Title
Deep Learning-Based Multi-Class Skin Cancer Detection: Methods, Challenges, and Future Directions
Authors
G. Nandhini, T. Ananth Kumar, P. Kanimozhi
Abstract
Skin cancer is one of the most common cancers in the world and late diagnosis can result in sub-optimal treatment outcomes and mortality. The skin cancer incidence rate is steadily rising annually, which makes it even more crucial to develop automated diagnostic tools that can assist dermatologists in making correct and timely choices. In recent years, Artificial Intelligence (AI), including deep learning and machine learning have been increasingly adopted for the analysis of dermatoscopic skin images for skin cancer detection. This paper summarizes all the recently available techniques for multi-class skin cancer detection based on AI, including the datasets utilized by researchers, the deep learning architecture developed and the different classification methods proposed. The strengths and limitations of each method are also considered, and the major challenges and open research issues are explored. The primary aim of this work is to provide researchers with a clear depiction of the current state of the field and to highlight future research priorities.
Keywords
Skin Cancer Detection; Deep Learning; Multi-Class Classification; Convolutional Neural Networks, Explainable Artificial Intelligence
Full Text
References
[1] A. H. Roky, M. M. Islam, A. M. F. Ahasan, M. S. Mostaq, M. Z. Mahmud, M. N. Amin, and M. A. Mahmud,
“Overview of skin cancer types and prevalence rates across continents,” Cancer Pathogens and Therapy, vol. 3,
no. 2, pp. 89–100, 2024, doi: 10.1016/j.cpt.2024.08.002.
[2] A. Naeem, T. Anees, M. Khalil, K. Zahra, R. A. Naqvi, and S. W. Lee, “SNC Net: Skin cancer detection by
integrating handcrafted and deep learning-based features using dermoscopy images,” Mathematics, vol. 12, no.
7, p. 1030, 2024, doi: 10.3390/math12071030.
[3] M. Tahir, A. Naeem, H. Malik, J. Tanveer, R. A. Naqvi, and S. W. Lee, “DSCC Net: Multi-classification deep
learning models for diagnosing skin cancer using dermoscopic images,” Cancers, vol. 15, no. 7, p. 2179, 2023,
doi: 10.3390/cancers15072179.
[4] B. Vasudevan, D. Mitra, and V. Deshwal, “Artificial intelligence in dermatology: The shifting paradigms,” Medical
Journal Armed Forces India, 2025, doi: 10.1016/j.mjafi.2025.08.016.
[5] H. Naseri and A. A. Safaei, “Diagnosis and prognosis of melanoma from dermoscopy images using machine
learning and deep learning: A systematic literature review,” BMC Cancer, vol. 25, no. 1, p. 75, 2025, doi:
10.1186/s12885-024-13423-y.
[6] A. M. Zbrzezny and T. Krzywicki, “Artificial intelligence in dermatology: A review of methods, clinical
applications, and perspectives,” Applied Sciences, vol. 15, no. 14, p. 7856, 2025, doi: 10.3390/app15147856.
[7] S. Górecki, A. Tatka, and J. Brusey, “Artificial intelligence and new technologies in melanoma diagnosis: A
narrative review,” Cancers, vol. 17, no. 24, p. 3896, 2025, doi: 10.3390/cancers17243896.
[8] X. Chen, Z. Zhou, H. Ding, H. Zheng, and Y. Ge, “Transforming aesthetic dermatology: The role of artificial
intelligence in skin health,” Dermatology and Therapy, vol. 15, pp. 1999–2013, 2025, doi: 10.1007/s13555-025-
01459-2.
[9] S. S. Han, S. I. Cho, S. E. Chang, S. H. Kim, and J. I. Na, “Validation of an AI for skin diseases in Korea and global
usage statistics,” medRxiv, 2025, doi: 10.1101/2025.01.27.25321219.
[10] N. Nadour, T. Duguet, S. Zahedi, H. Figoni, and R. Liard, “Diagnostic accuracy of artificial intelligence compared
to family physicians and dermatologists for skin conditions: A systematic review and meta-analysis,” BMC
Primary Care, vol. 26, p. 384, 2025, doi: 10.1186/s12875-025-03073-9.
[11] N. Verma, Ranvijay, and D. K. Yadav, “A comprehensive review on step-based skin cancer detection using
machine learning and deep learning methods,” Archives of Computational Methods in Engineering, vol. 32, pp.
4359–4412, 2025, doi: 10.1007/s11831-025-10275-y.
[12] G. Narmadha and M. Sivasakthi, “An investigation on the detection and classification of skin cancer using deep
learning,” in Artificial Intelligence and Technology: Systems Management, Decisions and Control for
Sustainability in the Digital Age, pp. 567–576, 2026.
[13] Y. K. Ahmed and A. N. A. Naji, “Smart feature extraction using deep learning for early diagnosis of chronic
diseases,” Network Modeling Analysis in Health Informatics and Bioinformatics, vol. 14, p. 140, 2025, doi:
10.1007/s13721-025-00641-y.
[14] A. Alrabai, A. Echtioui, and F. Kallel, “Explainable transfer learning for skin cancer diagnosis,” Network
Modeling Analysis in Health Informatics and Bioinformatics, vol. 15, no. 1, p. 30, 2026, doi: 10.1007/s13721-
025-00709-9.
[15] O. Attallah, “Skin-CAD: Explainable deep learning classification of skin cancer from dermoscopic images by
feature selection of dual high-level CNN features and transfer learning,” Computers in Biology and Medicine,
vol. 178, p. 108798, 2024, doi: 10.1016/j.compbiomed.2024.108798.
[16] H. AbuAlkebash, R. A. A. Saleh, and H. M. Ertunc, “Explainable depth-wise and channel-wise fusion models for
multi-class skin lesion classification,” PLoS One, vol. 21, no. 1, p. e0340901, 2026, doi:
10.1371/journal.pone.0340901.
[17] A. A. Noor, A. Manzoor, M. D. M. Qureshi, M. A. Qureshi, and W. Rashwan, “Unveiling explainable AI in
healthcare: Current trends, challenges, and future directions,” Wiley Interdisciplinary Reviews: Data Mining and
Knowledge Discovery, 2025, doi: 10.1002/widm.70018.
[18] O. Uwishema, M. Ghezzawi, N. Charbel, S. Alawieh, S. Roy, M. Wojtara, C. M. Hakayuwa, I. K. Ja’afar, and M.
Prasad, “Diagnostic performance of artificial intelligence for dermatological conditions: A systematic review,”
International Journal of Emergency Medicine, vol. 18, p. 172, 2025, doi: 10.1186/s12245-025-00975-4.
[19] M. S. Yaswanth and C. M. K. Reddy, “Unresolved challenges and future directions in AI-based skin cancer
detection,” in Proc. IEEE ICECA, 2025, doi: 10.1109/ICECA66444.2025.11382676.
[20] T. Majtner, S. Yildirim-Yayilgan, and J. Y. Hardeberg, “Combining deep learning and hand-crafted features for
skin lesion classification,” in Proc. 6th Int. Conf. Image Processing Theory, Tools and Applications (IPTA),
2016, pp. 1–6, doi: 10.1109/IPTA.2016.7821017.
[21] E. M. Senan and M. E. Jadhav, “Diagnosis of dermoscopy images for the detection of skin lesions using SVM and
KNN,” in Proc. 3rd Int. Conf. Sustainable Computing (SUSCOM 2021), Singapore: Springer, 2022, pp. 125–
134.
[22] M. A. Kassem, K. M. Hosny, R. Damaševičius, and M. M. Eltoukhy, “Machine learning and deep learning
methods for skin lesion classification and diagnosis: A systematic review,” Diagnostics, vol. 11, no. 8, p. 1390,
2021, doi: 10.3390/diagnostics11081390.
[23] T. R. S. Fernandes, A. S. Teles, J. R. N. Fernandes, L. D. B. Lima, D. L. Sousa, and R. C. Soares, “External
validation of AI models for skin diseases: A systematic review,” IEEE Access, 2025, doi:
10.1109/ACCESS.2025.3584904.
[24] M. Naqvi, S. Q. Gilani, T. Syed, O. Marques, and H.-C. Kim, “Skin cancer detection using deep learning—A
review,” Diagnostics, vol. 13, no. 11, p. 1911, 2023, doi: 10.3390/diagnostics13111911.
[25] M. A. Albahar, “Skin lesion classification using convolutional neural network with novel regularizer,” IEEE
Access, vol. 7, pp. 38306–38313, 2019, doi: 10.1109/ACCESS.2019.2906241.
[26] R. Tanna and T. Sharma, “Binary classification of melanoma skin cancer using SVM and CNN,” in Proc. Int.
Conf. Artificial Intelligence and Machine Vision (AIMV), 2021, pp. 1–4, doi:
10.1109/AIMV53313.2021.9670894.
[27] N. M. Rashad, N. M. M. Abdelnapi, A. F. Seddik, and M. A. Sayedelahl, “Automating skin cancer screening: A
deep learning approach,” Journal of Engineering and Applied Science, vol. 72, p. 6, 2025, doi: 10.1186/s44147-
024-00573-w.
[28] G. Durga Devi and D. M. D. Preethi, “Design and analysis of melanoma skin cancer using deep learning and
machine learning,” in Proc. IEEE ICDICI, 2025, doi: 10.1109/ICDICI66477.2025.11134986.
[29] K. Wanjari and P. Verma, “A review on the applications of machine learning and deep learning techniques for
skin cancer detection,” in Proc. IEEE ICSADL, 2025, doi: 10.1109/ICSADL65848.2025.10933289.
[30] S. A. H. Shah, S. T. H. Shah, R. Khaled, A. Buccoliero, S. B. H. Shah, A. Di Terlizzi, G. Di Benedetto, and M.
A. Deriu, “Explainable AI-based skin cancer detection using CNN, particle swarm optimization and machine
learning,” Journal of Imaging, vol. 10, no. 12, p. 332, 2024, doi: 10.3390/jimaging10120332.
[31] Z. M. Gohil and M. B. Desai, “Revolutionizing dermatology: A comprehensive survey of AI-enhanced early skin
cancer diagnosis,” Archives of Computational Methods in Engineering, vol. 31, pp. 4521–4531, 2024, doi:
10.1007/s11831-024-10121-7.
[32] H. A. Haenssle, C. Fink, R. Schneiderbauer, F. Toberer, T. Buhl, A. Blum, A. Kalloo, A. B. H. Hassen, L. Thomas,
A. Enk, and L. Uhlmann, “Man against machine: Diagnostic performance of a deep learning convolutional neural
network for dermoscopic melanoma recognition in comparison to 58 dermatologists,” Annals of Oncology, vol.
29, no. 8, pp. 1836–1842, 2018, doi: 10.1093/annonc/mdy166.
[33] H. Feng and L. Zhang, “EfficientNet-based skin cancer classification and recognition,” in Proc. SPIE, vol. 13800,
p. 138001Q, 2025, doi: 10.1117/12.3076682.
[34] A. Mahbod, G. Schaefer, C. Wang, G. Dorffner, R. Ecker, and I. Ellinger, “Transfer learning using a multi-scale
and multi-network ensemble for skin lesion classification,” Computer Methods and Programs in Biomedicine,
vol. 193, p. 105475, 2020, doi: 10.1016/j.cmpb.2020.105475.
[35] M. Zakariah, M. Al-Razgan, and T. Alfakih, “Skin cancer detection with MobileNet-based transfer learning and
MixNets for enhanced diagnosis,” Neural Computing and Applications, vol. 36, pp. 21383–21413, 2024, doi:
10.1007/s00521-024-10227-w.
[36] A. Bello, S.-C. Ng, and M.-F. Leung, “Skin cancer classification using fine-tuned transfer learning of DenseNet121,” Applied Sciences, vol. 14, no. 17, p. 7707, 2024, doi: 10.3390/app14177707.
[37] V. Anand, S. Gupta, S. R. Nayak, D. Koundal, D. Prakash, and K. D. Verma, “An automated deep learning model
for classification of skin disease using dermoscopy images: A comprehensive study,” Multimedia Tools and
Applications, vol. 81, no. 26, pp. 37379–37401, 2022, doi: 10.1007/s11042-021-11628-y.
[38] M. Ennab and H. Mcheick, “Advancing AI interpretability in medical imaging: A comparative analysis of pixellevel interpretability and Grad-CAM models,” Machine Learning and Knowledge Extraction, vol. 7, no. 1, p.
12, 2025, doi: 10.3390/make7010012.
[39] S. Sharma, A. Sharma, A. Verma, N. K. Fuloria, and A. Sharma, “Enhancing skin cancer detection with transfer
learning and deep learning,” in AI and IoT-Enhanced Skin Cancer Detection and Care, pp. 233–272, 2025, doi:
10.2174/97988988119521250101.
[40] C. Revanth, T. Balaji, N. K. Kumar, D. S. K. Goud, G. S. Haritha, and B. Praveen, “Optimizing skin cancer
detection using MobileNet,” in Proc. IEEE AIMLA, 2025, doi: 10.1109/AIMLA63829.2025.11040509.
[41] M. Badr, A. Elkasaby, M. Alrahmawy, and S. El-Metwally, “A multi-model deep learning architecture for
diagnosing multi-class skin diseases,” Journal of Imaging Informatics in Medicine, vol. 38, no. 3, pp. 1776–
1795, 2025, doi: 10.1007/s10278-024-01300-w.
[42] P. Dubey, P. Bande, R. Shukla, S. S. Kshatri, M. Nayak, and R. Sharma, “AI-powered dermatology:
Revolutionizing skin cancer detection through machine learning algorithms,” in Proc. Int. Conf. Cybernation
and Computation (CYBERCOM), IEEE, 2024, doi: 10.1109/CYBERCOM63683.2024.10803244.
[43] B. Ozdemir and I. Pacal, “A robust deep learning framework for multiclass skin cancer classification,” Scientific
Reports, vol. 15, no. 1, p. 4938, 2025, doi: 10.1038/s41598-025-89230-7.
[44] H. Chelabi, B. Chikhaoui, A. Kermi, and M. T. Khadir, “Classification of skin lesion images based on attention
mechanism,” in Proc. IEEE/ACS 22nd Int. Conf. Computer Systems and Applications (AICCSA), 2025, pp. 1–7,
doi: 10.1109/AICCSA66935.2025.11315191.
[45] S. Liu, Z. Zhuang, Y. Zheng, and S. Kolmaniče, “A VAN-based multi-scale cross-attention mechanism for skin
lesion segmentation network,” IEEE Access, vol. 11, pp. 81953–81964, 2023, doi:
10.1109/ACCESS.2023.3298826.
[46] A. R. Nath and S. Bhattacharjee, “A novel vision transformer based approach for skin lesion classification: A deep
learning framework,” in Proc. Int. Conf. Quantum Photonics, Artificial Intelligence, and Networking (QPAIN),
2025, pp. 1–5, doi: 10.1109/QPAIN66474.2025.11172224.
[47] R. Abdel Mawgoud and C. Posch, “Visual explainability of 250 skin diseases viewed through the eyes of an AIbased, self-supervised vision transformer—A clinical perspective,” JEADV Clinical Practice, vol. 4, no. 1, pp.
145–155, 2025, doi: 10.1002/jvc2.580.
[48] S. P. Jeong, J. Liu, and S. J. Fong, “Bridging deep learning in skincare and dermatology: Opportunities for hybrid
models and early skin cancer prediction,” International Journal of Contents, vol. 21, no. 2, p. 24, 2025, doi:
10.5392/ijoc.2025.21.2.024.
[49] I. Pacal, B. Ozdemir, J. Zeynalov, H. Gasimov, and N. Pacal, “A novel CNN-ViT-based deep learning model for
early skin cancer diagnosis,” Biomedical Signal Processing and Control, vol. 104, p. 107627, 2025, doi:
10.1016/j.bspc.2025.107627.
[50] A. El Mahdi, R. Abla, E. Assia, S. Abdelouahed, and A. Abdellah, “Fusion of Vision Transformer and VGG
features for enhanced skin lesion classification,” in Proc. ICCSC 2025, 2025, doi:
10.1109/ICCSC66714.2025.11135399.
[51] M. Z. Hasan and F. Y. Rifat, “Hybrid ensemble of segmentation-assisted classification and GBDT for skin cancer
detection,” arXiv preprint arXiv:2506.03420, 2025, doi: 10.48550/arXiv.2506.03420.
[52] K. Palaniswamy, M. MeenakshiSundaram, S. Arige, and T. B. Krishna, “Multimodal deep learning for skin cancer
diagnosis using images and metadata,” in Proc. IEEE CICT, 2025, doi: 10.1109/CICT67193.2025.11398859.
[53] P. G. Dhange and P. William, “Exploring deep neural architectures for skin lesion detection and classification,”
in Proc. Int. Conf. Computational Intelligence and Knowledge Economy (ICCIKE), IEEE, 2025, pp. 1–6, doi:
10.1109/ICCIKE67021.2025.11318236.
[54] B. Soundarya and C. Poongodi, “A novel hybrid feature fusion approach using hand-crafted features with transfer
learning model for enhanced skin cancer classification,” Computers in Biology and Medicine, vol. 190, p.
110104, 2025, doi: 10.1016/j.compbiomed.2025.110104.
[55] K. Sreekala, N. Rajkumar, R. Sugumar, K. V. D. Sagar, R. Shobarani, K. P. Krishnamoorthy, A. K. Saini, H.
Palivela, and A. Yeshitla, “Skin diseases classification using a hybrid AI-based localization approach,”
Computational Intelligence and Neuroscience, vol. 2022, p. 6138490, 2022, doi: 10.1155/2022/6138490.
[56] H. H. Aras and N. Doğan, “A comparative analysis for skin cancer detection by using explainable deep learning,”
Neural Computing and Applications, vol. 38, no. 4, p. 50, 2026.
[57] M. B. Jan, M. Rashid, R. Vavekanand, and V. Singh, “Integrating explainable AI for skin lesion classifications:
A systematic literature review,” SMHS Journal, vol. 2, no. 1, 2023, doi: 10.48185/smhs.v2i1.1422.
[58] D. Khurana, R. Kaur, R. K. Roul, and S. Batra, “Explainable artificial intelligence in cancer care: A domain-wise
review of adoption, challenges and opportunities,” Expert Systems, 2026, doi: 10.1111/exsy.70238.
[59] Y. He, J. Zheng, and E. Zou, “SHARPEN-CAM: Efficient hierarchical SHAP-based visual explanation for deep
CNNs,” Multimedia Systems, vol. 32, p. 3, 2026.
[60] S. Suara, A. Jha, P. Sinha, and A. A. Sekh, “Is Grad-CAM explainable in medical images?” in Computer Vision
and Image Processing (CVIP 2023), pp. 124–135, 2024.
[61] A. W. Lubis, A. G. Putrada, and A. Q. Istiqfarri, “ReXNet-150 and Grad-CAM for efficient and explainable
melanoma skin cancer detection,” in Proc. 5th Int. Conf. Intelligent Cybernetics Technology and Applications
(ICICyTA), 2025, pp. 137–142, doi: 10.1109/ICICyTA68677.2025.11362836.
[62] C. C. Ukwuoma, D. Cai, E. O. Eziefuna, A. Oluwasanmi, S. F. Abdi, G. W. Muoka, D. Thomas, and K. Sarpong,
“Enhancing histopathological medical image classification for early cancer diagnosis using deep learning and
explainable AI—LIME and SHAP,” Biomedical Signal Processing and Control, vol. 100, p. 107014, 2025, doi:
10.1016/j.bspc.2024.107014.
[63] Z. S. A. Wahid, Z. C. Oleiwi, and R. F. Kadhem, “An analytical and comparative study of modern intelligent
models for skin cancer diagnosis,” Journal of Quantum Computational Science Modeling, vol. 18, p. 12671,
2026, doi: 10.29304/jqcsm.2026.18.12671.
[64] J. W. Tjiu and C. F. Lu, “Equity and generalizability of artificial intelligence for skin-lesion diagnosis using
clinical, dermoscopic, and smartphone images: A systematic review and meta-analysis,” Medicina, vol. 61, no.
12, p. 2186, 2025, doi: 10.3390/medicina61122186.
[65] B. Cassidy, C. Kendrick, A. Brodzicki, J. Jaworek-Korjakowska, and M. H. Yap, “Analysis of the ISIC image
datasets: Usage, benchmarks and recommendations,” Medical Image Analysis, vol. 75, p. 102305, 2022, doi:
10.1016/j.media.2021.102305.
[66] D. Ajabani, Z. A. Shaikh, A. Yousef, K. Ali, and M. A. Albahar, “Enhancing skin lesion classification with human
baseline comparison,” PeerJ Computer Science, vol. 11, p. e2795, 2025, doi: 10.7717/peerj-cs.2795.
[67] A. M. Shabu, M. S. Ansari, and A. Aslam, “A generative AI approach for reducing skin tone bias in skin cancer
classification,” arXiv preprint arXiv:2602.14356, 2026, doi: 10.48550/arXiv.2602.14356.
[68] A. Solak, “HCHS-Net: A multimodal handcrafted feature and metadata framework for interpretable skin lesion
classification,” Biomimetics, vol. 11, no. 2, p. 154, 2026, doi: 10.3390/biomimetics11020154.
[69] N.-Y. Tran-Van and K.-H. Le, “A multimodal skin lesion classification through cross-attention fusion and
collaborative edge computing,” Computerized Medical Imaging and Graphics, vol. 124, p. 102588, 2025, doi:
10.1016/j.compmedimag.2025.102588.
[70] M. M. Yaqoob, M. Alsulami, M. A. Khan, and coauthors, “Symmetry in privacy-based health-care: A review of
skin cancer detection using federated learning,” Symmetry, vol. 15, no. 7, p. 1369, 2023, doi:
10.3390/sym15071369.
[71] S. Chamarthi, K. Fogelberg, T. J. Brinker, and J. Niebling, “Mitigating the influence of domain shift in skin lesion
classification: A benchmark study of unsupervised domain adaptation methods,” Informatics in Medicine
Unlocked, vol. 44, p. 101430, 2024, doi: 10.1016/j.imu.2023.101430.
[72] N. Alipour, T. Burke, and J. Courtney, “Skin type diversity in skin lesion datasets: A review,” Current
Dermatology Reports, vol. 13, pp. 198–210, 2024, doi: 10.1007/s13671-024-00440-0.
[73] Y. Diao, X. Chen, Z. Huang, Q. Tan, M. Yang, H. Yu, Y. Xu, and X. Hu, “Cutting-edge AI technologies in skin
cancer applications,” Cancer Letters, p. 218256, 2026, doi: 10.1016/j.canlet.2026.218256.
[74] N. Veeramani and P. Jayaraman, “A promising AI-based super-resolution image reconstruction technique for
early diagnosis of skin cancer,” Scientific Reports, vol. 15, p. 5084, 2025, doi: 10.1038/s41598-025-89693-8.
[75] N. M. Eshwarappa, H. Baghban, C. H. Hsu, P. Y. Hsu, R. H. Hwang, and M. Y. Chen, “Communication-efficient
federated learning for medical image classification,” Journal of Cloud Computing, vol. 14, p. 44, 2025, doi:
10.1186/s13677-025-00734-z.
[76] C. Supriyanto, A. Salam, J. Zeniarja, D. W. Utomo, I. N. Dewi, C. Paramita, A. Wijaya, and N. Z. M. Safar, “A
bibliometric review of deep learning approaches in skin cancer research,” Computation, vol. 13, no. 3, p. 78,
2025, doi: 10.3390/computation13030078.
[77] M. Alsaidi, M. T. Jan, A. Altaher, H. Zhuang, and X. Zhu, “Tackling class imbalance in dermoscopic image
classification using data augmentation and GAN,” Multimedia Tools and Applications, vol. 83, pp. 49121–
49147, 2024, doi: 10.1007/s11042-023-17067-1.
[78] V. Sydorskyi, I. Krashenyi, and O. Yakubenko, “Multimodal system for skin cancer detection,” arXiv preprint
arXiv:2601.14822, 2026, doi: 10.48550/arXiv.2601.14822.
[79] J. Kawahara, S. Daneshvar, G. Argenziano, and G. Hamarneh, “Seven-point checklist and skin lesion classification
using multitask multimodal neural networks,” IEEE Journal of Biomedical and Health Informatics, vol. 23, no.
2, pp. 538–546, 2019, doi: 10.1109/JBHI.2018.2824327.
[80] Pulaski, H., Harrison, S.A., Mehta, S.S., Sanyal, A.J., Vitali, M.C., Manigat, L.C., Hou, H., Madasu Christudoss,
S.P., Hoffman, S.M., Stanford-Moore, A. and Egger, R., 2025. Clinical validation of an AI-based pathology tool
for scoring of metabolic dysfunction-associated steatohepatitis. Nature Medicine, 31(1), pp.315-322., doi:
10.1038/s41591-024-03301-2.