GhostNet tabanlı derin öğrenme yaklaşımı ile cilt kanseri sınıflandırması

Yıl 2025, Cilt: 15 Sayı: 4, 1099 - 1111, 15.12.2025

Heba Almasani , Hayati Türe

https://doi.org/10.17714/gumusfenbil.1744530

Öz

Cilt kanseri, özellikle melanom, erken teşhis edildiğinde büyük ölçüde tedavi edilebilir olmasına rağmen, hâlâ kanserle ilişkili ölümlerin orantısız bir kısmına neden olmaktadır. Bu çalışma, GhostNet ailesine ait evrişimli sinir ağlarını (CNN) kullanarak dermoskopik görüntülerin malign veya benign olarak ikili sınıflandırmasını hedefleyen verimli bir derin öğrenme sistemi sunmaktadır. Eğitim için kamuya açık ISIC-2019 veri kümesi kullanılmış; bu veri kümesindeki aşırı sınıf dengesizliği, Sentetik Azınlık Aşırı Örnekleme Tekniği (SMOTE) ile Düzenlenmiş En Yakın Komşular (ENN) yönteminin birleşimi olan hibrit SMOTE-ENN yöntemiyle giderilmiştir. GhostNet’in üç varyantı (V1, V2, V3) uçtan uca şekilde ince ayar yapılarak eğitilmiş, adil bir model karşılaştırması için ön işleme, veri artırımı ve hiperparametreler sabit tutulmuştur. Değerlendirme metrikleri olarak doğruluk, kesinlik, duyarlılık, F1-skoru ve alıcı işletim karakteristik eğrisi altında kalan alan (AUC) kullanılmıştır. Derinlik-odaklı dikkat mekanizmasıyla güçlendirilmiş GhostNet V2, %95 doğruluk, malign sınıf için %94 duyarlılık, %94.3 F1-skoru ve 0.99 AUC ile en güçlü sonuçları vermiştir. GhostNetV2, gerçek zamanlı mobil donanım üzerinde çalışmayı engelleyecek kadar yüksek bir parametre sayısına sahip olmadan, hem V1 hem de V3 taban modellerinden daha iyi performans göstermiştir. Doğruluk, duyarlılık, F1-skoru, AUC ve geri çağırma sonuçları, hedefe yönelik dengesizlik düzeltmesi ile birleştirildiğinde, GhostNet gibi verimli mimarilerin daha ağır modellerin hesaplama gereksinimlerine ihtiyaç duymadan dermatolog seviyesinde duyarlılığa ulaşabileceğini ve bu nedenle birinci basamak sağlık hizmetlerinde veya kaynakların kısıtlı olduğu ortamlarda kullanılabilir olduğunu göstermektedir. Bu çalışma, diğer CNN modelleriyle yaptığımız önceki çalışmalarla tamamlayıcı niteliktedir ve modeller arası doğrudan karşılaştırmaya olanak tanımaktadır.

Anahtar Kelimeler

Dermoskopi görüntüleme , GhostNet , Melanom , Cilt kanseri , SMOTE-ENN

Proje Numarası

2

A deep learning approach powered by GhostNet for skin cancer classification

Öz

Skin cancer, especially melanoma, continues to cause a disproportionate percentage of cancer-related deaths despite its high curability when detected early. This work presents an efficient deep-learning system aimed at the binary classification of dermoscopic images as malignant or benign using the GhostNet family of convolutional neural networks (CNNs). The publicly available ISIC-2019 dataset was used for training its extreme class imbalance was corrected using a hybrid Synthetic Minority Over-sampling Technique coupled with Edited Nearest Neighbours (SMOTE-ENN). End-to-end fine-tuning of three variants of GhostNet (V1, V2, V3) was performed with preprocessing, augmentation, and hyper-parameters kept constant to ensure a fair model-to-model comparison. Evaluation metrics used were accuracy, precision, recall, F1-score, and the area under the receiver-operating-characteristic curve (AUC). GhostNet V2, augmented with depth-wise attention, gave the strongest results: 95% accuracy, 94% malignant class recall, F1-score of 94.3%, and an AUC of 0.99. GhostNetV2 performed better than both the baseline V1 and V3 without having a parameter count that would prevent real-time inference on mobile hardware. These results of Accuracy, sensitivity, F1-score, AUC and recall show that, when combined with targeted imbalance correction, efficient architectures such as GhostNet are capable of dermatologist-level sensitivity without the computational requirements of heavier models, and thus are feasible for point-of-care or resource-constrained settings. It is complementary to our previous work with other CNNs, allowing model-to-model comparison directly.

Anahtar Kelimeler

Dermoscopy imaging , GhostNet , Melanoma , Skin cancer , SMOTE-ENN

Proje Numarası

2

Kaynakça

• Abayomi-Alli, O. O., Dada, A. A., Adepoju, A. A., Ogunlusi, F., & Ekundayo, B. B. (2021). Malignant skin melanoma detection using image augmentation by oversampling and undersampling techniques. Turkish Journal of Electrical Engineering & Computer Sciences, 29(5), 2425–2441. https://doi.org/10.3906/elk-2103-56.
• Ali, S. M., Li, S., Ziaei, M. N., & Golkar, H. (2022). Skin cancer classification with deep learning: A systematic review. International Journal of Molecular Sciences, 23(19), 10701. https://doi.org/10.3390/ijms23191070.
• Almasani, H., & Ture, H. (2025). Improving skin lesion classification with pre-trained deep learning models. In IEEE SIU 2025 Conference, Şile, Türkiye.
• American Cancer Society. (2023). Cancer facts & figures 2023. American Cancer Society. https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annual-cancer-facts-and-figures/2023/2023-cancer-facts-and-figures.pdf.
• Batista, G. E. A. P. A., Prati, R. C., & Monard, M. C. (2004). A study of the behavior of several methods for balancing machine-learning training data. SIGKDD Explorations, 6(1), 20–29. https://doi.org/10.1145/1007730.1007735.
• Chatterjee, R., Lahiri, A., & Dey, M. (2022). A ResNet50-based approach for skin-lesion classification. In Proceedings of the IEEE International Conference on Computing, Communication and Networking Technologies (pp. 1–5). https://doi.org/10.1109/ICCCNT55788.2022.9945732.
• Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357. https://doi.org/10.1613/jair.953.
• Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017). Dermatologist-level classification of skin cancer with deep neural networks. Nature, 542(7639), 115–118. https://doi.org/10.1038/nature21056.
• Gessert, N., Nielsen, M., Shaikh, M., Werner, R., & Schlaefer, A. (2019). Skin lesion classification using ensembles of multi-resolution EfficientNets with meta data. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) ISIC 2019 Challenge Workshop (pp. 1–8). https://openaccess.thecvf.com/content_CVPRW_2019/html/ISIC/Gessert_Skin_Lesion_Classification_Using_Ensembles_CVPRW_2019_paper.html (Journal version: Artificial Intelligence in Medicine, 2020).
• Haenssle, H. A., Fink, C., Schneiderbauer, R., Toberer, F., Buhl, T., Blum, T., ... Klode, J. (2018). Man against machine: Diagnostic performance of a deep-learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists. Annals of Oncology, 29(8), 1836–1842. https://doi.org/10.1093/annonc/mdy166.
• Han, H., Wang, W.-Y., & Mao, B.-H. (2005). Borderline-SMOTE: A new over-sampling method in imbalanced data sets learning. In Lecture Notes in Computer Science (LNCS 3644): Advances in Intelligent Computing (pp. 878–887). Springer. https://doi.org/10.1007/11538059_91.
• Han, K., Wang, Y., Tian, Q., Guo, J., Xu, C., & Xu, C. (2020). GhostNet: More features from cheap operations. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp. 1580–1589). https://doi.org/10.1109/CVPR42600.2020.00165.
• Han, S., Pool, J., Tran, J., & Dally, W. J. (2015). Learning both weights and connections for efficient neural networks. In Advances in Neural Information Processing Systems, 28 (pp. 1135–1143).
• He, H., Bai, Y., Garcia, E. A., & Li, S. (2008). ADASYN: Adaptive synthetic sampling approach for imbalanced learning. Proceedings of the IEEE International Joint Conference on Neural Networks (IJCNN), 1322–1328. https://doi.org/10.1109/IJCNN.2008.4633969.
• He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 770–778). https://doi.org/10.1109/CVPR.2016.90
• Hinton, G., Vinyals, O., & Dean, J. (2015). Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531. https://arxiv.org/abs/1503.02531.
• Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., ... Adam, H. (2017). MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint. https://arxiv.org/abs/1704.04861.
• Howard, A., Sandler, M., Chu, G., Chen, L., Chen, B., Tan, M., ... Adam, H. (2019). Searching for MobileNetV3. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 1314–1324). https://doi.org/10.1109/ICCV.2019.00140.
• Husain, G., Chauhan, S. S., & Agarwal, M. (2025). SMOTE vs. SMOTE-ENN: A study on the performance of resampling algorithms for addressing class imbalance in regression models. Algorithms, 18(1), 37. https://doi.org/10.3390/a18010037.
• Khan, M., Jhanjhi, N. Z., & Imran, M. (2023). DenseNet-201 for skin-lesion classification. IEEE Access, 11, 20365–20375. https://doi.org/10.1109/ACCESS.2023.3244603.
• Kassem, M. A., Hosny, K. M., & Damaševičius, R. (2020). Skin lesions classification into eight classes for ISIC 2019 using deep convolutional neural network and transfer learning [Preprint]. ResearchGate. https://www.researchgate.net/publication/342325762.
• Kubat, M., & Matwin, S. (1997). Addressing the curse of imbalanced training sets: One-sided selection. In Proceedings of the 14th International Conference on Machine Learning (pp. 179–186). Morgan Kaufmann.
• Kumari, M., & Subbarao, N. (2022). A hybrid resampling algorithm (SMOTE and ENN)-based deep-learning models for identification of Marburg virus inhibitors. Future Medicinal Chemistry, 14(10), 701–715. https://doi.org/10.4155/fmc-2021-0285.
• Liu, Z., Hao, Z., Han, K., Tang, Y., & Wang, Y. (2024). GhostNetV3: Exploring the training strategies for compact models. arXiv preprint arXiv:2404.11202. https://arxiv.org/abs/2404.11202.
• Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., ... Sánchez, C. I. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42, 60–88. https://doi.org/10.1016/j.media.2017.07.005.
• Nguyen, V. D., Bui, N. D., & Do, H. K. (2022). Skin-lesion classification on imbalanced data using deep learning with soft attention. Sensors, 22(19), 7530. https://doi.org/10.3390/s22197530 .
• Pang, B., Chen, L., Tao, Q., Wang, E., & Yu, Y. (2024). GA-UNet: A lightweight Ghost and attention U-Net for medical image segmentation. Journal of Imaging Informatics in Medicine, 37(4), 1874–1888. https://doi.org/10.1007/s10278-024-01070-5.
• Rastegari, M., Ordonez, V., Redmon, J., & Farhadi, A. (2016). XNOR-Net: ImageNet classification using binary convolutional networks. In Computer Vision – ECCV 2016 (Vol. 9908, pp. 525–542). Springer. https://doi.org/10.1007/978-3-319-46493-0_32.
• Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L. (2018). MobileNetV2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 4510–4520). https://doi.org/10.1109/CVPR.2018.00474.
• Siegel, R. L., Miller, K. D., Fuchs, H. E., & Jemal, A. (2022). Cancer statistics, 2022. CA: A Cancer Journal for Clinicians, 72(1), 7–33. https://doi.org/10.3322/caac.21708.
• Tang, Y., Han, K., Guo, J., Xu, C., Xu, C., & Wang, Y. (2022). GhostNetV2: Enhance cheap operation with long-range attention. In Advances in Neural Information Processing Systems (NeurIPS 2022), Vol. 35. Proceedings page: https://proceedings.neurips.cc/paper_files/paper/2022/hash/40b60852a4abdaa696b5a1a78da34635-Abstract-Conference.html.
• Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks. In Proceedings of the 36th International Conference on Machine Learning (Vol. 97, pp. 6105–6114). PMLR.
• Tschandl, P., Rosendahl, C., & Kittler, H. (2018). The HAM10000 dataset: A large collection of multi-source dermatoscopic images of common pigmented skin lesions. Scientific Data, 5, 180161. https://doi.org/10.1038/sdata.2018.161.
• Wilson, D. L. (1972). Asymptotic properties of nearest neighbor rules using edited data. IEEE Transactions on Systems, Man, and Cybernetics, 2(3), 408–421. https://doi.org/10.1109/TSMC.1972.4309137.