomic data from IMvigor210 and GSE32894 cohorts. Interpretability used Grad-CAM on tumor patches.
RESULTS AND DISCUSSION: The Resnet50 model achieved high and consistent performance in distinguishing tumor and normal patches. The analysis identified four BCa subtypes, with distinct clinical outcomes. Clusters 2/3 were predominantly luminal-like (cluster 3 was enriched for FGFR3 and had the lowest TMB), whereas clusters 0/1 exhibited mixed luminal/ basal features with higher immune/stromal scores; regulon and pathway profiles diverged (e.g., EGFR/FOXM1/STAT3 in cluster 1 vs. FGFR3/FOXA1/PPARG in cluster 3). Grad-CAM highlighted distinct nuclear morphologies supporting interpretability.
CONCLUSION: We present a pathology-based subtype of BCa that is portable, prognostically informative, and mechanistically concordant with RNA-defined biology. The approach offers a cost-effective and accessible alternative to traditional molecular profiling methods, with potential applications in personalized treatment and improved patient stratification.