Current issue
Journal Information
Paper submission
Volumes and Issues
Contact
2/2026 vol. 20
Stats
CC BY-NC-ND 4.0
Get citation
RESEARCH PAPER
CNN-based Ensemble Architectures with Explanainable AI for Cutaneous Melanoma Identification
Adrian SZYMCZYK 1
,
 
Maria SKUBLEWSKA-PASZKOWSKA 1
,
 
Paweł POWROŹNIK 1
 
 
 
More details
Hide details
1
Department of Computer Science, Lublin University of Technology, Poland
 
 
Submission date: 2025-12-23
 
 
Final revision date: 2026-04-20
 
 
Acceptance date: 2026-05-02
 
 
Publication date: 2026-06-05
 
 
Corresponding author
Maria SKUBLEWSKA-PASZKOWSKA   

Department of Computer Science, Lublin University of Technology, Nadbystrzycka 38D, Nadbystrzycka 36B, 20-618, Lublin, Poland
 
 
Acta Mechanica et Automatica 2026;20(2):317-330
DOI: https://doi.org/10.65731/ama/2026-0033
 
 
Article (PDF)
References (53)
 
HIGHLIGHTS
  • Grad-CAM to highlight the most significant area for model classification
  • Preparing dataset consisting of multiple ISIC datasets from 2018, 2019, and 2020
  • Comparing the effectiveness of pre-trained ResNet152, DenseNet201, EfficientNet-B4
  • Applying ensemble methods with soft voting, hard voting, stacking
KEYWORDS
melanoma
ensemble learning
deep learning
convolutional neural networks
Grad-CAM
skin cancer identification
 
TOPICS
ABSTRACT
Malignant melanoma, a highly aggressive form of skin cancer, poses a significant global health challenge due to its rapid progression and high mortality rate if not detected on time. Early diagnosis is crucial for improving patient outcomes. The effectiveness of skin cancer detec-tion still faces serious challenges, like visual inspection that is less accurate and time-consuming. However, deep learning-based models provide early and accurate diagnosis, serving as a supporting tool for dermatologists. Thus, this study focuses on indicating the most suit-able model for skin diseases identification. Three prominent, pre-trained deep learning models, ResNet152, DenseNet201 and EfficientNet-B4, were involved in order to detect benign and malignant melanoma skin lesions. The study was performed utilizing a combined ISIC da-tasets gathered between 2018 and 2020 that consist of dermoscopic images. The above-mentioned deep learning algorithms were verified using accuracy, precision, recall, and F1-score metrics. Moreover, in this study the performance of skin cancer detection was enhanced uti-lizing soft, hard voting, and XGBoost ensemble learning methods. Combining two and three models were verified. The single models ob-tained accuracy at the level of 89.20%, 88.20%, and 90.40% for ResNet152, DenseNet201 and EfficientNet-B4, respectively. The soft vot-ing ensemble, merging ResNet-152 with EfficientNet-B4 or all three models, achieved the highest absolute accuracy of 91.30%, demonstrat-ing superior performance in melanoma diagnosis compared to individual models. Hard voting and XGBoost stated to be less effective in melanoma diagnosis. To confirm that the models were making decisions based on the significant image regions representing skin lesions, a visual explainable technique was applied. Gradient-weighted Class Activation Mapping proved the models to focus their attention to the relevant disease features. These findings underscore the potential of combining individual model strengths through ensemble learning to achieve superior diagnostic performance in melanoma detection, supporting clinicians in making more accurate and timely diagnosis.
REFERENCES (53)
1.
Meedeniya D, De Silva S, Gamage L, Isuranga U. (2024). Skin cancer identification utilizing deep learning: A survey. IET Image Processing. 2024; 18(13): 3731-3749.
Google Scholar
 
 
2.
Arnold M, Singh D, Laversanne M, Vignat J, Vaccarella S, Meheus F, Cust AE, De Vries E, Whiteman DC, Bray F. Global Burden of Cuta-neous Melanoma in 2020 and Projections to 2040. JAMA Dermatol. 2022;158(5):495.
Google Scholar
 
 
3.
Melanoma Awareness Month 2022 [Internet]. Available from: https://www.iarc.who.int/news-....
Google Scholar
 
 
4.
Shakya M, Patel R, Joshi S. A comprehensive analysis of deep learn-ing and transfer learning techniques for skin cancer classification. Sci Rep [Internet]. 2025;15(1). Available from: https://www.nature.com/article....
Google Scholar
 
 
5.
Morton, Mackie. Clinical accuracy of the diagnosis of cutaneous malig-nant melanoma. Br J Dermatol. 1998;138(2):283–7.
Google Scholar
 
 
6.
Powroznik P, Skublewska-Paszkowska M, Dziedzic K, Barszcz M. Feature Fusion Graph Consecutive-Attention Network for Skeleton-Based Tennis Action Recognition. Appl Sci. 2025;15(10):5320.
Google Scholar
 
 
7.
Powroźnik P, Skublewska-Paszkowska M, Nowomiejska K, Aristidou A, Panayides A, Rejdak R. Deep convolutional generative adversarial networks in retinitis pigmentosa disease images augmentation and de-tection. Adv Sci Technol Res J. 2024;19(2):321–40.
Google Scholar
 
 
8.
Skublewska-Paszkowska M, Powroznik P, Rejdak R, Nowomiejska K. Application of Convolutional Gated Recurrent Units U-Net for Distin-guishing between Retinitis Pigmentosa and Cone–Rod Dystrophy. Acta Mech Autom. 2024;18(3):505–13.
Google Scholar
 
 
9.
Nawaz K, Zanib A, Shabir I, Li J, Wang Y, Mahmood T, Rehman A. Skin cancer detection using dermoscopic images with convolutional neural network. Sci Rep [Internet]. 2025;15(1). Available from: https://www.nature.com/article....
Google Scholar
 
 
10.
Ahmad M, Ahmed I, Chehri A, Jeon G. Fusion of metadata and dermo-scopic images for melanoma detection: Deep learning and feature im-portance analysis. Inf Fusion. 2025;124:103304.
Google Scholar
 
 
11.
Esmaeili V, Mohassel Feghhi M, Seyedarabi H. Automatic melanoma detection using an optimized five-stream convolutional neural network. Sci Rep [Internet]. 2025;15(1). Available from: https://www.nature.com/article....
Google Scholar
 
 
12.
Zhang R. Melanoma Detection Using Convolutional Neural Network. In: 2021 IEEE International Conference on Consumer Electronics and Computer Engineering (ICCECE) [Internet]. Guangzhou, China: IEEE; 2021;75–8. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
13.
Ahmed A, Sun G, Bilal A, Li Y, Ebad SA. (2025). Precision and effi-ciency in skin cancer segmentation through a dual encoder deep learn-ing model. Scientific Reports. 2025; 15(1): 4815.
Google Scholar
 
 
14.
Ozdemir B, Pacal I. A robust deep learning framework for multiclass skin cancer classification. Scientific Reports. 2025; 15(1): 4938.
Google Scholar
 
 
15.
SMJ, PM, Aravindan C, Appavu R. Classification of skin cancer from dermoscopic images using deep neural network architectures. Multimed Tools Appl. 2023;82(10):15763–78.
Google Scholar
 
 
16.
Kwiatkowska D, Kluska P, Reich A. Convolutional neural networks for the detection of malignant melanoma in dermoscopy images. Adv Der-matol Allergol. 2021;38(3):412–20.
Google Scholar
 
 
17.
Yu L, Chen H, Dou Q, Qin J, Heng PA. Automated Melanoma Recog-nition in Dermoscopy Images via Very Deep Residual Networks. IEEE Trans Med Imaging. 2017;36(4):994–1004.
Google Scholar
 
 
18.
Ali R, Manikandan A, Lei R, Xu J. A novel SpaSA based hyper-parameter optimized FCEDN with adaptive CNN classification for skin cancer detection. Sci Rep [Internet]. 2024;14(1). Available from: https://www.nature.com/article....
Google Scholar
 
 
19.
Nida N, Irtaza A, Javed A, Yousaf MH, Mahmood MT. Melanoma lesion detection and segmentation using deep region based convolution-al neural network and fuzzy C-means clustering. Int J Med Inf. 2019;124:37–48.
Google Scholar
 
 
20.
Hosny KM, Kassem MA, Foaud MM. Classification of skin lesions using transfer learning and augmentation with Alex-net. Zhang J. PLOS ONE. 2019;14(5):e0217293.
Google Scholar
 
 
21.
Qin Z, Liu Z, Zhu P, Xue Y. A GAN-based image synthesis method for skin lesion classification. Comput Methods Programs Biomed. 2020;195:105568.
Google Scholar
 
 
22.
Mahbod A, Schaefer G, Wang C, Ecker R, Ellinge I. Skin Lesion Classification Using Hybrid Deep Neural Networks. In: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) [Internet]. Brighton. United Kingdom: IEEE. 2019; 1229–33. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
23.
Veeramani N, Jayaraman P, Krishankumar R, Ravichandran KS, Gandomi AH. DDCNN-F: double decker convolutional neural network “F” feature fusion as a medical image classification framework. Sci Rep [Internet]. 2024;14(1). Available from: https://www.nature.com/article....
Google Scholar
 
 
24.
Wu H, Chen S, Chen G, Wang W, Lei B, Wen Z. FAT-Net: Feature adaptive transformers for automated skin lesion segmentation. Med Im-age Anal. 2022;76:102327.
Google Scholar
 
 
25.
Khan MA, Mazhar T, Ali MD, Khattak UF, Shahzad T, Saeed MM, Hamam H. Automatic melanoma and non-melanoma skin cancer diag-nosis using advanced adaptive fine-tuned convolution neural networks. Discov Oncol [Internet]. 2025;16(1). Available from: https://link.springer.com/10.1....
Google Scholar
 
 
26.
Aftab RS, Hamim SA, Ahmed MM, Shafi SMA, Mazid-Ul-Haque Md. SkinScanNet: A CNN-Based Model with Explainable AI for Reliable and Transparent Skin Cancer Detection. In: 2024 27th International Conference on Computer and Information Technology (ICCIT) [Inter-net]. Cox’s Bazar. Bangladesh: IEEE. 2024; 2846–51. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
27.
Kumar MK, Vaishnavi J, Anjibabu T, Jayasri V. Skin Cancer Detection: A Hybrid Approach Combining CNNs and Explainable AI. In: 2025 In-ternational Conference on Machine Learning and Autonomous Systems (ICMLAS) [Internet]. Prawet, Thailand: IEEE; 2025:1102–7. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
28.
Gamage L, Isuranga U, Meedeniya D, De Silva S, Yogarajah P. Mela-noma Skin Cancer Identification with Explainability Utilizing Mask Guid-ed Technique. Electronics. 2024;13(4):680.
Google Scholar
 
 
29.
Que SKT. Research Techniques Made Simple: Noninvasive Imaging Technologies for the Delineation of Basal Cell Carcinomas. J Invest Dermatol. 2016;136(4):e33–8.
Google Scholar
 
 
30.
The International Skin Imaging Collaboration [Internet]. Available from: https://www.isic-archive.com/.
Google Scholar
 
 
31.
Caffery LJ, Rotemberg V, Weber J, Soyer HP, Malvehy J, Clunie D. The Role of DICOM in Artificial Intelligence for Skin Disease. Front Med [Internet]. 2021;7. Available from: https://www.frontiersin.org/ar....
Google Scholar
 
 
32.
Shorten C, Khoshgoftaar TM. A survey on Image Data Augmentation for Deep Learning. J Big Data [Internet]. 2019;6(1). Available from: https://journalofbigdata.sprin....
Google Scholar
 
 
33.
Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg AC, Fei-Fei L. ImageNet Large Scale Visual Recognition Challenge [Internet]. arXiv; 2014. Avail-able from: https://arxiv.org/abs/1409.057....
Google Scholar
 
 
34.
He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition [Internet]. arXiv. 2015. Available from: https://arxiv.org/abs/1512.033....
Google Scholar
 
 
35.
Huang G, Liu Z, van der Maaten L, Weinberger KQ. Densely Connect-ed Convolutional Networks [Internet]. arXiv; 2016. Available from: https://arxiv.org/abs/1608.069....
Google Scholar
 
 
36.
Tan M, Le QV. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. 2019. Available from: https://arxiv.org/abs/1905.119....
Google Scholar
 
 
37.
Atila Ü, Uçar M, Akyol K, Uçar E. Plant leaf disease classification using EfficientNet deep learning model. Ecol Inform. 2021;61:101182.
Google Scholar
 
 
38.
Zhang C, Ma Y, editors. Ensemble Machine Learning: Methods and Applications [Internet]. New York, NY: Springer New York. 2012. Available from: https://link.springer.com/10.1....
Google Scholar
 
 
39.
Mienye ID, Sun Y. A Survey of Ensemble Learning: Concepts, Algo-rithms, Applications and Prospects. IEEE Access. 2022;10:99129–49.
Google Scholar
 
 
40.
Rokach L. Ensemble-based classifiers. Artif Intell Rev. 2010 Feb;33(1–2):1–39.
Google Scholar
 
 
41.
Chen T, Guestrin C. XGBoost: A Scalable Tree Boosting System. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining [Internet]. San Francisco Cali-fornia USA: ACM; 2016; 785–94. Available from: https://dl.acm.org/doi/10.1145....
Google Scholar
 
 
42.
Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D.. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision. 2017; 618-626.
Google Scholar
 
 
43.
Dogan A, Birant D. A Weighted Majority Voting Ensemble Approach for Classification. In: 2019 4th International Conference on Computer Sci-ence and Engineering (UBMK) [Internet]. Samsun. Turkey: IEEE; 2019; 1–6. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
44.
Al-masni MA, Al-antari MA, Park HM, Park NH, Kim TS. A Deep Learning Model Integrating FrCN and Residual Convolutional Networks for Skin Lesion Segmentation and Classification. In: 2019 IEEE Eura-sia Conference on Biomedical Engineering, Healthcare and Sustainabil-ity (ECBIOS) [Internet]. Okinawa, Japan: IEEE. 2019; 95–8. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
45.
Daghrir J, Tlig L, Bouchouicha M, Sayadi M. Melanoma skin cancer detection using deep learning and classical machine learning techniques: A hybrid approach. In: 2020 5th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP) [Internet]. Sousse. Tunisia: IEEE; 2020;1–5. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
46.
Shahin AH, Kamal A, Elattar MA. Deep Ensemble Learning for Skin Lesion Classification from Dermoscopic Images. In: 2018 9th Cairo In-ternational Biomedical Engineering Conference (CIBEC) [Internet]. Cai-ro. Egypt: IEEE. 2018; 150–3. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
47.
Kaymak S, Esmaili P, Serener A. Deep Learning for Two-Step Classi-fication of Malignant Pigmented Skin Lesions. In: 2018 14th Symposi-um on Neural Networks and Applications (NEUREL) [Internet]. Bel-grade: IEEE. 2018; 1–6. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
48.
Yu Z, Jiang X, Zhou F, Qin J, Ni D, Chen S, Lei B, Wang T. Melano-ma Recognition in Dermoscopy Images via Aggregated Deep Convolu-tional Features. IEEE Trans Biomed Eng. 2019;66(4):1006–16.
Google Scholar
 
 
49.
Kumar, V, Choudhury, T. Real-Time Recognition of Malignant Skin Lesions using Ensemble Modeling. J Sci Ind Res. 2029;78:148–53.
Google Scholar
 
 
50.
Deep Learning Solutions for Skin Cancer Detection and Diagnosis. In: Learning and Analytics in Intelligent Systems [Internet]. Cham: Springer International Publishing; 2020; 159–82. Available from: http://link.springer.com/10.10....
Google Scholar
 
 
51.
Rezaoana N, Hossain MS, Andersson K. Detection and Classification of Skin Cancer by Using a Parallel CNN Model. In: 2020 IEEE Interna-tional Women in Engineering (WIE) Conference on Electrical and Com-puter Engineering (WIECON-ECE) [Internet]. Bhubaneswar, India: IEEE. 2020; 380–6. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
52.
Sedigh P, Sadeghian R, Masouleh MT. Generating Synthetic Medical Images by Using GAN to Improve CNN Performance in Skin Cancer Classification. In: 2019 7th International Conference on Robotics and Mechatronics (ICRoM) [Internet]. Tehran. Iran: IEEE. 2019; 497–502. Available from: https://ieeexplore.ieee.org/do....
Google Scholar
 
 
53.
Winkler JK, Blum A, Kommoss K, Enk A, Toberer F, Rosenberger A, Haenssle HA. Assessment of diagnostic performance of dermatologists cooperating with a convolutional neural network in a prospective clinical study: human with machine. JAMA dermatology. 2023; 159(6): 621-627.
Google Scholar
 
 
Share
Send by email
RELATED ARTICLE
Bearing Fault Detection and Diagnosis Based on Densely Connected Convolutional Networks
Application of Convolutional Gated Recurrent Units U-Net for Distinguishing between Retinitis Pigmentosa and Cone–Rod Dystrophy
Automatic Method of Macular Diseases Detection Using Deep CNN-GRU Network in OCT Images
Indexes
 
Keywords index
Topics index
Authors index
eISSN:2300-5319
ISSN:1898-4088
Journals System - logo
© 2006-2026 Journal hosting platform by Bentus
Scroll to top