[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-353912-105":59,"doc-detail-353912-en":130},{"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":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","incremental-learning-approach-for-semantic-segmentation-of-skin-histology-images","Incremental learning approach for semantic segmentation of skin histology images","","This study presents an incremental learning framework to strengthen the generalization and robustness of transformer-based deep learning models for segmenting skin cancer and related tissue structures. Instead of relying on static training distributions, the method incrementally incorporates new data while reducing catastrophic forgetting through a biologically inspired strategy. Multiple loss functions preserve prior knowledge while adapting to additional magnification levels. Experiments on an indistribution test set show consistent accuracy gains across magnification combinations, improving empirical generalization in skin cancer segmentation.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":35,"@type":76,"position":81},"https://docshare.wps.com/document/healthcare/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/incremental-learning-approach-for-semantic-segmentation-of-skin-histology-images/353912/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/incremental-learning-approach-for-semantic-segmentation-of-skin-histology-images/353912.png","ImageObject",300,407,{"name":92,"@type":93},"Jake","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-23","2026-09-22",true,{"@type":102,"interactionType":103,"userInteractionCount":34},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does the incremental learning framework address in skin cancer segmentation?","Question",{"text":112,"@type":113},"Deep learning models often degrade when encountering novel scenarios outside their training distributions, limiting clinical usefulness. The framework incrementally adapts to new data while improving robustness.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does the proposed method reduce catastrophic forgetting?",{"text":117,"@type":113},"It combines multiple loss functions with knowledge distillation, transferring knowledge from a previously trained teacher to a student. It also includes mutual distillation loss for better performance in incremental semantic segmentation.",{"name":119,"@type":110,"acceptedAnswer":120},"What performance improvements were reported on the indistribution test set?",{"text":121,"@type":113},"The approach achieved 89.05% accuracy with 10× magnification, 92.68% with 10× and 5× combined, and 95.53% when incorporating 10×, 5×, and 2× magnifications.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},353912,1790659096,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":34,"category_name":35,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":34,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145},962084928904,"https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nIncremental learning approach for semantic segmentation of skin histology images  \nSana Fatima1,2, Anum Abdul Salam2, Muhammad Usman Akram2, Ibrahim A. Hameed3 &  \nSaad B. Ahmed1􀀍  \nThis study presents an incremental learning framework to enhance the generalization and robustness of transformer-based deep learning models for segmenting skin cancer and related tissue structures. While deep learning models often perform well on data distributions similar to their training sets, their accuracy typically degrades when exposed to novel scenarios–limiting their clinical utility in skin cancer diagnosis. To address this, we propose a biologically inspired incremental learning strategy tailored for skin cancer classification and segmentation, allowing the model to incorporate new data progressively while reducing catastrophic forgetting. Our approach integrates multiple loss functions to preserve existing knowledge while adapting to additional magnification levels. Experimental results on the indistribution test set demonstrate consistent performance improvements: achieving 89.05% accuracy with 10 × magnification, 92.68% with 10 × and 5 × combined, and 95.53% when incorporating  \n10 ×, 5 ×, and 2 × magnifications. These findings highlight the potential of our method to improve the adaptability and reliability of deep learning systems for empirical generalization in skin cancer classification tasks.  \nKeywords Skin cancer, Deep learning, Incremental learning, Knowledge distillation, Mutual distillation loss, Transformer-based architecture, Data incremental learning  \nEvaluating the advances and shortcomings related to the Sustainable Development Goals (SDGs) today is vital for decision makers seeking to improve population health1. One of the latest estimates is that the global burden of cancer will continue to increase during the next two decades2. Skin cancer is one of the most common diseaseson the globe. There are two general skin cancer types: non-melanoma skin cancer and melanoma skin cancer. In the United States, non-melanoma skin cancers (NMSCs), such as squamous cell carcinoma (SCC) and basal cell carcinoma (BCC), make up over 98% of all skin cancer cases3. The Cancer Society of America states that melanoma skin cancer makes up just 1% of all cancer cases, and its death rate is higher4. A timely and precise diagnosis is still essential for successful treatment5.  \nSkin cancer diagnosis  \nTo correctly identify and treat skin cancer, pathologists examine several tissue slides that have different histomorphological traits. For differentiation among different histological layouts, all excision specimens must be focused, cut into 3 mm sections, and dyed with hematoxylin and eosin (H&E)6,7. However, earlier approaches that digitize skin cancer diagnosis using binary classification provide limited clinical and economic value, as they offer less informative outputs compared to multi-class semantic segmentation. In contrast, segmentation-based methods can deliver more detailed insights, enhancing decision-making confidence for both professionals and patients, where accurate detection and diagnosis are critical. While multi-class semantic segmentation methods offer more detailed reports suitable for clinical decision-making, binary classification methods often lack this level of clarity. Furthermore, Fig 1 shows the regions that are the most affected by skin cancer according to global statistics, with North America representing approximately fifty percent of all instances8.  \nOur proposed research uses histology images, and microscopic images oftissues used to analyze the disease, particularly in cancer diagnosis. Using the histology images, we proposed an incremental learning semantic segmentation of a multi-class framework on NMSC. Our study’s contribution is as follows;  \n1Department of Computer Science, Lakehead University, Thunder Bay, ON P7B 5E1, Canada","cbCaitLYWDv5W5PY","https://ap.wps.com/l/cbCaitLYWDv5W5PY","pdf",8345866,17,"English","# Skin cancer diagnosis\n# Proposed incremental learning semantic segmentation\n# Literature review","[{\"question\":\"What problem does the incremental learning framework address in skin cancer segmentation?\",\"answer\":\"Deep learning models often degrade when encountering novel scenarios outside their training distributions, limiting clinical usefulness. The framework incrementally adapts to new data while improving robustness.\"},{\"question\":\"How does the proposed method reduce catastrophic forgetting?\",\"answer\":\"It combines multiple loss functions with knowledge distillation, transferring knowledge from a previously trained teacher to a student. It also includes mutual distillation loss for better performance in incremental semantic segmentation.\"},{\"question\":\"What performance improvements were reported on the indistribution test set?\",\"answer\":\"The approach achieved 89.05% accuracy with 10× magnification, 92.68% with 10× and 5× combined, and 95.53% when incorporating 10×, 5×, and 2× magnifications.\"}]","Incremental learning approach for semantic segmentation of skin histology images | PDF",1790107881,43]