[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-detail-369888-en":59,"doc-seo-369888-105":81},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":5,"data":60},{"doc_id":61,"user_id":62,"nickname":63,"user_avatar":64,"doc_module":4,"category_id":34,"category_name":35,"doc_title":65,"doc_description":66,"doc_content":67,"file_id":68,"file_url":69,"file_type":70,"file_size":71,"view_count":44,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":72,"language":73,"language_code":74,"site_id":75,"html_lang":74,"table_of_contents":76,"faqs":77,"seo_title":78,"seo_description":66,"update_tm":79,"read_time":80},369888,7421720224475,"Mary Man","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c","Attention-enhanced hybrid U-Net for prostate cancer grading and explainability","Attention-enhanced hybrid U-Net (TAH U-Net) is proposed to improve prostate cancer histopathology segmentation for reliable Gleason Grade assessment. Existing CNN or conventional U-Net approaches lack contextual awareness and explainability, reducing robustness on heterogeneous tissues and weakening boundary delineation. The method combines ResNet50-based local feature extraction with Vision Transformer blocks for long-range dependencies, adds attention-guided skip connections, and applies a multi-stage hierarchical guided loss for boundary refinement. Explainable AI validation using LIME, Occlusion Sensitivity, and PDP supports clinical transparency. On SICAPv2, it achieves state-of-the-art results, outperforming traditional U-Net with higher Dice and IoU and improved precision and recall.","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nAttention-enhanced hybrid U-Net for prostate cancer grading and explainability  \nAhmad Nawaz Zaheer1􀀍, Muhammad Farhan2,6, Guilin Min1,3,6, Faiz Abdullah Alotaibi4 &  \nMrim M. Alnfiai5  \nProstate cancer remains a leading cause of mortality, necessitating precise histopathological segmentation for accurate Gleason Grade assessment. However, existing deep learning-based segmentation models lack contextual awareness and explainability, leading to inconsistent performance across heterogeneous tissue structures. Conventional U-Net architectures and CNNbased approaches struggle with capturing long-range dependencies and fine-grained histopathological patterns, resulting in suboptimal boundary delineation and model generalizability. To address these limitations, we propose a transformer-attention hybrid U-Net (TAH U-Net), integrating hybrid CNNtransformer encoding, attention-guided skip connections, and a multi-stage guided loss mechanism for enhanced segmentation accuracy and model interpretability. The ResNet50-based convolutional layers efficiently capture local spatial features, while Vision Transformer (ViT) blocks model global contextual dependencies, improving segmentation consistency. Attention mechanisms are incorporated into skip connections and decoder pathways, refining feature propagation by suppressing irrelevant tissue noise while enhancing diagnostically significant regions. A novel hierarchical guided loss function optimizes segmentation masks at multiple decoder stages, improving boundary refinement and gradient stability. Additionally, Explainable AI (XAI) techniques such as LIME, Occlusion Sensitivity, and Partial Dependence Analysis (PDP), validate the model’s decision-making transparency, ensuring clinical reliability. The experimental evaluation on the SICAPv2 dataset demonstrates state-of-theart performance, surpassing traditional U-Net architectures with a 4.6% increase in Dice Score, 5.1% gain in IoU, along with notable improvements in Precision (+ 4.2%) and Recall (+ 3.8%). This research significantly advances AI-driven prostate cancer diagnostics by providing an interpretable and highly accurate segmentation framework, enhancing clinical trust in histopathology-based grading within medical imaging and computational pathology.  \nKeywords Prostate cancer segmentation, Transformer-attention hybrid U-Net, Histopathology imaging, Explainable AI (XAI), Multi-stage guided loss, Gleason grade classification  \nHowever, prostate cancer is still one of the most common and frightening cancers in men the world over. Grading prostate cancer early and accurately to determine more effective treatment plans, which is important for better patient outcomes1. As medical imaging and artificial intelligence (AI) advances, deep learning-based methods become widely used for automated cancer detection, grading, and prognosis prediction. Convolutional neural networks (CNN) have shown strong capability in extracting spatial features from medical images among these2,3. Although they added to the existing work on prostate cancer grading, challenges remain in terms of heterogeneity among tumor regions, heterogeneity in tissue morphology, and the requirement for transparent and interpretable AI-driven diagnostic systems4. As prostate cancer grade is sensitive to image focus, there is a need for models that can focus on critical regions in medical images and give clear, interpretable reasoning behind their predictions. Despite AI that improves medical imaging processes, prostate cancer grading is still problematic. Due to variability in tissue morphology5–7, and staining protocols, or because of imaging equipment,  \n1BOYA International College, Jiangxi University of Technology, 115 Ziyang Avenue, Nanchang 330022, Jiangxi, China. 2Department of Computer Science, COMSATS University Islamabad, Sahiwal Campus, Sahiwal, Pakistan.  \n3College of Finance and Economic","cbCaiq3X2iCMdZ5t","https://ap.wps.com/l/cbCaiq3X2iCMdZ5t","pdf",5659695,21,"English","en",105,"# Attention-enhanced hybrid U-Net for prostate cancer grading and explainability\n## Motivation and limitations\n## Proposed TAH U-Net architecture\n## Explainable AI validation\n## Experimental results and improvements","[{\"question\":\"Why is explainability important for prostate cancer grading models?\",\"answer\":\"Conventional deep learning segmentation models are often opaque, which can reduce clinical trust. TAH U-Net includes Explainable AI methods to validate decision transparency for reliable histopathology-based grading.\"},{\"question\":\"What architectural components does TAH U-Net use?\",\"answer\":\"TAH U-Net integrates hybrid CNN-transformer encoding, with ResNet50 layers capturing local spatial features and Vision Transformer (ViT) blocks modeling global contextual dependencies. Attention-guided skip connections and decoder pathways refine feature propagation.\"},{\"question\":\"How does the multi-stage guided loss improve segmentation?\",\"answer\":\"A hierarchical guided loss function optimizes segmentation masks at multiple decoder stages, improving boundary refinement and stabilizing gradients. This leads to better segmentation accuracy than traditional U-Net variants.\"}]","Attention-enhanced hybrid U-Net for prostate cancer grading and explainability | PDF",1790185086,53,{"code":4,"msg":82,"data":83},"ok",{"site_id":75,"language":74,"slug":84,"title":65,"keywords":85,"description":66,"schema_data":86,"social_meta":140,"head_meta":142,"extra_data":144,"updated_unix":145},"attention-enhanced-hybrid-u-net-for-prostate-cancer-grading-and-explainability","",{"@graph":87,"@context":139},[88,102,122],{"@type":89,"itemListElement":90},"BreadcrumbList",[91,95,97,100],{"item":92,"name":93,"@type":94,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":96,"name":9,"@type":94,"position":14},"https://docshare.wps.com/document/",{"item":98,"name":35,"@type":94,"position":99},"https://docshare.wps.com/document/healthcare/",3,{"item":101,"name":65,"@type":94,"position":19},"https://docshare.wps.com/document/attention-enhanced-hybrid-u-net-for-prostate-cancer-grading-and-explainability/369888/",{"url":101,"name":65,"@type":103,"image":104,"author":109,"headline":65,"publisher":111,"fileFormat":114,"inLanguage":74,"description":66,"dateModified":115,"datePublished":116,"encodingFormat":114,"isAccessibleForFree":117,"interactionStatistic":118},"DigitalDocument",{"url":105,"@type":106,"width":107,"height":108},"https://docshare.wps.com/thumbnails/attention-enhanced-hybrid-u-net-for-prostate-cancer-grading-and-explainability/369888.png","ImageObject",300,407,{"name":63,"@type":110},"Person",{"url":92,"name":112,"@type":113},"DocShare","Organization","application/pdf","2026-10-27","2026-09-23",true,{"@type":119,"interactionType":120,"userInteractionCount":44},"InteractionCounter",{"@type":121},"ViewAction",{"@type":123,"mainEntity":124},"FAQPage",[125,131,135],{"name":126,"@type":127,"acceptedAnswer":128},"Why is explainability important for prostate cancer grading models?","Question",{"text":129,"@type":130},"Conventional deep learning segmentation models are often opaque, which can reduce clinical trust. TAH U-Net includes Explainable AI methods to validate decision transparency for reliable histopathology-based grading.","Answer",{"name":132,"@type":127,"acceptedAnswer":133},"What architectural components does TAH U-Net use?",{"text":134,"@type":130},"TAH U-Net integrates hybrid CNN-transformer encoding, with ResNet50 layers capturing local spatial features and Vision Transformer (ViT) blocks modeling global contextual dependencies. Attention-guided skip connections and decoder pathways refine feature propagation.",{"name":136,"@type":127,"acceptedAnswer":137},"How does the multi-stage guided loss improve segmentation?",{"text":138,"@type":130},"A hierarchical guided loss function optimizes segmentation masks at multiple decoder stages, improving boundary refinement and stabilizing gradients. This leads to better segmentation accuracy than traditional U-Net variants.","https://schema.org",{"og:url":101,"og:type":141,"og:title":65,"og:site_name":112,"og:description":66},"article",{"robots":143,"canonical":101},"index,follow",{"doc_id":61,"site_id":75},1791287852]