基于深度学习的皮肤病分类研究进展

doi:10.3778/j.issn.1002-8331.2410-0450

摘要/Abstract

摘要： 皮肤是人体最大的器官，因参与多种生理活动导致发病率较高。皮肤病的准确诊断对患者健康至关重要，然而传统诊断方法却存在主观性强和效率低下等问题。近年来，深度学习技术的迅猛发展为皮肤病分类提供了新的解决方案。系统性地回顾了基于深度学习的皮肤病分类研究进展，总结了近年来皮肤病诊断常用数据集，进而深入探讨了卷积神经网络（convolutional neural network，CNN）和Transformer等主流算法在皮肤病分类中的应用，并特别关注了综合算法的研究进展，此外还进行了典型方法的性能对比分析，总结了当前研究方法存在的各种挑战以及未来研究方向，为后续的发展提供启示和指导。

关键词: 皮肤病分类, 深度学习, 医学图像分析, 模型对比, 智能诊断, 皮肤图像数据集

Abstract: The skin is the largest organ of the human body and has a high incidence rate due to its involvement in various physiological activities. Accurate diagnosis of skin diseases is crucial for patient health, however, traditional diagnostic methods suffer from subjectivity and inefficiency. In recent years, the rapid development of deep learning technologies has provided new solutions for skin disease classification. This paper systematically reviews the progress of deep learning-based skin disease classification research. It summarizes the commonly used datasets for skin disease diagnosis in recent years, then delves into the applications of mainstream algorithms such as convolutional neural networks (CNN) and Transformers in skin disease classification, with a particular focus on the research progress of hybrid algorithms. Additionally, a performance comparison of typical methods is presented. Finally, the paper summarizes the challenges in current research methods and outlines future research directions, providing insights and guidance for subsequent developments.

Key words: skin disease classification, deep learning, medical image analysis, model comparison, intelligent diagnosis, skin image datasets

韩佩宏, 魏德健, 李传莉, 姜良, 曹慧. 基于深度学习的皮肤病分类研究进展[J]. 计算机工程与应用, 2025, 61(18): 61-77.

HAN Peihong, WEI Dejian, LI Chuanli, JIANG Liang, CAO Hui. Research Progress on Skin Disease Classification Based on Deep Learning[J]. Computer Engineering and Applications, 2025, 61(18): 61-77.

参考文献

[1] HAMEED N, RUSKIN A, ABU HASSAN K, et al. A comprehensive survey on image-based computer aided diagnosis systems for skin cancer[C]//Proceedings of the 2016 10th International Conference on Software, Knowledge, Information Management & Applications. Piscataway: IEEE, 2016: 205-214.
[2] HAMEED N, SHABUT A, HOSSAIN M A. A computer-aided diagnosis system for classifying prominent skin lesions using machine learning[C]//Proceedings of the 2018 10th Computer Science and Electronic Engineering. Piscataway: IEEE, 2018: 186-191.
[3] TSCHANDL P, CODELLA N, AKAY B N, et al. Comparison of the accuracy of human readers versus machine-learning algorithms for pigmented skin lesion classification: an open, web-based, international, diagnostic study[J]. The Lancet Oncology, 2019, 20(7): 938-947.
[4] RODRIGUES M, EZZEDINE K, HAMZAVI I, et al. New discoveries in the pathogenesis and classification of vitiligo[J]. Journal of the American Academy of Dermatology, 2017, 77(1): 1-13.
[5] SINGH M, KOTNIS A, JADEJA S D, et al. Cytokines: the Yin and Yang of vitiligo pathogenesis[J]. Expert Review of Clinical Immunology, 2019, 15(2): 177-188.
[6] BURLANDO M, MURACCHIOLI A, COZZANI E, et al. Psoriasis, vitiligo, and biologic therapy: case report and narrative review[J]. Case Reports in Dermatology, 2021, 13(2): 372-378.
[7] LUCIERI A, BAJWA M N, BRAUN S A, et al. ExAID: a multimodal explanation framework for computer-aided diagnosis of skin lesions[J]. Computer Methods and Programs in Biomedicine, 2022, 215: 106620.
[8] WAN L, CHEN J B, WU H, et al. Deep learning for inflammatory diseases classification based on reflectance confocal microscopy[J]. Journal of the American Academy of Dermatology, 2023, 88(5): e283-e284.
[9] SARGEN M R, LUK K M, STOFF B K, et al. Diagnostic accuracy of whole slide imaging for cutaneous, soft tissue, and melanoma sentinel lymph node biopsies with and without immunohistochemistry[J]. Journal of Cutaneous Pathology, 2018, 45(8): 597-602.
[10] YANG J C, JIAO Y B, XIONG N X, et al. Fast face gender recognition by using local ternary pattern and extreme learning machine[J]. Ksii Transactions on Internet and Information Systems, 2013, 7(7): 1705-1720.
[11] 王慧, 戚倩倩, 李雪, 等. 皮肤肿瘤图像自动分类的研究进展[J]. 计算机工程与应用, 2022, 58(16): 31-48.
WANG H, QI Q Q, LI X, et al. Research progress in automatic classification of skin lesions image[J]. Computer Engineering and Applications, 2022, 58(16): 31-48.
[12] ZHANG B, ZHOU X, LUO Y C, et al. Opportunities and challenges: classification of skin disease based on deep learning[J]. Chinese Journal of Mechanical Engineering, 2021, 34(1): 112.
[13] NANCY V A O, PRABHAVATHY P, ARYA M S. Role of artificial intelligence and deep learning in skin disease prediction: a systematic review and meta-analysis[J]. Annals of Data Science, 2024, 11(6): 2109-2139.
[14] LI H F, PAN Y N, ZHAO J, et al. Skin disease diagnosis with deep learning: a review[J]. Neurocomputing, 2021, 464: 364-393.
[15] LYAKHOVA U A, LYAKHOV P A. Systematic review of approaches to detection and classification of skin cancer using artificial intelligence: development and prospects[J]. Computers in Biology and Medicine, 2024, 178: 108742.
[16] RIAZ S, NAEEM A, MALIK H, et al. Federated and transfer learning methods for the classification of melanoma and nonmelanoma skin cancers: a prospective study[J]. Sensors, 2023, 23(20): 8457.
[17] RAY A, SARKAR S, SCHWENKER F, et al. Decoding skin cancer classification: perspectives, insights, and advances through researchers’lens[J]. Scientific Reports, 2024, 14(1): 30542.
[18] GUTMAN D, CODELLA N C F, CELEBI E, et al. Skin lesion analysis toward melanoma detection: a challenge at the international symposium on biomedical imaging (ISBI) 2016, hosted by the international skin imaging collaboration (ISIC)[J]. arXiv:1605.01397, 2016.
[19] CODELLA N C F, GUTMAN D, CELEBI M E, et al. Skin lesion analysis toward melanoma detection: a challenge at the 2017 International symposium on biomedical imaging (ISBI), hosted by the international skin imaging collaboration (ISIC)[C]//Proceedings of the 2018 IEEE 15th International Symposium on Biomedical Imaging. Piscataway: IEEE, 2018: 168-172.
[20] TSCHANDL P, ROSENDAHL C, KITTLER H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions[J]. Scientific Data, 2018, 5: 180161.
[21] COMBALIA M, CODELLA N C F, ROTEMBERG V, et al. BCN20000: dermoscopic lesions in the wild[J]. arXiv:1908.
02288, 2019.
[22] ROTEMBERG V, KURTANSKY N, BETZ-STABLEIN B, et al. A patient-centric dataset of images and metadata for identifying melanomas using clinical context[J]. Scientific Data, 2021, 8: 34.
[23] BAJWA M N, MUTA K, MALIK M I, et al. Computer-aided diagnosis of skin diseases using deep neural networks[J]. Applied Sciences, 2020, 10(7): 2488.
[24] BALLERINI L, FISHER R B, ALDRIDGE B, et al. A color and texture based hierarchical K-NN approach to the classification of non-melanoma skin lesions[M]//Color medical image analysis. Dordrecht: Springer Netherlands, 2013: 63-86.
[25] GIOTIS I, MOLDERS N, LAND S, et al. MED-NODE: a computer-assisted melanoma diagnosis system using non-dermoscopic images[J]. Expert Systems with Applications, 2015, 42(19): 6578-6585.
[26] MENDON?A T, FERREIRA P M, MARQUES J S, et al. PH2-A dermoscopic image database for research and benchmarking[C]//Proceedings of the 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway: IEEE, 2013: 5437-5440.
[27] KAWAHARA J, DANESHVAR S, ARGENZIANO G, et al. Seven-point checklist and skin lesion classification using multitask multimodal neural nets[J]. IEEE Journal of Biomedical and Health Informatics, 2018, 23(2): 538-546.
[28] HAN S S. SNU dataset+Quiz[DB/OL]. (2021-08-16)[2021-11-03]. https://doi.org/10.6084/m9.figshare.6454973.v11.
[29] HAN S S, KIM M S, LIM W, et al. Classification of the clinical images for benign and malignant cutaneous tumors using a deep learning algorithm[J]. Journal of Investigative Dermatology, 2018, 138(7): 1529-1538.
[30] XIE B, HE X Y, ZHAO S, et al. XiangyaDerm: a clinical image dataset of Asian race for skin disease aided diagnosis[M]//Large-scale annotation of biomedical data and expert label synthesis and hardware aware learning for medical imaging and computer assisted intervention. Cham: Springer International Publishing, 2019: 22-31.
[31] ZHANG J P, ZHONG F, HE K Q, et al. Recent advancements and perspectives in the diagnosis of skin diseases using machine learning and deep learning: a review[J]. Diagnostics, 2023, 13(23): 3506.
[32] 王师玮, 陈俊, 易才键. 基于改进ResNet50的皮肤病变图像分类[J]. 软件工程, 2023, 26(6): 50-54.
WANG S W, CHEN J, YI C J. Image classification of skin lesions based on improved ResNet50[J]. Software Engineering, 2023, 26(6): 50-54.
[33] ESKANDARI A, SHARBATDAR M. Efficient diagnosis of psoriasis and lichen planus cutaneous diseases using deep learning approach[J]. Scientific Reports, 2024, 14: 9715.
[34] KOUSIS I, PERIKOS I, HATZILYGEROUDIS I, et al. Deep learning methods for accurate skin cancer recognition and mobile application[J]. Electronics, 2022, 11(9): 1294.
[35] ALI JINNA MATHINA KANI M, PARVATHY M S, MAAJITHA BANU S, et al. Classification of skin lesion images using modified Inception V3 model with transfer learning and augmentation techniques[J]. Journal of Intelligent & Fuzzy Systems, 2023, 44(3): 4627-4641.
[36] KARTHIK R, VAICHOLE T S, KULKARNI S K, et al. Eff2Net: an efficient channel attention-based convolutional neural network for skin disease classification[J]. Biomedical Signal Processing and Control, 2022, 73: 103406.
[37] VENUGOPAL V, RAJ N I, NATH M K, et al. A deep neural network using modified EfficientNet for skin cancer detection in dermoscopic images[J]. Decision Analytics Journal, 2023, 8: 100278.
[38] MADURANGA M, NANDASENA D. Mobile-based skin disease diagnosis system using convolutional neural networks (CNN)[J]. International Journal of Image, Graphics and Signal Processing, 2022, 14(3): 47-57.
[39] SAHIN V H, OZTEL I, YOLCU OZTEL G. Human monkeypox classification from skin lesion images with deep pre-trained network using mobile application[J]. Journal of Medical Systems, 2022, 46(11): 79.
[40] KUMAR L, SINGH K U, SHARMA B K, et al. Deep learning and crow search to detect skin cancer in dermoscopic images[C]//Proceedings of the 2024 IEEE International Conference on Contemporary Computing and Communications. Piscataway: IEEE, 2024: 1-7.
[41] RADHIKA V, CHANDANA B S. MSCDNet-based multi-class classification of skin cancer using dermoscopy images[J]. PeerJ Computer Science, 2023, 9: e1520.
[42] ALI M S, MIAH M S, HAQUE J, et al. An enhanced technique of skin cancer classification using deep convolutional neural network with transfer learning models[J]. Machine Learning with Applications, 2021, 5: 100036.
[43] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems, 2017.
[44] ARKIN E, YADIKAR N, MUHTAR Y, et al. A survey of object detection based on CNN and Transformer[C]//Proceedings of the 2021 IEEE 2nd International Conference on Pattern Recognition and Machine Learning. Piscataway: IEEE, 2021: 99-108.
[45] HIMEL G M S, ISLAM M M, AL-AFF K A, et al. Skin cancer segmentation and classification using vision Transformer for automatic analysis in dermatoscopy-based noninvasive digital system[J]. International Journal of Biomedical Imaging, 2024, 2024(1): 3022192.
[46] XIN C, LIU Z F, ZHAO K Y, et al. An improved transformer network for skin cancer classification[J]. Computers in Biology and Medicine, 2022, 149: 105939.
[47] FLOSDORF C, ENGELKER J, KELLER I, et al. Skin cancer detection utilizing deep learning: classification of skin lesion images using a vision Transformer[J]. arXiv:2407. 18554, 2024.
[48] AHMAD I, AMIN J, IKRAMULLAH LALI M, et al. A novel Deeplabv3+ and vision-based transformer model for segmentation and classification of skin lesions[J]. Biomedical Signal Processing and Control, 2024, 92: 106084.
[49] REN Y G, XU W Q, MAO Y X, et al. Few-shot learning for dermatological conditions with lesion area aware swin Transformer[J]. International Journal of Imaging Systems and Technology, 2023, 33(5): 1549-1560.
[50] TANG K, SU J, CHEN R H, et al. SkinSwinViT: a lightweight Transformer-based method for multiclass skin lesion classification with enhanced generalization capabilities[J]. Applied Sciences, 2024, 14(10): 4005.
[51] XIE J S, WU Z Z, ZHU R J, et al. Melanoma detection based on swin Transformer and SimAM[C]//Proceedings of the 2021 IEEE 5th Information Technology, Networking, Electronic and Automation Control Conference. Piscataway: IEEE, 2021: 1517-1521.
[52] SHEHZAD K, TAN Z H, SHOUKAT S, et al. A deep-ensemble-learning-based approach for skin cancer diagnosis[J]. Electronics, 2023, 12(6): 1342.
[53] KRISHNA G S, SUPRIYA K, SORGILE M. LesionAid: vision Transformers-based skin lesion generation and classification[J]. arXiv:2302.01104, 2023.
[54] PACAL I, ALAFTEKIN M, ZENGUL F D. Enhancing skin cancer diagnosis using swin Transformer with hybrid shifted window-based multi-head self-attention and SwiGLU-based MLP[J]. Journal of Imaging Informatics in Medicine, 2024, 37(6): 3174-3192.
[55] ANJALI T, ABHISHEK S, REMYA S. A novel framework for rosacea detection using swin Transformers and explainable artificial intelligence[J]. Alexandria Engineering Journal, 2025, 118: 36-58.
[56] PARADDY S, VIRUPAKSHAPPA. Addressing challenges in skin cancer diagnosis: a convolutional swin Transformer approach[J]. Journal of Imaging Informatics in Medicine, 2025, 38(3): 1755-1775.
[57] DWIVEDI T, CHAURASIA B K, SHUKLA M M. Lightweight vision image transformer (LViT) model for skin cancer disease classification[J]. International Journal of System Assurance Engineering and Management, 2024, 15(10): 5030-5055.
[58] SADIK R, MAJUMDER A, BISWAS A A, et al. An in-depth analysis of convolutional neural network architectures with transfer learning for skin disease diagnosis[J]. Healthcare Analytics, 2023, 3: 100143.
[59] SWETHA R N, SHRIVASTAVA V K, PARVATHI K. Multiclass skin lesion classification using image augmentation technique and transfer learning models[J]. International Journal of Intelligent Unmanned Systems, 2024, 12(2): 220-228.
[60] 赵宸, 帅仁俊, 马力, 等. 基于Self-Attention StyleGAN的皮肤癌图像生成与分类[J]. 计算机工程与应用, 2022, 58(18): 111-121.
ZHAO C, SHUAI R J, MA L, et al. Generation and classification of skin cancer images based on Self-Attention StyleGAN[J]. Computer Engineering and Applications, 2022, 58(18): 111-121.
[61] AHMAD B, SUN J, PALADE V, et al. Improving skin cancer classification using heavy-tailed student T-distribution in generative adversarial networks (TED-GAN)[J]. Diagnostics, 2021, 11(11): 2147.
[62] TEODORO A A M, SILVA D H, ROSA R L, et al. A skin cancer classification approach using GAN and RoI-based attention mechanism[J]. Journal of Signal Processing Systems, 2023, 95(2): 211-224.
[63] ALSAIDI M, JAN M T, ALTAHER A, et al. Tackling the class imbalanced dermoscopic image classification using data augmentation and GAN[J]. Multimedia Tools and Applications, 2024, 83(16): 49121-49147.
[64] SU Q C, HAMED H N A, ISA M A, et al. A GAN-based data augmentation method for imbalanced multi-class skin lesion classification[J]. IEEE Access, 2024, 12: 16498-16513.
[65] BATISTA L G, BUGATTI P H, SAITO P T M. Classification of skin lesion through active learning strategies[J]. Computer Methods and Programs in Biomedicine, 2022, 226: 107122.
[66] DENG Z P, YANG Y Q, SUZUKI K. Federated active learning framework for efficient annotation strategy in skin-lesion classification[J]. Journal of Investigative Dermatology, 2025, 145(2): 303-311.
[67] MANDAL S, GHOSH S, JANA N D, et al. Active learning with particle swarm optimization for enhanced skin cancer classification utilizing deep CNN models[J]. Journal of Imaging Informatics in Medicine, 2024:1-18.
[68] 杨怡光, 王钧程, 谢凤英, 等. 皮肤疾病辅助诊断系统中的数据与方法[J]. 协和医学杂志, 2023, 14(1): 168-176.
YANG Y G, WANG J C, XIE F Y, et al. Data and methods in computer-aided diagnosis systems of skin diseases[J]. Medical Journal of Peking Union Medical College Hospital, 2023, 14(1): 168-176.
[69] TANG P, YAN X T, NAN Y, et al. FusionM4Net: a multi-stage multi-modal learning algorithm for multi-label skin lesion classification[J]. Medical Image Analysis, 2022, 76: 102307.
[70] ZHANG Y J, HU Y F, LI K, et al. Exploring the influence of Transformer-based multimodal modeling on clinicians’diagnosis of skin diseases: a quantitative analysis[J]. Digital Health, 2024, 10: 20552076241257087.
[71] TANG P, YAN X T, NAN Y, et al. Joint-individual fusion structure with fusion attention module for multi-modal skin cancer classification[J]. Pattern Recognition, 2024, 154: 110604.
[72] MOHAMMED H M A, OMEROGLU A N, ORAL E A, et al. ISAFusionNet: involution and soft attention based deep multi-modal fusion network for multi-label skin lesion classification[J]. Computers and Electrical Engineering, 2025, 122: 109966.
[73] ZHOU J X, HE X N, SUN L Y, et al. Pre-trained multimodal large language model enhances dermatological diagnosis using SkinGPT-4[J]. Nature Communications, 2024, 15: 5649.
[74] RAZIA SULTHANA A, CHAMOLA V, HUSSAIN Z, et al. A novel end-to-end deep convolutional neural network based skin lesion classification framework[J]. Expert Systems with Applications, 2024, 246: 123056.
[75] ARUMUGAM R V, SARAVANAN S. Automated multi-class skin cancer classification using white shark optimizer with ensemble learning classifier on dermoscopy images[J]. Multimedia Tools and Applications, 2025, 84(8): 4857-4879.
[76] TABAGHSAR B J, TAVOLI R, TOOSI M M A. Presenting a three layer stacking ensemble classifier of deep learning and machine learning for skin cancer classification[J]. Multimedia Tools and Applications, 2025, 84(9): 6135-6156.
[77] 王泽彤, 张俊华, 王肖. SFA-ConvNeXt: 逐级聚合多尺度ConvNeXt的皮肤镜图像分类[J]. 计算机工程与应用, 2024, 60(20): 244-253.
WANG Z T, ZHANG J H, WANG X. SFA-ConvNeXt: dermoscopic image classification for stepwise aggregation of multiscale ConvNeXt[J]. Computer Engineering and Applications, 2024, 60(20): 244-253.
[78] HAGGERTY H, CHANDRA R. Self-supervised learning for skin cancer diagnosis with limited training data[J]. arXiv:2401.00692, 2024.
[79] ?ZDEMIR Z, KELES H Y, TANRI?VER ? ?. Meta-transfer derm-diagnosis: exploring few-shot learning and transfer learning for skin disease classification in long-tail distribution[J]. arXiv:2404.16814, 2024.
[80] CHEN T L, LIU Q, YANG J. Few-shot classification with multiscale feature fusion for clinical skin disease diagnosis[J]. Clinical, Cosmetic and Investigational Dermatology, 2024, 17: 1007-1026.
[81] ASIF S, QURRAT-UL-AIN, KHAN S U R, et al. SKINC-NET: an efficient lightweight deep learning model for multiclass skin lesion classification in dermoscopic images[J]. Multimedia Tools and Applications, 2025, 84(13): 12531-12557.
[82] MIDASALA V D, PRABHAKAR B, KRISHNA CHAITANYA J, et al. MFEUsLNet: skin cancer detection and classification using integrated AI with multilevel feature extraction-based unsupervised learning[J]. Engineering Science and Technology, an International Journal, 2024, 51: 101632.
[83] CINO L, DISTANTE C, MARTELLA A, et al. Skin lesion classification through test time augmentation and explainable artificial intelligence[J]. Journal of Imaging, 2025, 11(1): 15.
[84] KAMAL S, OATES T. MedGrad E-CLIP: enhancing trust and transparency in AI-driven skin lesion diagnosis[J]. arXiv:2501.06887, 2025.
[85] SAYED G I, SOLIMAN M M, HASSANIEN A E. A novel melanoma prediction model for imbalanced data using optimized SqueezeNet by bald eagle search optimization[J]. Computers in Biology and Medicine, 2021, 136: 104712.