ULTRASON GÖRÜNTÜLERİNDE TİROİD NODÜLLERİNİN SEGMENTASYONU

Abstract

Ülkemizde ve dünyada tiroid kanser miktarının yaygınlaşması neticesinde tiroid kanserlerinin tespit edilebilmesi için bilgisayar destekli farklı çalışmaların yapılması literatüre önemli bir katkı sağlamaktadır. Özellikle yapay zeka ve görüntü işleme konuları bu alanda sıklıkla kullanılan bir yöntemdir. Bunun nedeni, girişimsel olmayan yapısı, erişilebilirliği ve nodüllerin morfolojik özellikleri hakkında değerli bilgiler sağlayabilmesidir. Son yıllarda, tıbbi görüntülemede segmentasyon algoritmaları, tanısal doğruluğu artırma potansiyelleri nedeniyle büyük ilgi görmüştür. Tiroid nodüllerinin doğru segmentasyonu, yapay zeka destekli klinik karar destek sistemlerinin tiroid kanserinin tespiti ve teşhisi için geliştirilmesinde kritik bir ilk adımdır. Bu çalışmada tiroid nodüllerinin tespit edilebilmesi için yenilikçi bazı yöntemler kullanılmıştır. Örnek segmentasyonunda YOLOv5-Small algoritması ile doppler görüntüleri hariç tutulduğunda %79 dice skoru sağlanmıştır, sonrasında doppler görüntülerini içeren veri setinde YOLOv5-Large algoritması ile %91 test dice skoru elde edilmiştir. Semantik segmentasyonda Attention Unet++ ve Manet algoritması kullanılarak, doppler görüntüleri hariç tutulduğunda %89 test dice skoru elde edilirken, doppler görüntülerini içeren veri setinde %91 test dice skoruna ulaşılmıştır. Böylece normal şartlarda hekimler tarafından hariç tutulan görüntülerin de bilgisayarlı görüntü işleme sürecinde daha yüksek sonuçlar sunabileceği gösterilmiştir.

Keywords

• Tiroid Nodül Segmentasyonu
• Yapay Zeka
• Örnek Segmentasyonu
• Semantik Segmantasyon

SEGMENTATION OF THYROID NODULES ON ULTRASOUND IMAGES

Abstract

The increasing prevalence of thyroid cancer in our country and globally has led to the development of various computer-aided studies for its detection, contributing significantly to the literature. Artificial intelligence and image processing are particularly prominent methods in this field due to their non-invasive nature, accessibility, and ability to provide valuable information about the morphological characteristics of nodules. In recent years, segmentation algorithms in medical imaging have garnered substantial interest for their potential to enhance diagnostic accuracy. Accurate segmentation of thyroid nodules is a critical first step in the development of AI-assisted clinical decision support systems for the detection and diagnosis of thyroid cancer. In this study, innovative methods were employed to detect thyroid nodules. A dice score of 79% was achieved in instance segmentation using the YOLOv5-Small algorithm when doppler images were excluded, while a dice score of 91% was obtained using the YOLOv5-Large algorithm on a dataset that included doppler images. In semantic segmentation, the Attention Unet++ and Manet algorithms achieved a dice score of 89% when doppler images were excluded, and 91% when they were included. These results demonstrate that images typically excluded by physicians could potentially offer better outcomes in computerized image processing.

Keywords

• Thyroid Nodule Segmentation
• Artificial Intelligence
• Instance Segmentation
• Semantic Segmantation

Supporting Institution

Istanbul Medeniyet University

Ethical Statement

This study received an official approval from the Clinical Ethics Committee of Istanbul Medeniyet University on 07.09.2022.

Thanks

This study was supported by the Scientific and Technological Research Council of Turkey (TÜBİTAK) under the 2214-A International PhD Research Fellowship Program with the number 1059B142100676 and the 2211-C Domestic PhD Fellowship Program for Priority Areas with the number 1649B032001140.

References

1. Abdolali, F., Kapur, J., Jaremko, J. L., Noga, M., Hareendranathan, A. R., & Punithakumar, K. (2020). Automated thyroid nodule detection from ultrasound imaging using deep convolutional neural networks. Computers in Biology and Medicine, 122, 103871. https://doi.org/10.1016/j.compbiomed.2020.103871
2. Aytaç, Z., Iseri, İ., & Dandıl, B. (2021). Derin Öğrenme Kullanarak Tiroid Kanseri Teşhisi. Avrupa Bilim ve Teknoloji Dergisi, (29), 292-298.
3. Buda, M., Wildman-Tobriner, B., Castor, K., Hoang, J. K., & Mazurowski, M. A. (2020). Deep learning-based segmentation of nodules in thyroid ultrasound: improving performance by utilizing markers present in the images. Ultrasound in medicine & biology, 46(2), 415-421.
4. Chen, J., You, H., & Li, K. (2020). A review of thyroid gland segmentation and thyroid nodule segmentation methods for medical ultrasound images. Computer methods and programs in biomedicine, 185, 105329.
5. Demetriou, E., Fokou, M., Frangos, S., Papageorgis, P., Economides, P. A., & Economides, A. (2023). Thyroid nodules and obesity. Life, 13(6), 1292.
6. Eloy, C., Russ, G., Suciu, V., Johnson, S. J., Rossi, E. D., Pantanowitz, L., & Vielh, P. (2022). Preoperative diagnosis of thyroid nodules: An integrated multidisciplinary approach. Cancer Cytopathology, 130(5), 320-325.
7. Inan, N. G., Kocadağlı, O., Yıldırım, D., Meşe, İ., & Kovan, Ö. (2024). Multi-class classification of thyroid nodules from automatic segmented ultrasound images: Hybrid ResNet based UNet convolutional neural network approach. Computer Methods and Programs in Biomedicine, 243, 107921.
8. Gong, H., Chen, J., Chen, G., Li, H., Li, G., & Chen, F. (2023). Thyroid region prior guided attention for ultrasound segmentation of thyroid nodules. Computers in biology and medicine, 155, 106389.

9. Hettihewa, K., Kobchaisawat, T., Tanpowpong, N., & Chalidabhongse, T. H. (2023). MANet: a multi-attention network for automatic liver tumor segmentation in computed tomography (CT) imaging. Scientific Reports, 13(1), 20098. Hicks, S. A., Strümke, I., Thambawita, V., Hammou, M., Riegler, M. A., Halvorsen, P., & Parasa, S. (2022). On evaluation metrics for medical applications of artificial intelligence. Scientific reports, 12(1), 5979.
10. Hoang, J. K., Middleton, W. D., Farjat, A. E., Langer, J. E., Reading, C. C., Teefey, S. A., & Tessler, F. N. (2018). Reduction in thyroid nodule biopsies and improved accuracy with American College of Radiology Thyroid Imaging Reporting and Data System. Radiology, 287(1), 185-193.
11. Hoang, J. K., Middleton, W. D., Farjat, A. E., Teefey, S. A., Abinanti, N., Boschini, F. J., Bronner, A. J., Dahiya, N., Hertzberg, B. S., Newman, J. R., Scanga, D., Vogler, R. C., & Tessler, F. N. (2018). Interobserver Variability of Sonographic Features Used in the American College of Radiology Thyroid Imaging Reporting and Data System. AJR. American Journal of Roentgenology, 211(1), 162–167. https://doi.org/10.2214/AJR.17.19192
12. Jocher, G. (2020). YOLOv5 by Ultralytics (Version 7.0) [Python]. https://doi.org/10.5281/zenodo.3908559
13. Kunapinun, A., Dailey, M. N., Songsaeng, D., Parnichkun, M., Keatmanee, C., &Ekpanyapong, M. (2023). Improving GAN Learning Dynamics for Thyroid Nodule Segmentation. Ultrasound in Medicine & Biology, 49(2), 416–430. https://doi.org/10.1016/j.ultrasmedbio.2022.09.010
14. Qureshi, I., Yan, J., Abbas, Q., Shaheed, K., Riaz, A. B., Wahid, A., Khan, M. W. J., & Szczuko, P. (2023). Medical image segmentation using deep semantic-based methods: A review of techniques, applications and emerging trends. Information Fusion, 90, 316–352. https://doi.org/10.1016/j.inffus.2022.09.031
15. Sharma, R., Saqib, M., Lin, C. T., & Blumenstein, M. (2022). A Survey on Object Instance Segmentation. https://opus.lib.uts.edu.au/handle/10453/167620
16. Tessler, F. N., Middleton, W. D., Grant, E. G., Hoang, J. K., Berland, L. L., Teefey, S. A., Cronan, J. J., Beland, M. D., Desser, T. S., Frates, M. C., Hammers, L. W., Hamper, U. M., Langer, J. E., Reading, C. C., Scoutt, L. M., & Stavros, A. T. (2017). ACR Thyroid Imaging, Reporting and Data System (TI-RADS): White Paper of the ACR TI-RADS Committee. Journal of the American College of Radiology: JACR, 14(5), 587–595. https://doi.org/10.1016/j.jacr.2017.01.046
17. Ultralytics. (2024, June 27). YOLO Performance Metrics. https://docs.ultralytics.com/guides/yolo-performance-metrics
18. Yamashita, R., Kapoor, T., Alam, M. N., Galimzianova, A., Syed, S. A., Ugur Akdogan, M., Alkim, E., Wentland, A. L., Madhuripan, N., Goff, D., Barbee, V., Sheybani, N. D., Sagreiya, H., Rubin, D. L., & Desser, T. S. (2022). Toward Reduction in False-Positive Thyroid Nodule Biopsies with a Deep Learning-based Risk Stratification System Using US Cine-Clip Images. Radiology. Artificial Intelligence, 4(3), e210174. https://doi.org/10.1148/ryai.210174
19. Yang, D., Xia, J., Li, R., Li, W., Liu, J., Wang, R., Qu, D., & You, J. (2024). Automatic Thyroid Nodule Detection in Ultrasound Imaging With Improved YOLOv5 Neural Network. IEEE Access, 12, 22662–22670. IEEE Access. https://doi.org/10.1109/ACCESS.2024.3359367
20. Zaidi, S. S. A., Ansari, M. S., Aslam, A., Kanwal, N., Asghar, M., & Lee, B. (2022). A survey of modern deep learning based object detection models. Digital Signal Processing, 126, 103514. https://doi.org/10.1016/j.dsp.2022.103514
21. Zhou, H., Wang, R., Zhou, M., Fu, P., & Bai, Y. (2022). A Deep Learning-Based Cascade Automatic Classification System for Malignant Thyroid Nodule Recognition in Ultrasound Image. Journal of Physics: Conference Series, 2363(1), 012029. https://doi.org/10.1088/1742-6596/2363/1/012029
22. Zou, K. H., Warfield, S. K., Bharatha, A., Tempany, C. M. C., Kaus, M. R., Haker, S. J., Wells, W. M., Jolesz, F. A., & Kikinis, R. (2004). Statistical Validation of Image Segmentation Quality Based on a Spatial Overlap Index. Academic Radiology, 11(2), 178–189. https://doi.org/10.1016/S1076-6332(03)00671-8