[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-351550-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-351550-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","advanced-hybrid-transformer-cnn-framework-for-improved-skin-lesion-classification-and-segmentation","Advanced hybrid transformer CNN framework for improved skin lesion classification and segmentation","","Advanced hybrid transformer and CNN architectures target reliable skin lesion classification and segmentation for skin cancer detection. Deep learning shows promise for early identification, yet struggles with nonhomogeneous illumination, low contrast, and hair-on-skin dermoscopic images. The approach introduces a dual pipeline: GlobalSkinNet for classification using CNN-Transformer fusion optimized with Adam, evaluated on PH2, ISIC-2019/2020, and HAM10000; followed by SkinFormNet for segmentation using SegFormer feature extraction feeding an attention-based U-Net, assessed on PH2 and ISIC and HAM10000 datasets.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/advanced-hybrid-transformer-cnn-framework-for-improved-skin-lesion-classification-and-segmentation/351550/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/advanced-hybrid-transformer-cnn-framework-for-improved-skin-lesion-classification-and-segmentation/351550.png","ImageObject",300,407,{"name":42,"@type":43},"Melati","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-27","2026-09-22",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",12,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What challenges limit current deep learning skin cancer detection?","Question",{"text":63,"@type":64},"Models can struggle with nonhomogeneous luminance, low contrast, and hair-on-skin images, which affect accurate lesion detection.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How does the proposed framework perform skin lesion classification?",{"text":68,"@type":64},"It uses a transformer-based dual architecture where GlobalSkinNet combines convolutional and transformer components, trained with selected parameters and the Adam optimizer.",{"name":70,"@type":61,"acceptedAnswer":71},"How is skin lesion segmentation carried out in the proposed method?",{"text":72,"@type":64},"After classification, SkinFormNet processes the images: it extracts features using pre-trained SegFormer and feeds them into an attention-based U-Net for segmentation.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},351550,1790619383,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":140,"language":141,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":12,"update_tm":145,"read_time":146,"preview_page_count":33,"preview_urls":147},962085570644,"https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nAdvanced hybrid transformer CNN framework for improved skin lesion classification and segmentation  \nNazish Yousaf1,5, JavariaAmin2,5, Wasi Haider Butt1,5, Amad Zafar3,5 & Seonghan Kim3,4􀀍  \nAs reported by the World Health Organization (WHO), Skin Cancer (SC) is a growing medical concern with millions of cases diagnosed each year. Deep Learning methods show promising results in  \nearly detection and classification of SC, but still face challenges in accurate detection due to nonhomogeneous luminance, low contrast, and hair-on-skin images. Therefore, a novel transformer-based dual-architecture is proposed, comprising models for skin lesion classification and segmentation. The proposed classification model is based on the Global Contextual Vision Transformer (GlobalSkinNet), a combination of convolutional and transformer models. The proposed GlobalSkinNet model is trained using the selected parameters and the Adam optimizer for skin lesion classification. The GlobalSkinNet’s performance is assessed on four selected benchmark datasets: PH2, ISIC-2019, ISIC-2020, and HAM10000, achieving the accuracies of 100%, 98%, 97%, and 98%, respectively. The proposed SkinFormNet segmentation model subsequently processes the classified images. In this model, feature extraction is performed using the pre-trained SegFormer model, which is then input tothe U-Net architecture with an attention mechanism. The segmentation model’s performance is evaluated on five benchmark datasets, including PH2, ISIC-2016, ISIC-2017, ISIC-2018, and HAM10000, achieving a dice coefficient of 0.97, 0.98, 0.94, 0.99, and 0.96, respectively.  \nKeywords Skin lesion analysis, Skin lesion segmentation, Skin cancer classification, Vision Transformer, CNN, Deep learning  \nAccording to the WHO, SC is becoming a common public health risk with millions of cases diagnosed every year. SC originates in skin cells and usually results from abnormal cell growth, forming a lesion or mass. It can be lethal depending on its intensity and spread. Another primary reason for SC is direct exposure to UV (ultraviolet) radiation. UV radiation is harmful and can cause skin cancer, leading to death1. UV radiation is harmful to humans and is usually the main cause of death due to SC. According to research, 46,000 new SC cases are reported annually in the United Kingdom2. Research shows that fair-skinned races are more vulnerable to skin cancer than dark-skinned races, as the pigments in the skin protect the directly exposed skin layers3.  \nSC has two primary categories, melanoma and non-melanoma. Typically, melanoma is the most lethal type and tends to penetrate early. Symptoms usually include a mole with irregular borders that changes color, shape, or size. Malignant melanoma is a fatal and deadly type of cancer that affects a smaller number of the population but has the highest death rate. The risk of malignant melanoma is mainly related to UV exposure, behavior, personal characteristics, or genetics. It has a greater tendency to spread to its surroundings, making it fatal. Non-melanoma or non-cancerous types are further classified into Basal cell carcinoma, Squamous cell carcinoma, Merkel cell carcinoma, Skin adnexal tumors, Cutaneous lymphoma, Kaposi sarcoma, and numerous other sarcoma types. These types of cancers are more prevalent but are usually benign, having a low inclination to spread, and can be cured easily as compared to melanoma cancers4. Hence, early detection and diagnosis are crucial to prevent the spread of skin cancer and initiate treatment immediately.  \nTraditionally, the biopsy procedure has been the gold standard for diagnosing skin cancer. It relies on dermatologists and their histopathological examinations, which are time-consuming and laborious processes. Alternatively, rule-based procedures such as ABCDE (asymmetry, borders, colors, diameter, evolution of lesions) and a 7-point checklist ","cbCailatos2V5BlH","https://ap.wps.com/l/cbCailatos2V5BlH","pdf",3428097,21,"English","# Overview\n## Skin cancer background and challenges\n## Proposed dual-architecture framework\n# Classification approach\n## GlobalSkinNet design and training\n## Benchmark datasets and results\n# Segmentation approach\n## SkinFormNet design and evaluation\n## SegFormer and attention-based U-Net","[{\"question\":\"What challenges limit current deep learning skin cancer detection?\",\"answer\":\"Models can struggle with nonhomogeneous luminance, low contrast, and hair-on-skin images, which affect accurate lesion detection.\"},{\"question\":\"How does the proposed framework perform skin lesion classification?\",\"answer\":\"It uses a transformer-based dual architecture where GlobalSkinNet combines convolutional and transformer components, trained with selected parameters and the Adam optimizer.\"},{\"question\":\"How is skin lesion segmentation carried out in the proposed method?\",\"answer\":\"After classification, SkinFormNet processes the images: it extracts features using pre-trained SegFormer and feeds them into an attention-based U-Net for segmentation.\"}]","Advanced hybrid transformer CNN framework for improved skin lesion classification and segmentation | PDF",1790094679,53,[148,152,156,160],{"page_no":22,"url":149,"width":150,"height":151},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/9d620bf0f32dad8f07bc797d203f71f8.webp",983,1291,{"page_no":26,"url":153,"width":154,"height":155},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/9f70ad1e70a60844f9c62096d599a03b.webp",982,1290,{"page_no":30,"url":157,"width":158,"height":159},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/ad34aa6aa237b87fb93a7ff984c53b36.webp",981,1289,{"page_no":33,"url":161,"width":162,"height":163},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/7d76ae15c177df483d3f8806fd7bb6ff.webp",980,1287]