[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-detail-351630-en":59,"doc-seo-351630-105":99},{"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":39,"category_name":40,"doc_title":65,"doc_description":66,"doc_content":67,"file_id":68,"file_url":69,"file_type":70,"file_size":71,"view_count":52,"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,"preview_page_count":19,"preview_urls":81},351630,1374402968488,"Đào","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0","Interpretable CRAM-Enhanced Lightweight Dual-Branch CNN for Real-Time Breast Cancer Histopathology in Internet-of-Medical-Things Environments","Breast cancer remains a major global health challenge, and histopathological image analysis is the diagnostic gold standard. Manual microscopy is slow and subjective, while common deep models are often too complex and opaque for real-time Internet of Medical Things (IoMT) deployment. An interpretable lightweight hybrid model is proposed by combining MobileNetV2 and EfficientNet-B0 with a contextual recurrent attention module (CRAM). CRAM reweights fused features to emphasize diagnostically relevant regions, reaching 99.9% accuracy and AUC of 1.00 with ~12 M parameters. Interpretability is provided via Grad-CAM and SHAP, enabling transparent, resource-aware support for point-of-care diagnostics and real clinical workflows, including future extension to tumor subtype classification.","Small  \n[www.small-journal.com](www.small-journal.com)  \nRESEARCH ARTICLE   \nInterpretable CRAM‑Enhanced Lightweight Dual‑Branch CNN for Real‑Time Breast Cancer Histopathology in Internet‑of‑Medical‑Things Environments  \nRoseline Oluwaseun Ogundokun1, 2, 3   Rotimi-Williams Bello2  Pius Adewale Owolawi2  Rytis Maskelinas4  Abdulsatar Abduljabbar Sultan5  \n1 Department of Multimedia Engineering, Kaunas University of Technology, Kaunas, Lithuania  2 Department of Computer Systems Engineering, Tshwane University of Technology, Pretoria, South Africa  3 Department of Computer Science, Redeemer’s University, Ede, Osun, Nigeria  4 Department of Software  \nEngineering, Kaunas University of Technology, Kaunas, Lithuania  5 Business Management Department, Catholic University in Erbil, Erbil, Kurdistan, Iraq Correspondence: Roseline Oluwaseun Ogundokun ([roseline.ogundokun@ktu.lt](roseline.ogundokun@ktu.lt))  \nReceived: 28 July 2025  Revised: 1 March 2026  Accepted: 4 March 2026  \nKeywords: breast cancer | histopathology | internet‑of‑medical‑things | interpretable | lightweight deep learning | smart diagnostic systems  \nABSTRACT  \nBreast cancer remains a primary global health concern, with histopathological image analysis serving as the diagnostic gold standard. However, manual microscopy is time-consuming and often subjective. While deep learning offers a powerful solution, existing models are typically too complex and opaque for real-time use in Internet of Medical Things (IoMT) environments. To address this, we propose an interpretable and lightweight hybrid deep learning model that combines MobileNetV2 and EfficientNet-B0, enhanced by a novel contextual recurrent attention module (CRAM) . CRAM refines fused features through attention-based weighting, improving focus on diagnostically relevant regions. The model achieved 99.9% classification accuracy and an AUC of 1.00, outperforming standalone baselines while remaining efficient (∼12 M parameters) and suitable for IoMT deployment. Interpretability is ensured through integrated Grad-CAM and SHAP analyses, which visually and quantitatively explain predictions by highlighting malignant tissue features that align with pathologist judgment. This balance of accuracy, efficiency, and transparency enables real-time, trustworthy diagnostics for resource-limited and point-of-care settings. Future work includes extending to multi-class tumor subtypes and clinical validation in real-world workflows. The proposed system represents a significant step toward making AI in digital pathology more accessible and explainable.  \n1  Introduction  \nBreast cancer remains one of the most prevalent and deadly cancers among women worldwide [1] . Early and accurate diagnosis, achieved through tissue biopsy and histopathological examination, is crucial for effective treatment and improved survival [2, 3] . In histopathology, pathologists examine thin tissue sections under a microscope and stain them to identify malignant morphological features [4] . This manual analysis is labor-intensive, time-consuming, and subjective, often resulting in inter-observer  \nvariability and diagnostic delays [5] . The integration of digital pathology with artificial intelligence offers a promising avenue to assist pathologists by automating the detection of cancerous patterns in histology images, thereby enhancing accuracy and consistency [6] . Deep learning, particularly convolutional neural networks (CNNs), has demonstrated remarkable performance in image classification tasks, including those in medical imaging and pathology [7, 8] . CNN-based models can automatically learn complex feature hierarchies from raw pixel data, obviating the need for manual feature engineering.  \nThis is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.  \n© 2026 The Author(s). Small published by Wiley-VCH Gm","cbCaikkvHrg1Afee","https://ap.wps.com/l/cbCaikkvHrg1Afee","pdf",2245083,25,"English","en",105,"# Introduction\n## Motivation: automated histopathology and diagnostic delays\n## Limits of deep learning in IoMT settings\n## Proposed interpretable lightweight CRAM-enhanced approach\n## Interpretability and explainable AI for trust","[{\"question\":\"Why is histopathology important in breast cancer diagnosis?\",\"answer\":\"Histopathological examination of stained tissue biopsy images is the diagnostic gold standard because it reveals malignant morphological features.\"},{\"question\":\"What problem does the proposed CRAM-enhanced model address?\",\"answer\":\"It targets the gap between high-performing but complex, opaque deep models and the need for lightweight, efficient, real-time inference with interpretability in IoMT environments.\"},{\"question\":\"How does the model improve interpretability of its predictions?\",\"answer\":\"It uses integrated Grad-CAM and SHAP analyses to visually and quantitatively explain which malignant tissue features drive the predicted class.\"}]","Interpretable CRAM-Enhanced Lightweight Dual-Branch CNN for Real-Time Breast Cancer Histopathology in Internet-of-Medical-Things Environments | PDF",1790095054,63,[82,86,90,95],{"page_no":8,"url":83,"width":84,"height":85},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/d4872a100a7c795495a7284a79f8adf1.webp",983,1283,{"page_no":14,"url":87,"width":88,"height":89},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/5392a67a2f777acb5b92efcf6210a9ba.webp",982,1281,{"page_no":91,"url":92,"width":93,"height":94},3,"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/d7ac2af96c86c51b3437366f072e5432.webp",981,1280,{"page_no":19,"url":96,"width":97,"height":98},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/5a2e5aa2f0ea17b6d96644ee174cbb09.webp",980,1279,{"code":4,"msg":100,"data":101},"ok",{"site_id":75,"language":74,"slug":102,"title":65,"keywords":103,"description":66,"schema_data":104,"social_meta":157,"head_meta":159,"extra_data":161,"updated_unix":162},"interpretable-cram-enhanced-lightweight-dual-branch-cnn-for-real-time-breast-cancer-histopathology-in-internet-of-medical-things-environments","",{"@graph":105,"@context":156},[106,119,139],{"@type":107,"itemListElement":108},"BreadcrumbList",[109,113,115,117],{"item":110,"name":111,"@type":112,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":114,"name":9,"@type":112,"position":14},"https://docshare.wps.com/document/",{"item":116,"name":40,"@type":112,"position":91},"https://docshare.wps.com/document/research-report/",{"item":118,"name":65,"@type":112,"position":19},"https://docshare.wps.com/document/interpretable-cram-enhanced-lightweight-dual-branch-cnn-for-real-time-breast-cancer-histopathology-in-internet-of-medical-things-environments/351630/",{"url":118,"name":65,"@type":120,"image":121,"author":126,"headline":65,"publisher":128,"fileFormat":131,"inLanguage":74,"description":66,"dateModified":132,"datePublished":133,"encodingFormat":131,"isAccessibleForFree":134,"interactionStatistic":135},"DigitalDocument",{"url":122,"@type":123,"width":124,"height":125},"https://docshare.wps.com/thumbnails/interpretable-cram-enhanced-lightweight-dual-branch-cnn-for-real-time-breast-cancer-histopathology-in-internet-of-medical-things-environments/351630.png","ImageObject",300,407,{"name":63,"@type":127},"Person",{"url":110,"name":129,"@type":130},"DocShare","Organization","application/pdf","2026-11-03","2026-09-22",true,{"@type":136,"interactionType":137,"userInteractionCount":52},"InteractionCounter",{"@type":138},"ViewAction",{"@type":140,"mainEntity":141},"FAQPage",[142,148,152],{"name":143,"@type":144,"acceptedAnswer":145},"Why is histopathology important in breast cancer diagnosis?","Question",{"text":146,"@type":147},"Histopathological examination of stained tissue biopsy images is the diagnostic gold standard because it reveals malignant morphological features.","Answer",{"name":149,"@type":144,"acceptedAnswer":150},"What problem does the proposed CRAM-enhanced model address?",{"text":151,"@type":147},"It targets the gap between high-performing but complex, opaque deep models and the need for lightweight, efficient, real-time inference with interpretability in IoMT environments.",{"name":153,"@type":144,"acceptedAnswer":154},"How does the model improve interpretability of its predictions?",{"text":155,"@type":147},"It uses integrated Grad-CAM and SHAP analyses to visually and quantitatively explain which malignant tissue features drive the predicted class.","https://schema.org",{"og:url":118,"og:type":158,"og:title":65,"og:site_name":129,"og:description":66},"article",{"robots":160,"canonical":118},"index,follow",{"doc_id":61,"site_id":75},1790203198]