A Comparison of Different Convolutional Neural Network Models for Skin Cancer Diagnosis

Yıl 2025, Cilt: 15 Sayı: 1, 25 - 38, 01.03.2025

İbrahim Aruk , Ahmet Nusret Toprak

https://doi.org/10.21597/jist.1575214

Öz

In recent years, a notable rise in the prevalence of skin cancer has been seen worldwide. Early and correct diagnosis of skin cancer improves treatment success rates and substantially enhances patients' quality of life. Traditional skin cancer diagnostic techniques generally depend on visual evaluations and include a subjective methodology. On the other hand, deep learning algorithms provide effective solutions to improve the accuracy and time of skin cancer diagnostics. This study evaluates the efficacy of advanced Convolutional Neural Network (CNN) models, including EfficientNet, VGG, Inception, DenseNet, and DarkNet, in the classification of skin cancer. Twenty-one CNN models were trained and extensively analyzed on the ISIC 2017 dataset utilizing data augmentation and transfer learning methodologies. The results of the study demonstrated that the EfficientNet-b0 model attained superior performance with an accuracy of 84.00%, precision of 83.63%, sensitivity of 74.96%, and an F1-score of 78.59%. This comprehensive study shows the efficacy of CNN-based models in skin cancer diagnosis and illustrates the promise of these algorithms for future research.

Anahtar Kelimeler

Skin cancer classification, Convolutional neural networks, Melanoma

Cilt Kanseri Tanısı için Farklı Evrişimsel Sinir Ağı Modellerinin Karşılaştırılması

Öz

Son yıllarda, dünya genelinde cilt kanseri görülme oranında önemli bir artış gözlemlenmektedir. Cilt kanserinin zamanında ve doğru bir şekilde teşhis edilmesi, tedavi başarı oranlarını artırmakta ve aynı zamanda hastaların yaşam kalitesinin iyileşmesine büyük katkı sağlamaktadır. Geleneksel cilt kanseri tanı yöntemleri genellikle görsel değerlendirmelere dayanmakta ve öznel bir yaklaşım içermektedir. Bununla birlikte, derin öğrenme algoritmaları, cilt kanseri teşhislerinin doğruluğunu ve verimliliğini artırmak için etkili çözümler sunmaktadır. Bu çalışmada, EfficientNet, VGG, Inception, DenseNet ve DarkNet gibi gelişmiş Evrişimsel Sinir Ağı (CNN) modellerinin cilt kanseri sınıflandırmasındaki performansları incelenmiştir. Toplamda yirmi bir CNN modeli, ISIC 2017 veri seti üzerinde, veri artırma ve transfer öğrenme teknikleri kullanılarak eğitilmiş ve detaylı bir şekilde değerlendirilmiştir. Deneysel sonuçlar, EfficientNet-b0 modelinin %84.00 doğruluk, %83.63 kesinlik, %74.96 duyarlılık ve %78.59 F1-skoru ile en yüksek performansı sergilediğini göstermiştir. Bu kapsamlı analiz, CNN tabanlı modellerin cilt kanseri teşhisindeki etkinliğini göstermekte ve gelecekteki araştırmalar için bu algoritmaların potansiyelini ortaya koymaktadır.

Anahtar Kelimeler

Cilt kanseri sınıflandırması, Evrişimsel sinir ağları, Melanom

Kaynakça

• Adegun, A. A. ve Viriri, S. (2020) “FCN-Based DenseNet Framework for Automated Detection and Classification of Skin Lesions in Dermoscopy Images”, IEEE Access, 8, 150377-150396.
• Al-masni, M. A., Kim, D. H. ve Kim, T. S. (2020) “Multiple skin lesions diagnostics via integrated deep convolutional networks for segmentation and classification”, Computer Methods and Programs in Biomedicine, 190.
• Ashraf, H., Waris, A., Ghafoor, M. F., Gilani, S. O. ve Niazi, I. K. (2022) “Melanoma segmentation using deep learning with test-time augmentations and conditional random fields”, Scientific Reports, 12(1).
• Chollet, F. (2016) “Xception: Deep Learning with Depthwise Separable Convolutions”.
• Codella, N. C. F., Gutman, D., Celebi, M. E., Helba, B., Marchetti, M. A., Dusza, S. W., Kalloo, A., Liopyris, K., Mishra, N., Kittler, H. ve Halpern, A. (2018) “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)”, 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), 168-172.
• Cong, S. ve Zhou, Y. (2023) “A review of convolutional neural network architectures and their optimizations”, Artificial Intelligence Review, 56(3), 1905-1969.
• Dhillon, A. ve Verma, G. K. (2020) “Convolutional neural network: a review of models, methodologies and applications to object detection”, Progress in Artificial Intelligence, 85-112.
• Dillshad, V., Khan, M. A., Nazir, M., Saidani, O., Alturki, N. ve Kadry, S. (2023) “D2LFS2Net: Multi-class skin lesion diagnosis using deep learning and variance-controlled Marine Predator optimisation: An application for precision medicine”, CAAI Transactions on Intelligence Technology, 1-16.
• Gajera, H. K., Nayak, D. R. ve Zaveri, M. A. (2023) “A comprehensive analysis of dermoscopy images for melanoma detection via deep CNN features”, Biomedical Signal Processing and Control, 79.
• Hameed, M., Zameer, A. ve Zahoor Raja, M. A. (2024) “A Comprehensive Systematic Review: Advancements in Skin Cancer Classification and Segmentation Using the ISIC Dataset”, CMES - Computer Modeling in Engineering and Sciences, 2131-2164.
• Hayat, S. N. (2024) “Skin Cancer Detection Approach Using Convolutional Neural Network Artificial Intelligence”, IJIIS: International Journal of Informatics and Information Systems, 7(2), 46-54.
• Hermosilla, P., Soto, R., Vega, E., Suazo, C. ve Ponce, J. (2024) “Skin Cancer Detection and Classification Using Neural Network Algorithms: A Systematic Review”, Diagnostics.
• Huang, G., Liu, Z., van der Maaten, L. ve Weinberger, K. Q. (2016) “Densely Connected Convolutional Networks”.
• Hussain, S. I. ve Toscano, E. (2024) “An Extensive Investigation into the Use of Machine Learning Tools and Deep Neural Networks for the Recognition of Skin Cancer: Challenges, Future Directions, and a Comprehensive Review”, Symmetry.
• Kaur, R., GholamHosseini, H., Sinha, R. ve Lindén, M. (2022) “Automatic lesion segmentation using atrous convolutional deep neural networks in dermoscopic skin cancer images”, BMC Medical Imaging, 22(1).
• Musthafa, M. M., T R, M., V, V. K. ve Guluwadi, S. (2024) “Enhanced skin cancer diagnosis using optimized CNN architecture and checkpoints for automated dermatological lesion classification”, BMC Medical Imaging, 24(1), 201.
• Naeem, A. ve Anees, T. (2024) “DVFNet: A deep feature fusion-based model for the multiclassification of skin cancer utilizing dermoscopy images”, PLoS ONE, 19(3 March), 1-27.
• Nancy, V. A. O., Prabhavathy, P., Arya, M. S. ve Ahamed, B. S. (2023) “Comparative study and analysis on skin cancer detection using machine learning and deep learning algorithms”, Multimedia Tools and Applications, 82(29), 45913-45957.
• Ozcan, T., Toprak, A. N., Aruk, I., Sahin, O. ve Ozcan, I. (2024) “Applications of deep learning techniques in healthcare systems: A review”, Journal of Clinical Practice & Research, 45(5).
• Raja Subramanian, R., Achuth, D., Shiridi Kumar, P., kumar Reddy, K. N., Amara, S. ve Chowdary, A. S. (2021) “Skin cancer classification using Convolutional neural networks”, Proceedings of the Confluence 2021: 11th International Conference on Cloud Computing, Data Science and Engineering, 13-19.
• Redmon, J. ve Farhadi, A. (2018) “YOLOv3: An Incremental Improvement”.
• Rezaoana, N., Hossain, M. S. ve Andersson, K. (2020) “Detection and Classification of Skin Cancer by Using a Parallel CNN Model”, Proceedings of 2020 IEEE International Women in Engineering (WIE) Conference on Electrical and Computer Engineering, WIECON-ECE 2020, 380-386.
• Shete, A. S., Sanjay Rane, A., Gaikwad, P. S. ve Patil, M. H. (2021) “DETECTION OF SKIN CANCER USING CNN ALGORITHM”, International Journal Of Advance Scientific Research And Engineering Trends, 6(5), 2456-0774.
• Siegel, R. L., Giaquinto, A. N. ve Jemal, A. (2024) “Cancer statistics, 2024”, CA: A Cancer Journal for Clinicians, 74(1), 12-49.
• Simonyan, K. ve Zisserman, A. (2014) “Very Deep Convolutional Networks for Large-Scale Image Recognition”.
• Subramanian, B., Muthusamy, S., Thangaraj, K., Panchal, H., Kasirajan, E., Marimuthu, A. ve Ravi, A. (2024) “A Novel Approach Using Transfer Learning Architectural Models Based Deep Learning Techniques for Identification and Classification of Malignant Skin Cancer”, Wireless Personal Communications, 134(4), 2183-2201.
• Szegedy, C., Vanhoucke, V., Ioffe, S. ve Shlens, J. (2016) Rethinking the Inception Architecture for Computer Vision.
• Tan, M. ve Le, Q. V. (2019) “EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks”.
• Tanna, R. ve Sharma, T. (2021) “Binary Classification of Melanoma Skin Cancer using SVM and CNN”, Proceedings - 2021 1st IEEE International Conference on Artificial Intelligence and Machine Vision, AIMV 2021.
• Toprak, A. N. ve Aruk, I. (2024) “A Hybrid Convolutional Neural Network Model for the Classification of Multi-Class Skin Cancer”, International Journal of Imaging Systems and Technology, 34(5), e23180.
• Wang, R., Chen, X., Wang, X., Wang, H., Qian, C., Yao, L. ve Zhang, K. (2024) “A novel approach for melanoma detection utilizing GAN synthesis and vision transformer”, Computers in Biology and Medicine, 176, 108572.
• Yilmaz, A., Kalebasi, M., Samoylenko, Y., Guvenilir, M. E. ve Uvet, H. (2021) “Benchmarking of Lightweight Deep Learning Architectures for Skin Cancer Classification using ISIC 2017 Dataset”.