Multi-Attention and Ensemble Learning for Precise Dermoscopic Diagnosis
Received: 18 October 2025 | Revised: 14 November 2025 and 28 November 2025 | Accepted: 29 November 2025 | Online: 6 June 2026
Corresponding author: Umar Farooq Khattak
Abstract
Skin cancer has high incidence rates, and early diagnosis is crucial to improving survival rates. Thus, efficient diagnostic tools are imperative. This study suggests a new way to classify skin cancer by integrating a multi-attention Convolutional Neural Network (CNN) with ensemble learning for increased diagnostic precision. Utilizing the HAM10000 dataset with 10,015 dermoscopic images of seven different lesions, this study addresses significant issues such as class imbalance, noise, and inadequate feature extraction. Advanced preprocessing techniques, including contrast correction, noise removal, and hair removal, were used to standardize images and eliminate artifacts. The multi-attention mechanism enables the CNN to weight significant lesion features, such as asymmetry, irregular margins, and color variation, which are critical to distinguishing malignant from benign lesions. In addition, ensemble learning additionally ensures model stability by ensembling multiple classifiers to reduce prediction bias. Experimental results demonstrate that the proposed model compares favorably with standard deep learning techniques in terms of accuracy and sensitivity in distinguishing melanoma and non-melanoma lesions. By facilitating non-invasive and precise diagnosis, this technique has the potential to assist clinicians in early diagnosis, thereby increasing survival rates and reducing treatment costs. The proposed automated skin cancer diagnosis system holds potential for global healthcare systems, but its clinical deployment is currently limited by high computational demands and reduced generalizability to non-dermoscopic images.
Keywords:
skin cancer, deep learning, convolutional neural network, multi-attention mechanism, melanoma, ensemble learning, dermoscopic images, HAM10000 dataset, early diagnosis, class imbalanceReferences
WHO - Cancer Country Profiles Malaysia 2020," ICCP Portal. https://www.iccp-portal.org/resources/who-cancer-country-profiles-malaysia-2020.
R. L. Siegel, K. D. Miller, and A. Jemal, "Cancer statistics, 2018," CA: A Cancer Journal for Clinicians, vol. 68, no. 1, pp. 7–30, 2018.
Y. S. Alsahafi, M. A. Kassem, and K. M. Hosny, "Skin-Net: a novel deep residual network for skin lesions classification using multilevel feature extraction and cross-channel correlation with detection of outlier," Journal of Big Data, vol. 10, no. 1, June 2023, Art. no. 105.
J. B. A. Das, D. Mishra, A. Das, M. N. Mohanty, and A. Sarangi, "Skin cancer detection using machine learning techniques with ABCD features," in 2022 2nd Odisha International Conference on Electrical Power Engineering, Communication and Computing Technology (ODICON), Nov. 2022.
W. Paja, J. Szkoła, K. Pancerz, J. Sarzyński, and M. żychowska, "A Preliminary Research on Automatic Identification of Melanocytic Skin Lesions from Digital Images," Procedia Computer Science, vol. 225, pp. 4706–4712, Jan. 2023.
W. Salma and A. S. Eltrass, "Automated deep learning approach for classification of malignant melanoma and benign skin lesions," Multimedia Tools and Applications, vol. 81, no. 22, pp. 32643–32660, Sept. 2022.
S. Qian, K. Ren, W. Zhang, and H. Ning, "Skin lesion classification using CNNs with grouping of multi-scale attention and class-specific loss weighting," Computer Methods and Programs in Biomedicine, vol. 226, Nov. 2022, Art. no. 107166.
I. A. Alfi, M. M. Rahman, M. Shorfuzzaman, and A. Nazir, "A Non-Invasive Interpretable Diagnosis of Melanoma Skin Cancer Using Deep Learning and Ensemble Stacking of Machine Learning Models," Diagnostics, vol. 12, no. 3, Mar. 2022.
V. Anand, S. Gupta, D. Koundal, and K. Singh, "Fusion of U-Net and CNN model for segmentation and classification of skin lesion from dermoscopy images," Expert Systems with Applications, vol. 213, Mar. 2023, Art. no. 119230.
J. Avanija, C. Chandra Mohan Reddy, C. Sri Chandan Reddy, D. Harshavardhan Reddy, T. Narasimhulu, and N. V. Hardhik, "Skin Cancer Detection using Ensemble Learning," in 2023 International Conference on Sustainable Computing and Smart Systems (ICSCSS), June 2023, pp. 184–189.
K. Sethanan et al., "Double AMIS-ensemble deep learning for skin cancer classification," Expert Systems with Applications, vol. 234, Dec. 2023, Art. no. 121047.
K. M. Selvaraj, S. Gnanagurusubbiah, R. R. R. Roy, J. H. J. Peter, and S. Balu, "Enhancing skin lesion classification with advanced deep learning ensemble models: a path towards accurate medical diagnostics," Current Problems in Cancer, vol. 49, Apr. 2024, Art. no. 101077.
M. M. Musthafa, T. R. Manesh, K. V. Vinoth, and S. Guluwadi, "Enhanced skin cancer diagnosis using optimized CNN architecture and checkpoints for automated dermatological lesion classification," BMC Medical Imaging, vol. 24, no. 1, Aug. 2024, Art. no. 201.
A. Panthakkan, S. M. Anzar, S. Jamal, and W. Mansoor, "Concatenated Xception-ResNet50 — A novel hybrid approach for accurate skin cancer prediction," Computers in Biology and Medicine, vol. 150, Nov. 2022, Art. no. 106170.
X. He, Y. Wang, S. Zhao, and X. Chen, "Co-Attention Fusion Network for Multimodal Skin Cancer Diagnosis," Pattern Recognition, vol. 133, Jan. 2023, Art. no. 108990.
P. Li, T. Han, Y. Ren, P. Xu, and H. Yu, "Improved YOLOv4-tiny based on attention mechanism for skin detection," PeerJ Computer Science, vol. 9, Mar. 2023, Art. no. e1288.
Y. Wan, Y. Cheng, and M. Shao, "MSLANet: multi-scale long attention network for skin lesion classification," Applied Intelligence, vol. 53, no. 10, pp. 12580–12598, May 2023.
D. Halabi, "Enhancing Skin Cancer Detection and Classification: Exploring the Impact of Attention Mechanisms in Transfer Learning Models." Medicine and Pharmacology, Dec. 13, 2023.
Y. Wang, J. Cai, D. C. Louie, Z. J. Wang, and T. K. Lee, "Incorporating clinical knowledge with constrained classifier chain into a multimodal deep network for melanoma detection," Computers in Biology and Medicine, vol. 137, Oct. 2021, Art. no. 104812.
G. Cai and N. Lynch, "A Geometry-Sensitive Quorum Sensing Algorithm for the Best-of-N Site Selection Problem," in Swarm Intelligence, 2022, pp. 1–13.
M. A. Khan et al., "Automatic melanoma and non-melanoma skin cancer diagnosis using advanced adaptive fine-tuned convolution neural networks," Discover Oncology, vol. 16, no. 1, Apr. 2025, Art. no. 645.
M. D. Ali et al., "Breast Cancer Classification through Meta-Learning Ensemble Technique Using Convolution Neural Networks," Diagnostics, vol. 13, no. 13, June 2023, Art. no. 2242.
M. A. Khan et al., "An Advanced Deep Learning Framework for Skin Cancer Classification," The Review of Socionetwork Strategies, vol. 19, no. 1, pp. 111–130, Apr. 2025.
G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, "Densely Connected Convolutional Networks," in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 2017, pp. 2261–2269.
J. Kawahara, A. BenTaieb, and G. Hamarneh, "Deep features to classify skin lesions," in 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI), Apr. 2016, pp. 1397–1400.
N. Bansal and S. Sridhar, "Skin Lesion Classification Using Ensemble Transfer Learning," in Second International Conference on Image Processing and Capsule Networks, 2022, pp. 557–566.
Y. C. Hum et al., "The development of skin lesion detection application in smart handheld devices using deep neural networks," Multimedia Tools and Applications, vol. 81, no. 29, pp. 41579–41610, Dec. 2022.
M. D. R. H. Khan, A. H. Uddin, A. A. Nahid, and A. K. Bairagi, "Skin Cancer Detection from Low-Resolution Images Using Transfer Learning," in Intelligent Sustainable Systems, 2022, pp. 317–334.
M. S. Al Huda, T. E. Shrestha, A. Hossain, N. B. Sharif, M. A. Ali, and T. I. Erdei, "DeepMelaNet: Advancing Melanoma Stage Classification in Skin Cancer Diagnosis," Engineering, Technology & Applied Science Research, vol. 15, no. 1, pp. 19627–19635, Feb. 2025.
P. Tschandl, C. Rosendahl, and H. Kittler, "The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions," Scientific Data, vol. 5, no. 1, Aug. 2018, Art. no. 180161.
P. Tschandl, "The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions." Harvard Dataverse, Feb. 07, 2023.
Downloads
How to Cite
License
Copyright (c) 2026 Muhammad Amir Khan, Razia Manan, Ali Khalid, Mohammad Shahid, Umar Farooq Khattak
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
