Artificial Intelligence–Based Classification of Monkeypox Disease from Skin Lesion Images

Öz

Monkeypox posed a significant global outbreak risk in 2022, leading the World Health Organization (WHO) and national authorities to implement rapid preventive measures. Although laboratory tests are primarily used for diagnosis, the high transmissibility of the disease highlights the need for supportive diagnostic approaches. In this study, a dataset consisting of 1,259 skin images, including 510 monkeypox and 749 non-monkeypox cases obtained from open sources, was expanded to 8,533 images through data augmentation techniques. Eight pre-trained deep learning architectures were trained and evaluated for the classification of monkeypox and non-monkeypox skin images. Based on the experimental results, a new stacking-based ensemble learning model was developed by combining feature vectors extracted from the high-performing DenseNet169, DenseNet201, and Xception architectures, achieving an accuracy of 99.30% on the test dataset. Detailed performance analyses were conducted, and Gradient-weighted Class Activation Mapping (Grad-CAM) was utilized to visualize the regions of interest influencing the model’s classification decisions

Anahtar Kelimeler

• Monkeypox
• Deep learning
• Ensemble learning
• Grad-CAM
• Image classification
• Transfer learning

Etik Beyan

We hereby declare that there is no conflict of interest, ethical misconduct, or any situation or process that may constitute an ethical violation among the authors or with any other individuals or institutions in this study.

Maymun Çiçeği Hastalığının Cilt Lezyon Görüntüleri Üzerinden Yapay Zekâ Algoritmaları İle Sınıflandırılması

Öz

Maymun çiçeği hastalığı, 2022 yılında küresel ölçekte ciddi bir salgın riski oluşturarak Dünya Sağlık Örgütü (DSÖ) ve ulusal otoritelerin hızlı önlem süreçleri geliştirmesine neden olmuştur. Teşhis aşamasında genellikle laboratuvar testleri kullanılmasına rağmen, hastalığın yüksek bulaşıcılığı nedeniyle destekleyici tanı yöntemlerine ihtiyaç duyulmaktadır. Bu çalışmada, açık kaynaklardan elde edilen 510 maymun çiçeği ve 749 maymun çiçeği olmayan cilt görüntüsünden oluşan toplam 1259 görsel, veri artırma yöntemleriyle zenginleştirilerek 8.533 görüntüden oluşan bir veri setine dönüştürülmüştür. Maymun çiçeği enfeksiyonu taşıyan ve taşımayan cilt görsellerinin sınıflandırılmasında sekiz farklı önceden eğitilmiş derin öğrenme mimarisi eğitilmiş ve test edilmiştir. Elde edilen sonuçlara göre, yüksek doğruluk oranlarına sahip DenseNet169, DenseNet201 ve Xception modellerinden çıkarılan özellik vektörleri birleştirilerek, yığınlama temelli topluluk öğrenmesi yaklaşımıyla yeni bir model tasarlanmış ve test veri seti üzerinde %99.30 doğruluğa ulaşmıştır. Modelin detaylı performans analizleri sunulmuş ve sınıflandırma yaparken odaklandığı bölgeleri tespit etmek için de Gradyan Ağırlıklı Sınıf Aktivasyon Haritalandırma (Grad-CAM) kullanılmıştır.

Anahtar Kelimeler

• Maymun çiçeği
• Derin öğrenme
• Topluluk öğrenmesi
• Grad-CAM
• Görüntü sınıflandırma
• Transfer öğrenme.

Etik Beyan

Çalışmamızda yazarlar arasında ya da diğer kişilerle veya kurumlarla herhangi bir çıkar çatışması , etik ihlali olabilecek bir durum - süreç bulunmadığını beyan ederiz.

Kaynakça

1. [1] Guarner, J., Del Rio, C. and Malani, P. N., “Monkeypox in 2022-What Clinicians Need to Know”. JAMA, 328(2), 139–140. https://doi.org/10.1001/jama.2022.10802 (2022).
2. [2] Kantele, A., Chickering, K., Vapalahti, O. and Rimoin, A.W., “Emerging diseases—the monkeypox epidemic in the Democratic Republic of the Congo” Clinical Microbiology and Infection, Volume 22, Issue 8, 658 – 659, 2016.
3. [3] Vaughan, A. M., Cenciarelli, O., Colombe, S., Alves de Sousa, L., Fischer, N., Gossner, C. M., Pires, J., Scardina, G., Aspelund, G., Avercenko, M., Bengtsson, S., Blomquist, P., Caraglia, A., Chazelle, E., Cohen, O., Díaz, A., Dillon, C., Dontsenko, I., Kotkavaara, K., Fafangel, M. and Haussig, J. M., “A large multi-country outbreak of monkeypox across 41 countries in the WHO European Region, 7 March to 23 August 2022” , Eurosurveillance, 27(36). https://doi.org/10.2807/1560-7917.ES.2022.27.36.2200920, 2022.
4. [4] Nakhaie, M., Arefinia, N., Charostad, J., Bashash, D., Abdolvahab, M.H. and Zarei, M., “Monkeypox virus diagnosis and laboratory testing,” Rev Med Virol, vol. 33, no. 1, Jan. 2023. https://doi.org/10.1002/rmv.2404
5. [5] Qu, J., Zhang, X., Liu, K., Li, Y., Wang, T., Fang, Z., Chen, C., Tan, X., Lin, Y., Xu, Q., Yang, Y., Wang, W., Huang, M., Guo, S., Chen, Z., Rao, W., Shi, X. and Peng, B., “A Comparative Evaluation of Three Diagnostic Assays for the Detection of Human Monkeypox”, Viruses, 16(8), 1286. https://doi.org/10.3390/v16081286
6. [6] Oladipo, E.K., Ajayi, A.F., Odeyemi, A.N., Akindiya, O.E., Adebayo, E.T., Oguntomi, A.S., Oyewole, M.P., Jimah, E.M., Oladipo, A.A., Ariyo, O.E., Oladipo, B.B. and Oloke, J.K., “Laboratory diagnosis of COVID-19 in Africa: availability, challenges and implications”. Drug Discoveries Therapeutics, vol. 14, no. 4, pp. 153–160 https://doi.org/10.5582/DDT.2020.03067
7. [7] Zhao, W., Jiang, W. and Qiu, X., “Deep learning for COVID-19 detection based on CT images”, Sci Rep 11, 14353 (2021). https://doi.org/10.1038/s41598-021-93832-2
8. [8] Jaradat, A. S., Al Mamlook, R. E., Almakayeel, N., Alharbe, N., Almuflih, A. S., Nasayreh, A., Gharaibeh, H., Gharaibeh, M., Gharaibeh, A. and Bzizi, H., “Automated Monkeypox Skin Lesion Detection Using Deep Learning and Transfer Learning Techniques” , International journal of environmental research and public health, 20(5), 4422, 2023. https://doi.org/10.3390/ijerph20054422

9. [9] Ali, S.N., Ahmed, T., Paul, J., Jahan, T., Sani, S.M.S., Noor, N. and Hasan, T., “Monkeypox Skin Lesion Detection Using Deep Learning Models: A Feasibility Study” , arXiv:2207.03342v1 [cs.CV], 2022. Dataset URL: https://github.com/ShamsNafisaAli/Monkeypox-Skin-Lesion-Dataset , Erişim Tarihi: 21 Aralık 2024
10. [10] Siar, M. and Teshnehlab, M., "Brain Tumor Detection Using Deep Neural Network and Machine Learning Algorithm," , 2019 9th International Conference on Computer and Knowledge Engineering (ICCKE), Mashhad, Iran, 2019, pp. 363-368, doi: 10.1109/ICCKE48569.2019.8964846
11. [11] Ahamed, A., Politza, A.J., Liu, T., Khalid, M.A.U., Zhang, H. and Guan, W., “CRISPR-based strategies for sample-to-answer monkeypox detection: current status and emerging opportunities” , Nanotechnology, Volume 36, Number 4, DOI 10.1088/1361-6528/ad892b
12. [12] McCollum, A. M. and Damon, I. K., “Human monkeypox”, Clinical infectious diseases : an official publication of the Infectious Diseases Society of America, 58(2), 260–267, 2014. https://doi.org/10.1093/cid/cit703
13. [13] Petti, C.A., Polage, C.R., Quinn, T.C., Ronald, A.R. and Sande, M.A., “Laboratory Medicine in Africa: A Barrier to Effective Health Care,” Clinical Infectious Diseases, vol. 42, no. 3, pp. 377–382, Feb. 2006, https://doi.org/10.1086/499363
14. [14] Ahsan, M. M., Uddin, M. R., Farjana, M., Sakib, A. N., Momin, K. A. and Luna, S. A., “Image Data collection and implementation of deep learning-based model in detecting Monkeypox disease using modified VGG16”, arXiv:2206.01862v1 [eess.IV] 4 Jun 2022. Dataset URL: https://github.com/mahsan2/Monkeypox-dataset-2022 , Erişim Tarihi: 21 Aralık 2024
15. [15] Bala, D., Hossain, M. S., Hossain, M. A., Abdullah, M. I., Rahman, M. M., Manavalan, B., Gu, N., Islam, M. S. and Huang, Z., “MonkeyNet : a robust deep convolutional neural network for monkeypox disease detection and classification” , Neural Networks, 161,757-775, 2023. doi.org/10.1016/j.neunet.2023.02.022, Dataset URL: https://data.mendeley.com/datasets/r9bfpnvyxr/6 Erişim Tarihi: 21 Aralık 2024
16. [16] Sahin, V.H., Oztel, I. and Yolcu Oztel, G., “Human Monkeypox Classification from Skin Lesion Images with Deep Pre-trained Network using Mobile Application” , J Med Syst 46, 79 ,2022. https://doi.org/10.1007/s10916-022-01863-7
17. [17] Sitaula, C. and Shahi, T.B., “Monkeypox Virus Detection Using Pre-trained Deep Learning-based Approaches” , J Med Syst 46, 78, 2022. https://doi.org/10.1007/s10916-022-01868-2
18. [18] Altun, M., Gürüler, H., Özkaraca, O., Khan, F., Khan, J. and Lee, Y., “Monkeypox Detection Using CNN with Transfer Learning” , Sensors, 23(4), 1783, 2023. https://doi.org/10.3390/s23041783
19. [19] Taspinar, Y. S., Cinar, I., Kursun, R. and Koklu, M., “Monkeypox Skin Lesion Detection with Deep Learning Models and Development of Its Mobile Application”. International Journal of Research in Engineering and Science, ISSN (Online): 2320-9364, ISSN (Print): 2320-9356, Vol(12), 1, pp 273-285, 2024.
20. [20] Akram, A., Jamjoom, A.A., Innab, N., Almujally, N.A., Umer5, M., Alsubai, S. and Fimiani, G., “SkinMarkNet: an automated approach for prediction of monkeyPox using image data augmentation with deep ensemble learning models” , Multimed Tools Appl 84, 20177–20193 (2025). https://doi.org/10.1007/s11042-024-19862-w
21. [21] Maqsood, S., Damaševičius, R., Shahid, S. and Forkert, N.D., “MOX-NET: Multi-stage deep hybrid feature fusion and selection framework for monkeypox classification,” Expert Syst Appl, vol. 255, p. 124584, Dec. 2024, https://doi.org/10.1016/j.eswa.2024.124584
22. [22] Asif, S., Zhao, M., Tang, F., Zhu, Y. And Zhao, B. “Metaheuristics optimization-based ensemble of deep neural networks for Mpox disease detection,” Neural Networks, vol. 167, pp. 342–359, Oct. 2023, https://doi.org/10.1016/j.neunet.2023.08.035
23. [23] Thieme, A.H., Zheng, Y., Machiraju, G. et al. , “A deep-learning algorithm to classify skin lesions from mpox virus infection”, Nat Med 29, 738–747 , 2023.
24. [24] Ahsan, M.M., Alam, T.E., Haque, M.A., Ali, S., Rifat, R.H., Nafi, A.N., Hossain, M. and Islam, K., “Enhancing Monkeypox diagnosis and explanation through modified transfer learning, vision transformers, and federated learning,” Inform Med Unlocked, vol. 45, p. 101449, 2024.
25. [25] Tripathi, M.K., Nath, A., Singh, T.P. et al., “Evolving scenario of big data and Artificial Intelligence (AI) in drug discovery” , Mol Divers 25, 1439–1460 , 2021. https://doi.org/10.1007/s11030-021-10256-w
26. [26] (2023). Dermnet. Accessed: Sep. 2023. [Online]. Available: http://www.dermnet.com/ , Dataset URL: https://dermnetnz.org/images/mpox-images Erişim Tarihi: 21 Aralık 2024
27. [27] (2022). A. Nasayrah, “Data Monkeypox,” Kaggle.com. (accessed: Sep. 8, 2025) , DatasetURL:https://www.kaggle.com/datasets/ahmadnasayrah/data-monkeypox , Erişim Tarihi: 21 Aralık 2024
28. [28] Wahid, M., Ahmad, N., Zafar, M.H. and Khan, S., “On combining MD5 for image authentication using LSB substitution in selected pixels,” International Conference on Engineering and Emerging Technologies (ICEET), IEEE, pp. 1–6, ISBN 9781538621707, 2018.
29. [29] Malviya, A. V. and Ladhake, S. A., “Region duplication detection using color histogram and moments in digital image” , International Conference on Inventive Computation Technologies (ICICT), Vol.1, pp. 1-4, 2016.
30. [30] Wong, S.C., Gatt, A., Stamatescu, V. and McDonnell, M.D., “Understanding Data Augmentation for Classification: When to Warp?” , nternational Conference on Digital Image Computing: Techniques and Applications (DICTA), IEEE, 2016, pp. 1-6, doi: 10.1109/DICTA.2016.7797091.
31. [31] Perez, L. and Wang, J. “The Effectiveness of Data Augmentation in Image Classification using Deep Learning,” arXiv:1712.04621v1 [cs.CV] 13 Dec 2017.
32. [32] Sönmez, D., “Geri yayılım algoritması’na matematiksel yaklaşım,” DerinOglenme.com, Jun. 28, 2018. [Online]. Available: “https://www.derinogrenme.com/2018/06/28/geri-yayilim-algoritmasinamatematiksel-yaklasim/.
33. [33] LeCun, Y., Bengio, Y. and Hinton, G., “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, May 2015.
34. [34] LeCun, Y., Bottou, L., Bengio, Y. And Haffner, P., “Gradient-based learning applied to document recognition,” Proceedings of the IEEE, vol. 86, no. 11, pp. 2278–2324, 1998. 10.1109/5.726791
35. [35] Goodfellow, I., Bengio, Y. and Courville, A., “Deep learning” , Cambridge, MA, USA: MIT Press, 2017.
36. [36] Zafar, A., Saba, N., Arshad, A., Alabrah, A., Riaz, S., Suleman, M., Zafar, S. and Nadeem, M., “Convolutional Neural Networks: A Comprehensive Evaluation and Benchmarking of Pooling Layer Variants” , Symmetry, 16(11), 1516, 2024. https://doi.org/10.3390/sym16111516
37. [37] Yao, J., “Vehicle Classification Enhancement Through Data Augmentation and Model Fusion: A Study of Deep Learning Approaches” , 4th International Conference on Digital Society and Intelligent Systems (DSInS), IEEE, pp. 449–453, 2024. DOI: 10.1109/DSInS64146.2024.10992194
38. [38] Kunc, V. And Kléma, J., “Three Decades of Activations: A Comprehensive Survey of 400 Activation Functions for Neural Networks” , arXiv:2402.09092v1 [cs.LG] 14 Feb 2024.