{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T06:55:52Z","timestamp":1790751352830,"version":"4.1.0"},"reference-count":25,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T00:00:00Z","timestamp":1790726400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Well trained segmentation and classification of skin lesions from dermoscopic images are crucial to an early detection of melanoma and other malignant skin disease. While many deep learning models reach impressive overall performance, most existing approaches treat segmentation and classification as separate tasks, leading to limited interaction between spatial lesion localization and semantic disease knowledge. This segregation results in a high variance prediction and would reduce the overall robustness of automated diagnostic systems. To mitigate this limitation, we proposed SkinNet-X: a unified multitask framework to reformulate the skin lesion segmentation and classification tasks jointly. We achieve this with a shared hierarchical feature representation to transfer knowledge between both tasks while introducing new Cross Task Consistency Learning (CTCL) strategy that enforces semantic consistency between lesion localization and disease prediction during joint optimization. In order to enhance the discriminative capability of feature expression and alleviate redundant representations, a dual stage multi label feature optimization module is further exploited, accompanied by an adaptive PolyLoss based optimization strategy stabilizing multitask learning. The effectiveness of the proposed SkinNet-X framework was evaluated using segmentation and classification experiments on the ISIC 2018, ISIC 2019, and HAM10000 benchmark datasets. We also performed additional analyses, namely ablation studies, computational complexity evaluation, sensitivity analysis and statistical validation to explore the contribution of each architectural component. Experimental results show that the proposed multitask learning strategy can consistently improve accuracy of lesion localization and performance of classification with robust generalizability across multiple datasets. These findings demonstrate that SkinNet-X is an effective and potentially useful framework for automated skin lesion analysis and computer-aided dermatological diagnosis.<\/jats:p>","DOI":"10.3389\/frai.2026.1886221","type":"journal-article","created":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T06:03:41Z","timestamp":1790748221000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["SkinNet-X: a task coupled feature interaction framework for joint skin lesion segmentation and classification"],"prefix":"10.3389","volume":"9","author":[{"given":"E.","family":"Babu","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering (SCOPE), Vellore Institute of Technology","place":["Vellore, Tamil Nadu, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"S.","family":"Murali","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering (SCOPE), Vellore Institute of Technology","place":["Vellore, Tamil Nadu, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,30]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"575","DOI":"10.1016\/j.jare.2025.08.039","article-title":"Skin cancer segmentation and recognition from dermoscopy images: a novel framework based on improved DeepLabV3+ and network-level fused deep architectures","volume":"83","author":"Arshad","year":"2025","journal-title":"J. Adv. Res"},{"key":"B2","doi-asserted-by":"publisher","first-page":"127721","DOI":"10.1016\/j.eswa.2025.127721","article-title":"A novel hybrid ConvNeXt based approach for enhanced skin lesion classification","volume":"283","author":"Aruk","year":"2025","journal-title":"Expert Syst. Appl."},{"key":"B3","doi-asserted-by":"publisher","first-page":"40690","DOI":"10.1038\/s41598-025-23830-1","article-title":"AttenUNeT X with iterative feedback mechanisms for robust deep learning skin lesion segmentation","volume":"15","author":"Babu","year":"2025","journal-title":"Sci. Rep."},{"key":"B4","first-page":"1","article-title":"Mentored learning: improving generalization and convergence of student learner","volume":"25","author":"Cao","year":"2024","journal-title":"J. Mach. Learn. Res."},{"key":"B5","doi-asserted-by":"publisher","first-page":"110352","DOI":"10.1016\/j.compbiomed.2025.110352","article-title":"Deep pixel wise supervision for skin lesion classification","volume":"193","author":"Dzieniszewska","year":"2025","journal-title":"Comput. Biol. Med"},{"key":"B6","doi-asserted-by":"publisher","first-page":"100546","DOI":"10.1016\/j.fraope.2026.100546","article-title":"Tri attention boosted scalable EfficientNet for skin lesion classification via triple stage gain ratioed averaging","volume":"15","author":"Efat","year":"2026","journal-title":"Franklin Open"},{"key":"B7","doi-asserted-by":"publisher","first-page":"107449","DOI":"10.1016\/j.bspc.2024.107449","article-title":"Multi residual attention network for skin lesion classification","volume":"103","author":"Ghazouani","year":"2025","journal-title":"Biomed. Signal Process. Control"},{"key":"B8","doi-asserted-by":"publisher","first-page":"20552076261462026","DOI":"10.1177\/20552076261462026","article-title":"Enhancing skin lesion classification using a Tri-Path attention stacked ensemble architecture with Cohen's Kappa proportioned averaging","volume":"12","author":"Hasan","year":"2026","journal-title":"Digit. Health"},{"key":"B9","doi-asserted-by":"publisher","first-page":"108594","DOI":"10.1016\/j.compbiomed.2024.108594","article-title":"Multi scale feature fusion and class weight loss for skin lesion classification","volume":"170","author":"Hu","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"B10","doi-asserted-by":"publisher","first-page":"100706","DOI":"10.1016\/j.eij.2025.100706","article-title":"Advanced skin lesion segmentation and classification using adaptive contextual GLCM and deep learning hybrid models","volume":"30","author":"Kumar","year":"2025","journal-title":"Egypt. Inf. J."},{"key":"B11","doi-asserted-by":"publisher","first-page":"107819","DOI":"10.1016\/j.bspc.2025.107819","article-title":"White patchy skin lesion classification using feature enhancement and interaction transformer module","volume":"107","author":"Li","year":"2025","journal-title":"Biomed. Signal Process. Control"},{"key":"B12","doi-asserted-by":"publisher","first-page":"104389","DOI":"10.1016\/j.cviu.2025.104389","article-title":"Multi-scale feature fusion based SAM for high-quality few-shot medical image segmentation","volume":"258","author":"Liu","year":"2025","journal-title":"Comput. Vis. Image Underst."},{"key":"B13","doi-asserted-by":"publisher","first-page":"37823","DOI":"10.1038\/s41598-025-21783-z","article-title":"A deep learning based dual branch framework for automated skin lesion segmentation and classification via dermoscopic images","volume":"15","author":"Owida","year":"2025","journal-title":"Sci. Rep."},{"key":"B14","doi-asserted-by":"publisher","first-page":"104201","DOI":"10.1016\/j.rineng.2025.104201","article-title":"Temporal integration of ResNet features with LSTM for enhanced skin lesion classification","volume":"25","author":"Padhy","year":"2025","journal-title":"Results Eng."},{"key":"B15","doi-asserted-by":"publisher","first-page":"474","DOI":"10.1049\/cit2.12410","article-title":"Advancing skin cancer detection integrating a novel unsupervised classification and enhanced imaging techniques","volume":"10","author":"Rahman","year":"2025","journal-title":"CAAI Trans. Intell. Technol."},{"key":"B16","doi-asserted-by":"publisher","first-page":"1808770","DOI":"10.3389\/frai.2026.1808770","article-title":"Improved CNN with BiLSTM model for early melanoma and skin lesion classification","volume":"9","author":"Rajeshkumar","year":"2026","journal-title":"Front. Artif. Intell."},{"key":"B17","doi-asserted-by":"publisher","first-page":"106380","DOI":"10.1016\/j.bspc.2024.106380","article-title":"A blockchain based deep learning system with optimization for skin disease classification","volume":"95","author":"Rokade","year":"2024","journal-title":"Biomed. Signal Process. Control"},{"key":"B18","doi-asserted-by":"publisher","first-page":"109081","DOI":"10.1016\/j.cmpb.2025.109081","article-title":"MultiExCam: a multi approach and explainable artificial intelligence architecture for skin lesion classification","volume":"273","author":"Ruga","year":"2026","journal-title":"Comput. Methods Programs Biomed."},{"key":"B19","doi-asserted-by":"publisher","first-page":"100844","DOI":"10.1016\/j.eij.2025.100844","article-title":"Hair removal and lesion segmentation of dermoscopic images for classification of skin cancer using deep neural networks","volume":"32","author":"Shahzaib","year":"2025","journal-title":"Egypt. Inf. J."},{"key":"B20","doi-asserted-by":"publisher","first-page":"15680","DOI":"10.1038\/s41598-025-00025-2","article-title":"An attention based hybrid approach using CNN and BiLSTM for improved skin lesion classification","volume":"15","author":"Shaik","year":"2025","journal-title":"Sci. Rep."},{"key":"B21","doi-asserted-by":"publisher","first-page":"100326","DOI":"10.1016\/j.health.2024.100326","article-title":"A hybrid Grasshopper optimization algorithm for skin lesion segmentation and melanoma classification using deep learning","volume":"5","author":"Thapar","year":"2024","journal-title":"Healthc. Anal."},{"key":"B22","doi-asserted-by":"publisher","first-page":"102588","DOI":"10.1016\/j.compmedimag.2025.102588","article-title":"A multimodal skin lesion classification through cross attention fusion and collaborative edge computing","volume":"124","author":"Tran-Van","year":"2025","journal-title":"Comput. Med. Imaging Graph."},{"key":"B23","doi-asserted-by":"publisher","first-page":"41150","DOI":"10.1038\/s41598-025-25480-9","article-title":"Effective skin cancer classification by modified and optimized Inception ResNet V2 model","volume":"15","author":"Umapathi","year":"2025","journal-title":"Sci. Rep."},{"key":"B24","doi-asserted-by":"publisher","first-page":"e70517","DOI":"10.1002\/eng2.70517","article-title":"Hidden feature extraction based on Mask R-CNN and transfer learning and its application in medical diagnosis","volume":"7","author":"Yang","year":"2025","journal-title":"Eng. Rep"},{"key":"B25","doi-asserted-by":"publisher","first-page":"364","DOI":"10.3390\/math11020364","article-title":"DeepLabv3+ based segmentation and best features selection using slime mould algorithm for multi class skin lesion classification","volume":"11","author":"Zafar","year":"2023","journal-title":"Mathematics"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1886221\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T06:03:41Z","timestamp":1790748221000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1886221\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,30]]},"references-count":25,"alternative-id":["10.3389\/frai.2026.1886221"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1886221","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,30]]},"article-number":"1886221"}}