Cilt Kanseri Görüntülerinde Gürültü Temizliği ve Lezyonun Dört Sınıfa Ayrılması

Yıl 2024, Cilt: 24 Sayı: 2, 284 - 293, 29.04.2024

Cihan Akyel Nursal Arıcı

https://doi.org/10.35414/akufemubid.1211510

Öz

Günümüzde cilt kanseri çevresel koşulların da etkisiyle artış göstermektedir. Cilt kanserinin birçok farklı türü olmasına rağmen melanom (MEL) kötü huylu ve en ölümcül olanıdır. Bazal hücre karsinomu (BHK) ve skuamöz hücre karsinomu (SHK) cilt kanserleri de diğer organlara yayılım eğilimi gösterebilmektedirler. Cilt kanserinde erken teşhis tedavi sürecinde çok önemlidir. Cilt kanseri renk geçişleri, yapısal durumu gibi özelliklere bakılarak sınıflandırılabilmektedir. Kanser teşhisinde derin öğrenme ve görüntü işleme algoritmalarının kullanımı yüksek başarı oranı ve insan hatasını bertaraf etmesinden dolayı kullanımı yaygınlaşmaktadır. Lezyon görüntülerinde bulunan kıl, mürekkep izi gibi gürültüler lezyonun bu yöntemlerle sınıflandırılmasında başarıyı düşürmektedir. Çalışmada LinkNetRCB7 modeli ve görüntü işleme algoritmaları ile lezyon görüntülerinde gürültü temizliği yapılmıştır. Bu aşamada %97 eğitim başarısı elde edilmiştir. Sınıflama aşamasında çalışmada BHK, SHK, MEL ve iyi huylu olmak üzere görüntüler ISIC 2019’a ait veri seti ile dört sınıfa ayrılmıştır. Bu aşamada %94.87 eğitim başarısı gözlemlenmiştir.

Anahtar Kelimeler

Derin öğrenme, Cilt kanseri, Sınıflama, LinkNet, ResNet, Classification

Noise Removal in Skin Cancer Images and Classification of Lesions into Four Classes

Öz

Today, skin cancer is increasing with the effect of environmental conditions. Although there are many different types of skin cancer, melanoma (MEL) is the most malignant and the most deadly. Basal cell carcinoma (BCC) and squamous cell carcinoma (SHC) skin cancers may also tend to spread to other organs. Early diagnosis of skin cancer is very important in the treatment process. Skin cancer can be classified by looking at features such as color transitions and structural status. The use of deep learning and image processing algorithms in cancer diagnosis is becoming widespread due to its high success rate and elimination of human error. Noises such as hair and ink traces in the lesion images reduce the success in classifying the lesion with these methods. In the study, noise cleaning was performed on lesion images with the LinkNetRCB7 model and image processing algorithms. At this stage, 97% educational success was achieved. In the classification phase, the images were divided into four classes with the data set of ISIC 2019: BHK, SHK, MEL and benign. At this stage, 94.87% educational success was observed.

Anahtar Kelimeler

Deep learning, Skin cancer, Classification, LinkNet, ResNet

Kaynakça

• Akyel, C., 2022. Görüntü işleme ve derin öğrenme yöntemleri ile cilt kanseri teşhisi için karar destek sisteminin geliştirilmesi. Doktora Tezi, Gazi Üniversitesi Bilişim Enstitüsü, Ankara, 79.
• Akyel, C., and ARICI, N., 2022. Hair Removal and Lesion Segmentation with FCN8-ResNetC and Image Processing in Images of Skin Cancer. Bilişim Teknolojileri Dergisi, 15(2), 231-238. https://doi.org/10.17671/gazibtd.1060330
• Akyel, C., and ARICI, N., 2022. LinkNet-B7: Noise Removal and Lesion Segmentation in Images of Skin Cancer. Mathematics, 10(5), 736-751. https://doi.org/10.3390/math10050736
• Bardou, D., Bouaziz, H., Lv, L., and Zhand, T., 2021. Hair removal in dermoscopy images using variational autoencoders. Skin Research Technology, 28, 445-454. https://doi.org/10.1111/srt.13145
• Bassel, A., Abdulkareem, A. B., and Alyasseri, Z. A. A., 2022. Automatic Malignant and Benign Skin Cancer Classification Using a Hybrid Deep Learning Approach. Diagnostics, 12(2472), 1-15. https://doi.org/10.3390/diagnostics12102472
• Cassidy, B., Kendrick, C., Brodzicki, A., Jaworek-Korjakowska, J., and Yap, M. H., 2022. Analysis of the ISIC image datasets: Usage benchmarks and recommendations. Medical Image Analysis, 75, 1–15. https://doi.org/10.1016/j.media.2021.102305
• Codella N., Rotemberg, V., Tschandl, P., Celebi, M. E., Dusza, S., Gutman, D:, Helba, B., Kalloo, A., Liopyris, K., Marchetti, M., Kittler, H., Halpern, A., 2018. Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC), Computer Vision and Pattern Recognition, 2018, 1-12. https://doi.org/10.48550/arXiv.1902.03368
• Grignaffini, F., Barbuto, F., Piazzo, L., Troiano, M., Simeoni, P., Mangini, F., Pellacani, G., Cantisani, C., and Frezza, F., 2022. Machine Learning Approaches for Skin Cancer Classification from Dermoscopic Images: A Systematic Review. Algorithms, 15, 1-30. https://doi.org/10.3390/a15110438
• Indraswari, R., Rokhana, R., and Herulambang, W., (2022). Melanom image classification based on MobileNetV2 network. Procedia Computer Science, 197, 198–207. https://doi.org/10.1016/j.procs.2021.12.132
• Jaisakthi S. M., Mirunalini, P., Chandrabose, A.,and Rajagopal, A., 2022. Classification of skin cancer from dermoscopic images using deep neural network architectures. Multimedia Tools and Applications, 2022,1-16. https://doi.org/10.1007/s11042-022-13847-3
• Kahia, M., Echtioui, A., Kallel, F., and Hamida, A. B., 2022. Skin Cancer Classification using Deep Learning Models. ICAART 2022, 1, 554-560. https://doi.org/10.5220/0010976400003116
• Kaur, R., Gholamhosseini, H., Sinha, R., and Lindén, M., 2022. Melanom Classification Using a Novel Deep Convolutional Neural Network with Dermoscopic Images. Sensors, 22(3), 1-15. https://doi.org/10.3390/s22031134
• Kausar, N., Hameed, A., Sattar, M., Ashraf, R., Imran, A. S., Abidin, M. Z., and Ali, A., 2021. Multiclass Skin Cancer Classification Using Ensemble of Fine-Tuned Deep Learning Models. Applied Sciences, 11, 1-20. https://doi.org/10.3390/app112210593
• Kaya, B., and Önal, M., 2021. COVID-19 Tespiti için Akciğer BT Görüntülerinin Bölütlenmesi. Avrupa Bilim Ve Teknoloji Dergisi, 28, 1296-1303. https://doi.org/10.31590/ejosat.1015061
• Lee, T., Ng, V., Gallagher, R., Coldman, A., and McLean, D., 1997. Dullrazor®: A software approach to hair removal from images, Computers in Biology and Medicine, 27(6), 533-543. https://doi.org/10.1016/S0010-4825(97)00020-6
• Lee, J. R. H., Pavlova, M., Famouri, M., and Wong, A., 2022. S Cancer-Net SCa: tailored deep neural network designs for detection of skin cancer from dermoscopy images. BMC Medical Imaging, 22(143), 1-12. https://doi.org/10.1186/s12880-022-00871-w
• Li, W., Raj, A. N. J., Tjahjadi, T., and Zhuang, Z., 2021. Digital hair removal by deep learning for skin lesion segmentation. Pattern Recognition, 117, 1-15. https://doi.org/10.1016/j.patcog.2021.107994
• Lu, X., and Zadeh, Y. A. F. A., 2022. Deep Learning-Based Classification for Melanoma Detection Using XceptionNet. Hindawi Journal of Healthcare Engineering, 2022, 1-10. https://doi.org/10.1155/2022/2196096
• Lopez A. R., Giro-i-Nieto X., Burdick J., Marques O., (2017). Skin Lesion Classification From Dermoscopic Images Using Deep Learning Techniques, 13th IASTED International Conference on Biomedical Engineering (BioMed), Manhattan, New York, U.S: Institute of Electrical and Electronics Engineers (IEEE), 49–54.
• Machlin, J., Machlin S. R., Ekwueme, D. U., Yabrof K. R., 2015. Prevalence and costs of skin cancer treatment in the U.S. American Journal of Preventive Medicine, 48, 183–187. https://doi.org/10.1016/j.amepre.2014.08.036
• Powers, D., Powers A., 2011. Evaluation: From precision, recall and F-measure to ROC, informedness, markedness & correlation. Journal of Machine Learning Technologies, 2, 2229-3981. https://doi.org/10.9735/2229-3981
• Rehman, M., Ahmed, F., Alsuhibany, S. A., Jamal, S. S., Ali, M. Z., and Ahmad, J., 2022. Classification of Skin Cancer Lesions Using Explainable Deep Learning. Sensors, 22, 1–14. https://doi.org/10.3390/s22186915
• Rotemberg, V., Kurtansky, N., Betz-Stablein, B., Caffery, L., Chousakos, E., Codella, N., Combalia, M., Dusza, S., Guitera, P., Gutman, D., Halpern, A., Helba, B., Kittler, H., Kose, K., Langer, S., Lioprys, K., Malvehy, J., Musthaq, S., Nanda, J., Reiter, O., Shih, G., Stratigos, A., Tschandl, P., Weber, J. and Soyer, P., 2021. A patient-centric dataset of images and metadata for identifying melanomas using clinical context. Scientific Data, 8(34), 1-8. https://doi.org/10.1038/s41597-021-00865-3
• Sokolova, M., Japkowicz, N., Szpakowicz, S., 2006. Beyond Accuracy, F-Score and ROC: A Family of Discriminant Measures for Performance Evaluation. Advances in Artificial Intelligence, 4304, 1015-1021. https://doi.org/10.1007/11941439_114
• Suresh, A., and Seeja, R., 2019. Deep Learning Based Skin Lesion Segmentation and Classification of Melanom Using Support Vector Machine (SVM). Asian Pacific Journal of Cancer Prevention, 20(5), 1555-1561, https://doi.org/10.31557/APJCP.2019.20.5.1555
• Şahin, N. ve Alpaslan, N., 2020. Seg-Net Mimarisi Kullanılarak Cilt Lezyon Bölütleme Performansının İyileştirilmesi. Avrupa Bilim ve Teknoloji Dergisi, special issue, 40-45. https://doi.org/10.31590/ejosat.araconf6
• Talavera-Martínez, L., Bibiloni, P., and González-Hidalgo, M., 2020. Hair Segmentation and Removal in Dermoscopic Images Using Deep Learning. IEEE Access, 9, 2694–2704. https://doi.org/10.1109/ACCESS.2020.3047258
• Tschandl P., Rosendahl C. And Kittler H., 2018. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Scientific Data, 5, 1-9. https://doi.org/10.1038/sdata.2018.161
• Tan, X., Lai, S., and Zhang, M., 2014. Green Channel Guiding Denoising on Bayer Image. The Scientific World Journal, 2014, 1-9. https://doi.org/10.1155/2014/979081