208214

Meme Ultrason Görüntülerinde Kanser Hücre Segmentasyonu için Yeni Bir FCN Modeli

Meme kanseri kadınlar arasında yaygın bir hastalıktır. Zamanında teşhis ve uygun müdahale, iyileşme beklentilerini önemli ölçüde artırır ve hastalığın ilerlemesini engeller. Meme kanserinin görsel görüntüleri, meme dokusunu kötü huylu ve kötü huylu olmayan alanlara ayırmaya hizmet ederek, kötü huylu hücrelerin ayırt edilmesi, tümör boyutlarının değerlendirilmesi ve neoplazmın evrelendirilmesi dahil olmak üzere çok önemli bilgiler sağlar. Meme kanseri taramaları, özellikle genç ve hamile kadınlar grubundaki hassasiyetleri iyileştirmek için yürürlüğe konmuştur. Bununla birlikte, radyografik tasvirlerin incelenmesinin yanlışlıkla bazı incelikleri gözden kaçırabileceği durumlar da vardır. Gelişmiş yapay zeka paradigmaları, sofistike hesaplama metodolojileri ile birleştiğinde, daha yüksek hassasiyette sonuçlar elde etmek için kullanılmaktadır. Bu bağlamda, ultrason teknolojisi tarafından kolaylaştırılan segmentasyon metodolojisi çok önemli bir müdahale olarak ortaya çıkmaktadır. Mevcut araştırmada, sapmaların tanımlanması ve kanserojen bölgelerin belirlenmesi için U-Net ve yenilikçi bir Fully Convolutional Network mimarisinden yararlanılmıştır. Bu çalışma kapsamında önerilen Fully Convolutional Network mimarisi, test görüntülerinde %77,2 MeanIoU, %71,6 precision, %77,7 recall ve %74,5 F1 skoru elde edilmiştir. U-Net modeli ile %76,4 MeanIoU, %67,6 precision, %80,4 recall ve %73 F1 skoru elde edilmiştir. Bulgular, önerilen Fully Convolutional Network mimarisinin U-Net modeline göre daha iyi performans sergilediğini ortaya koymuştur. Bu sonuçlar, meme kanseri teşhisi ve tedavisi için segmentasyon işleminin önemini vurgulamakta ve önerilen Fully Convolutional Network mimarisinin U-Net mimarisinden daha başarılı olduğunu ortaya koymaktadır.

Anahtar Kelimeler:

Fully Convolutional Network, U-Net, meme görüntüsü, kanserli hücre, segmentasyon

A New FCN Model for Cancer Cell Segmentation in Breast Ultrasound Images

Breast cancer stands as a widespread ailment among women. Timely diagnosis and appropriate intervention significantly enhance prospects of recovery and thwart disease progression. Visual representations of breast cancer serve to delineate breast tissue into malignant and non-malignant regions, offering crucial insights, including the discernment of malignant cells, assessment of tumor dimensions, and staging of the neoplasm. Breast cancer screenings are implemented to ameliorate sensitivities, especially within the cohort of young and pregnant women. However, instances exist wherein the examination of radiographic portrayals may inadvertently overlook certain intricacies. Advanced artificial intelligence paradigms, coupled with sophisticated computational methodologies, are harnessed to attain outcomes of heightened precision. In this context, the segmentation methodology facilitated by ultrasound technology emerges as a pivotal intervention. In the present study, U-Net and an innovative Fully Convolutional Network architecture are utilized to identify deviations and demarcate carcinogenic zones. Within the scope of this research, the proposed Fully Convolutional Network architecture achieves a MeanIoU of 77.2%, precision rates of 71.6%, recall rates of 77.7%, and an F1 score of 74.5% in testing images. Conversely, the U-Net model yields comparable performance metrics, manifesting a MeanIoU of 76.4%, precision of 67.6%, recall of 80.4%, and an F1 score of 73%. Notably, the findings underscore the superiority of the proposed Fully Convolutional Network paradigm over the U-Net architecture. These results accentuate the significance of segmentation methodologies in the realm of breast cancer diagnosis and treatment, thus affirming the superior efficacy of the suggested Fully Convolutional Network architecture relative to the U-Net architecture.

Keywords:

Fully Convolutional Network, U-Net, breast image, cancer cell, segmentation,

PDF

___

Loukas, C., Kostopoulos, S., Tanoglidi, A., Glotsos, D., Sfikas, C., Cavouras, D., 2013. Breast cancer characterization based on image classification of tissue sections visualized under low magnification. Computational and Mathematical Methods in Medicine, 2013, 7. doi:10.1155/2013/829461
Joy, J., Penhoet, E., & Petititti, D., 2005. Saving women’s lives: strategies for improving breast cancer detection and diagnosis. National Academies Press.
Veta, M., Pluim, J. P. W., Diest, P. J., Viergeve, M. A., 2014. Breast cancer histopathology image analysis: A review. IEEE Transactions on Biomedical Engineering, 61(5), 1400–1411.
Irshad, H., Veillard, A., Roux, L., Racoceanu, D., 2014. Methods for nuclei detection, segmentation, and classification in digital histopathology: A review-Current status and future potential. IEEE Reviews in Biomedical Engineering, 7, 97–114.
Ataş, M., Özdemi̇r, C., Ataş, İ., Ak, B., Özeroğlu, E., 2022. Biometric identification using panoramic dental radiographic images withfew-shot learning. TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES, 30(3), 1115–1126. doi:10.55730/1300-0632.3830
Dogan, Y., Keles, H. Y., 2022. Iterative facial image inpainting based on an encoder-generator architecture. Neural Computing & Applications, 34(12), 10001–10021. doi:10.1007/s00521-022-06987-y
Kılıç, Ş., Kaya, Y., Askerbeyli, İ., 2021. A new approach for human recognition through wearable sensor signals. Arabian Journal for Science and Engineering, 46, 4175-4189. Dogan, Y., 2023. A New Global Pooling Method for Deep Neural Networks: Global Average of Top-K Max-Pooling. Traitement du Signal, 40(2).
Motta, L. S., 2010. Obtenção automática da região de interesse em termogramas frontais da mama para o auxílio a detecção precoce de doenças. In Federal University Fluminense (UFF). Niterói (RJ, Brazil.
Marques, R. S., 2012. Segmentação automática das mamas em imagens térmicas. In Federal University Fluminense (UFF). Niterói (RJ, Brazil.
Baffa, M., Cheloni, D., Lattari, L., 2016. Segmentação Automática de Imagens Térmicas das Mamas Utilizando Limiarização com Refinamento Adaptativo. Anais Do XVI Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS 2016). Presented at the Simpósio Brasileiro de Computação Aplicada à Saúde, Brasil. doi:10.5753/sbcas.2016.9896
Marques RS, Conci A, Perez MG, Andaluz VH, Mejia TM. 2016. An approach for automatic segmentation of thermal imaging in computer aided diagnosis. IEEE Latin Am Trans. 14(4):1856–1865.
Melo, R., Conci, A., Vasconcelos, C. N., 2017. On the use of fully convolutional networks on evaluation of infrared breast image segmentations. In Anais do XVII Workshop de Informática Médica. São Paulo, Brazil: SBC.
Ozan, O., Jo, S., Le, F. L., Matthew, L., Mattias, H., Kazunari, M., … Steven, M., 2018. Kainz Bernhard Attention u-net: Learning where to look for the pancreas.
Zhou, Z., Rahman Siddiquee, M. M., Tajbakhsh, N., Liang, J., 2018. Unet++: A nested u-net architecture for medical image segmentation. In Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 20, 2018, Proceedings 4 (pp. 3-11). Springer International Publishing.
Olaf, R., Philipp, F., Thomas, B., 2015. Convolutional networks for biomedical image segmentation, International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 234–241). Springer.
Mendes, L. C., Rodrigues, E. O., Izidoro, S. C., Conci, A., & Liatsis, P., 2020. ROI extraction in thermographic breast images using genetic algorithms. 2020 International Conference on Systems, Signals and Image Processing (IWSSIP). Presented at the 2020 International Conference on Systems, Signals and Image Processing (IWSSIP), Niterói, Brazil. doi:10.1109/iwssip48289.2020.9145346
Baffa, M. de F. O., Coelho, A. M., Conci, A., 2021. Segmentação de Imagens Infravermelhas Para Detecção do Câncer de Mama Utilizando U-NET CNN. Anais Do XXI Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS 2021). Presented at the Simpósio Brasileiro de Computação Aplicada à Saúde, Brasil. doi:10.5753/sbcas.2021.16058
Yang, L., Xiuxiu, H., Jincao, Y., Tonghe, W., Lijing, W., Wei, L., … Xiaofeng, Y., 2021. Breast tumor segmentation in 3D automatic breast ultrasound using mask scoring R-CNN, Med. Med. Phys, 48(1), 204–214.
Al-Dhabyani, W., Gomaa, M., Khaled, H., & Fahmy, A., 2020. Dataset of breast ultrasound images. Data in Brief, 28(104863), 104863. doi:10.1016/j.dib.2019.
https://www.who.int/health-topics/cancer#tab=tab, (02.02.2023)
https://gco.iarc.fr/today, (02.02.2023)