{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T11:40:35Z","timestamp":1786534835408,"version":"build-2736575974"},"reference-count":44,"publisher":"MDPI AG","issue":"16","license":[{"start":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T00:00:00Z","timestamp":1786492800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004410","name":"Scientific and Technological Research Council of T\u00fcrkiye","doi-asserted-by":"publisher","award":["122E629"],"award-info":[{"award-number":["122E629"]}],"id":[{"id":"10.13039\/501100004410","id-type":"DOI","asserted-by":"publisher"}]},{"award":["122E629"],"award-info":[{"award-number":["122E629"]}],"id":[{"id":"https:\/\/ror.org\/04w9kkr77","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: Skin lesion classification remains challenging in real clinical settings because diagnostically relevant information is distributed across different imaging modalities, class imbalance is often severe, and overall performance alone may conceal important weaknesses in model behavior. In this study, a dual-view meta-aware model, termed DVM-SLC was developed for multi-class skin lesion classification on the MILK10k dataset. Methods: The model jointly processes paired clinical and dermoscopic images and incorporates patient-level contextual information through a dedicated metadata branch. Cross-view information is integrated by a gated fusion mechanism to preserve modality-specific representations while allowing adaptive interaction between the two views. The model was evaluated not only in terms of comparative classification performance, but also with respect to calibration, robustness under common image corruptions, and expert-supported explainability. Results: Among the configurations evaluated using pooled out-of-fold predictions, the gated DVM-SLC achieved the highest Accuracy (0.7002) and the highest observed Macro-AUC (0.8774), whereas the gate-free dual-view model with metadata achieved the highest Macro F1-score (0.4174). The accuracy difference between the gated model and the dermoscopic-only baseline, which provided the highest non-proposed Accuracy, was statistically significant, whereas the observed Macro-AUC difference was not statistically significant. These findings indicated a metric-dependent trade-off rather than uniform superiority of a single fusion strategy. The raw confidence estimates were already reasonably aligned with empirical correctness, and temperature scaling provided only a modest, fold-dependent improvement in aggregate calibration. In robustness analysis, the strongest degradation was observed under Gaussian noise, while the model remained comparatively stable under blur, JPEG compression, brightness variation, contrast variation, and color shift. The explainability analysis was complemented by a structured dermatologist review. Gradient-weighted Class Activation Mapping (Grad-CAM) outputs from the clinical and dermoscopic branches were examined to determine whether the highlighted regions localized the lesion and corresponded to clinically meaningful morphological or dermoscopic features. This assessment was qualitative and was not intended as quantitative localization validation. Conclusions: Overall, the findings present DVM-SLC as a broadly evaluated approach with metric-dependent strengths and clear limitations, particularly in minority-class recognition. Its contribution lies in combining comparative performance analysis with calibration, robustness, and structured qualitative explainability assessment.<\/jats:p>","DOI":"10.3390\/diagnostics16162543","type":"journal-article","created":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T11:02:11Z","timestamp":1786532531000},"page":"2543","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["DVM-SLC: A Dual-View Meta-Aware Model for Reliable Multi-Class Skin Lesion Classification from Clinical and Dermoscopic Images"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1999-2797","authenticated-orcid":false,"given":"G\u00f6k\u00e7en","family":"\u00c7etinel","sequence":"first","affiliation":[{"name":"Electrical and Electronics Engineering Department, Engineering Faculty, Sakarya University, 54050 Serdivan, T\u00fcrkiye"},{"name":"Sakarya University Technology Development Zones Management Inc. (Sakarya Teknokent), Sakarya University Esentepe Campus, 54050 Serdivan, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sevda","family":"G\u00fcl","sequence":"additional","affiliation":[{"name":"Department of Electronics and Automation, Adapazar\u0131 Vocational School, Sakarya University, 54050 Serdivan, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rabia","family":"\u00d6zta\u015f Kara","sequence":"additional","affiliation":[{"name":"Department of Dermatology, Sakarya University Training and Research Hospital, 54050 Serdivan, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,12]]},"reference":[{"key":"ref_1","unstructured":"World Health Organization (2025, December 16). Cancer. Available online: https:\/\/www.who.int\/news-room\/fact-sheets\/detail\/cancer."},{"key":"ref_2","unstructured":"MILK Study Team (2025). MILK10k. ISIC Arch."},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., and Batra, D. (2017, January 22\u201329). Grad-CAM: Visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy.","DOI":"10.1109\/ICCV.2017.74"},{"key":"ref_4","unstructured":"Lundberg, S.M., and Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems (NeurIPS), Curran Associates, Inc."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"12531","DOI":"10.1007\/s11042-024-19489-x","article-title":"SKINC-NET: An efficient lightweight deep learning model for multiclass skin lesion classification in dermoscopic images","volume":"84","author":"Asif","year":"2024","journal-title":"Multimed. Tools Appl."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"12799","DOI":"10.1007\/s10586-024-04540-1","article-title":"An effective multiclass skin cancer classification approach based on deep convolutional neural network","volume":"27","author":"Houssein","year":"2024","journal-title":"Clust. Comput."},{"key":"ref_7","first-page":"e88711","article-title":"Multiclass classification of pigmented skin lesions using a multimodal large language model","volume":"17","author":"Iinuma","year":"2025","journal-title":"Cureus"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"6268","DOI":"10.1038\/s41598-025-90423-3","article-title":"A fuzzy rank-based deep ensemble methodology for multi-class skin cancer classification","volume":"15","author":"Halder","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Mustahid, A.A.M., Rahman, M.S., Joy, M.I.K., Ishrak, M.F., Ahmed, N., and Muzahid, A.A.M. (2025, January 20\u201322). DenseNet-driven multi-class classification of skin lesions. Proceedings of the 17th International Conference on Computer and Automation Engineering (ICCAE), Perth, Australia.","DOI":"10.1109\/ICCAE64891.2025.10980560"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"4938","DOI":"10.1038\/s41598-025-89230-7","article-title":"A robust deep learning framework for multiclass skin cancer classification","volume":"15","author":"Ozdemir","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_11","first-page":"122","article-title":"CNN architectural models for multiclass classification of aesthetic facial skin disorders","volume":"12","author":"Rismayani","year":"2025","journal-title":"Int. J. Adv. Technol. Eng. Explor."},{"key":"ref_12","first-page":"294","article-title":"Deep CNN model for multi-class skin cancer classification","volume":"34","author":"Vemulamma","year":"2025","journal-title":"J. Comput. Anal. Appl."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"101843","DOI":"10.1016\/j.compmedimag.2020.101843","article-title":"Automated multi-class classification of skin lesions through deep convolutional neural network with dermoscopic images","volume":"88","author":"Iqbal","year":"2021","journal-title":"Comput. Med. Imaging Graph."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"773","DOI":"10.1007\/s11517-023-02969-x","article-title":"An efficient multi-class classification of skin cancer using optimized vision transformer","volume":"62","author":"Desale","year":"2024","journal-title":"Med. Biol. Eng. Comput."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"733","DOI":"10.1007\/s42979-024-03072-0","article-title":"Skin cancer detection and multi-class classification using deep CNN model with estimated disease probabilities","volume":"5","author":"Kumar","year":"2024","journal-title":"SN Comput. Sci."},{"key":"ref_16","first-page":"63","article-title":"A deep learning framework for multi-class skin cancer diagnosis","volume":"13","author":"Haque","year":"2025","journal-title":"Eur. J. Comput. Sci. Inf. Technol."},{"key":"ref_17","first-page":"12531","article-title":"SkinDWNet: A novel deep learning model for multiclass skin cancer classification","volume":"84","author":"Naeem","year":"2025","journal-title":"Multimed. Syst."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Alruwaili, M., and Mohamed, M. (2025). An integrated deep learning model with EfficientNet and ResNet for accurate multi-class skin disease classification. Diagnostics, 15.","DOI":"10.3390\/diagnostics15050551"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1776","DOI":"10.1007\/s10278-024-01300-w","article-title":"A multi-model deep learning architecture for diagnosing multi-class skin diseases","volume":"38","author":"Badr","year":"2024","journal-title":"J. Imaging Inform. Med."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"32505","DOI":"10.1007\/s11042-024-20503-5","article-title":"Custom CNN architectures for skin disease classification: Binary and multi-class performance","volume":"84","author":"Gupta","year":"2025","journal-title":"Multimed. Tools Appl."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"87105","DOI":"10.1007\/s11042-024-19817-1","article-title":"A deep learning-based architecture for multi-class skin cancer classification","volume":"83","author":"Mushtaq","year":"2024","journal-title":"Multimed. Tools Appl."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"e70102","DOI":"10.1002\/ima.70102","article-title":"JILDYA-Net: An efficient lightweight multi-class classification architecture for skin lesions","volume":"35","author":"Laouarem","year":"2025","journal-title":"Int. J. Imaging Syst. Technol."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Priyadarshini, P.S.M., and Khan, N.Z.A. (2024, January 26\u201327). Classification of multi-class skin cancer using deep learning algorithm. Proceedings of the ICDCECE, Ballari, India.","DOI":"10.1109\/ICDCECE60827.2024.10549142"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"113677","DOI":"10.1109\/ACCESS.2024.3432904","article-title":"Early detection of multiclass skin lesions using transfer learning-based IncepX-Ensemble model","volume":"12","author":"Chatterjee","year":"2024","journal-title":"IEEE Access"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Chiu, T.-M., Chi, I.C., Li, Y.C., and Tseng, M.H. (2025). Deep ensemble learning for multiclass skin lesion classification. Bioengineering, 12.","DOI":"10.3390\/bioengineering12090934"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Alshahrani, M., Al-Jabbar, M., Senan, E.M., Ahmed, I.A., and Mohammed Saif, J.A. (2024). Analysis of dermoscopy images of multi-class for early detection of skin lesions by hybrid systems based on integrating features of CNN models. PLoS ONE, 19.","DOI":"10.1371\/journal.pone.0298305"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"e23180","DOI":"10.1002\/ima.23180","article-title":"A hybrid CNN model for the classification of multi-class skin cancer","volume":"34","author":"Toprak","year":"2024","journal-title":"Int. J. Imaging Syst. Technol."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"143029","DOI":"10.1109\/ACCESS.2024.3468612","article-title":"EFAM-Net: A multi-class skin lesion classification model utilizing enhanced feature fusion and attention mechanisms","volume":"12","author":"Ji","year":"2024","journal-title":"IEEE Access"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"260","DOI":"10.1007\/s42979-025-03806-8","article-title":"Multi-class classification of class imbalanced skin lesion dataset using modified SMOTE-ENN","volume":"6","author":"Sahoo","year":"2025","journal-title":"SN Comput. Sci."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"16498","DOI":"10.1109\/ACCESS.2024.3360215","article-title":"A GAN-based data augmentation method for imbalanced multi-class skin lesion classification","volume":"12","author":"Su","year":"2024","journal-title":"IEEE Access"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Tang, K., Su, J., Chen, R., Huang, R., Dai, M., and Li, Y.S. (2024). SkinSwinViT: A lightweight transformer-based method for multiclass skin lesion classification. Appl. Sci., 14.","DOI":"10.3390\/app14104005"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"102391","DOI":"10.1016\/j.eclinm.2023.102391","article-title":"A deep learning fusion network trained with clinical and high-frequency ultrasound images","volume":"67","author":"Zhu","year":"2024","journal-title":"eClinicalMedicine"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"102307","DOI":"10.1016\/j.media.2021.102307","article-title":"FusionM4Net: A Multi-Stage Multi-Modal Learning Algorithm for Multi-Label Skin Lesion Classification","volume":"76","author":"Tang","year":"2022","journal-title":"Med. Image Anal."},{"key":"ref_34","unstructured":"Cheslerean-Boghiu, T., Fleischmann, M.-E., Willem, T., and Lasser, T. (2023). Transformer-Based Interpretable Multi-Modal Data Fusion for Skin Lesion Classification. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"108549","DOI":"10.1016\/j.compbiomed.2024.108549","article-title":"A Novel Multi-Task Learning Network for Skin Lesion Classification Based on Multi-Modal Clues and Label-Level Fusion","volume":"175","author":"Lin","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"ref_36","unstructured":"Zhang, Y., Xie, Y., Wang, H., Avery, J.C., Hull, M.L., and Carneiro, G. (March, January 26). A Novel Perspective for Multi-Modal Multi-Label Skin Lesion Classification. Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision, Tucson, AZ, USA."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"108729","DOI":"10.1016\/j.cmpb.2025.108729","article-title":"Self-Supervised Multi-Modality Learning for Multi-Label Skin Lesion Classification","volume":"265","author":"Wang","year":"2025","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_38","doi-asserted-by":"crossref","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":"Le","year":"2025","journal-title":"Comput. Med. Imaging Graph."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"103091","DOI":"10.1016\/j.artmed.2025.103091","article-title":"A Multi-Stage Multi-Modal Learning Algorithm with Adaptive Multimodal Fusion for Improving Multi-Label Skin Lesion Classification","volume":"162","author":"Zuo","year":"2025","journal-title":"Artif. Intell. Med."},{"key":"ref_40","unstructured":"Sellergren, A., Kazemzadeh, S., Jaroensri, T., Kiraly, A., Traverse, M., Kohlberger, T., Xu, S., Jamil, F., Hughes, C., and Lau, C. (2025). MedGemma Technical Report. arXiv."},{"key":"ref_41","unstructured":"Tan, M., and Le, Q.V. (2019, January 9\u201315). EfficientNet: Rethinking model scaling for convolutional neural networks. Proceedings of the 36th International Conference on Machine Learning (ICML), Long Beach, CA, USA."},{"key":"ref_42","unstructured":"Loshchilov, I., and Hutter, F. (2019, January 6\u20139). Decoupled weight decay regularization. Proceedings of the 7th International Conference on Learning Representations (ICLR), New Orleans, LA, USA."},{"key":"ref_43","unstructured":"Loshchilov, I., and Hutter, F. (2017, January 24\u201326). SGDR: Stochastic gradient descent with warm restarts. Proceedings of the 5th International Conference on Learning Representations (ICLR), Toulon, France."},{"key":"ref_44","unstructured":"Guo, C., Pleiss, G., Sun, Y., and Weinberger, K.Q. (2017, January 6\u201311). On calibration of modern neural networks. Proceedings of the 34th International Conference on Machine Learning (ICML), Sydney, Australia."}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/16\/2543\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T11:17:04Z","timestamp":1786533424000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/16\/2543"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,12]]},"references-count":44,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2026,8]]}},"alternative-id":["diagnostics16162543"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16162543","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,12]]}}}