Gelişimsel kalça displazisi ultrason görüntülerinin iki aşamalı derin öğrenme yaklaşımı ile kullanabilirlik analizinin yapılması

Yıl 2025, , 541 - 554, 16.08.2024

Muhammed Cihad Özdemir , Bahattin Kerem Aydın , Sadettin Çiftci , Murat Ceylan

https://doi.org/10.17341/gazimmfd.1318983

Öz

Gelişimsel kalça displazisini tespit etmek için ana standart ultrasonik görüntülemedir. Ancak bu teknik operatöre ve dış faktörlere bağlıdır. Bu çalışmada değişkenliği ortadan kaldırmak ve hataları en aza indirmek için yapay zekâ tabanlı bir sistem geliştirilmiştir. Derin öğrenme modeli ilk olarak 3 temel anatomik yapının bulunduğu alanı segmentlere ayırır, daha sonra segmente edilen alanı özelliklerine göre sınıflandırarak 2 aşamalı bir derin öğrenme modeli oluşturur. 3 temel anatomik yapı segmente edildikten sonra ultrason görüntüsünün analiz edilebilirliği kontrol edilir. Ultrason görüntüsünün analiz edilebilirliğinin kontrolü için önce 3 temel anatomik yapının tespit edilip edilmediği kontrol edilmektedir. 3 temel anatomik yapı tespit edilmişse, iliak kemiğin taban çizgisine paralelliği kontrol edilerek US görüntüsünün analiz edilebilirliği kontrol edilmektedir. Geliştirilen sistemde toplam 840 görüntü kullanılmıştır. Sistem US görüntülerinin kullanılabilirliğini %96 doğrulukla karar verdiği gözlemlenmiştir. Geliştirilen sistemin gelişimsel kalça displazisi tanısında doğruluk ve hız açısından doktorlara yardımcı olacağı düşünülmektedir.

Anahtar Kelimeler

: Derin Öğrenme, Gelişimsel Kalça Displazisi, Konvolüsyonel Sinir Ağları, Maskeli Bölge Tabanlı Evrişimli Sinir Ağı, U-NET

Kaynakça

• 1. H. Patel, "Canadian Task Force on Preventive Health Care. Preventive health care, 2001 update: Screening and management of developmental dysplasia of the hip in newborns," CMAJ, 164, 1669–1677, 2001.
• 2. V. Bialik, G. M. Bialik, S. Blazer, P. Sujov, F. Wiener, and M. Berant, "Developmental dysplasia of the hip: A new approach to incidence," Pediatrics, 103, 93–99, 1999.
• 3. H. Atalar, U. Sayli, O. Y. Yavuz, I. Uras, and H. Dogruel, Indicators of successful use of the Pavlik harness in infants with developmental dysplasia of the hip, Int. Orthop., 31, 145–150, 2007.
• 4. W. H. Harris, Etiology of osteoarthritis of the hip, Clin. Orthop. Relat. Res., 213, 20-33, 1986.
• 5. P. Tomà, M. Valle, U. Rossi, and G. M. Brunenghi, Paediatric hip—ultrasound screening for developmental dysplasia of the hip: A review, Eur. J. Ultrasound, 14, 45–55, 2001.
• 6. H. Omeroğlu, Use of ultrasonography in developmental dysplasia of the hip, J. Child Orthop., 8 (2), 105-113, 2014.
• 7. T. Centel, Doğumsal Kalça Çıkığı (DDH) Nedir?, [Online]. Erişim Tarihi: [Ziyaret Tarihi: 24 Haziran 2023]. Web Adresi: https://tuncaycentel.com/ddh/ddhTR7.htm.
• 8. H. Dogruel, H. Atalar, O. Y. Yavuz, and U. Sayli, Clinical examination versus ultrasonography in detecting developmental dysplasia of the hip, Int. Orthop., 32, 415–419, 2008.
• 9. R. Graf, The diagnosis of congenital hip-joint dislocation by the ultrasonic combbound treatment, Arch. Orthop. Trauma Surg., 97, 117–133, 1980.
• 10. S.-W. Lee, H.-U. Ye, K.-J. Lee, W.-Y. Jang, J.-H. Lee, S.-M. Hwang, and Y.-R. Heo, Reply to Çiftci, S.; Aydin, B.K. Comment on 'Accuracy of New Deep Learning Model-Based Segmentation and Key-Point Multi-Detection Method for Ultrasonographic Developmental Dysplasia of the Hip (DDH) Screening', Diagnostics, 12, 7, article 1739, 2022.
• 11. J. J. Dias, I. H. Thomas, A. C. Lamont, B. S. Mody, and J. R. Thompson, The reliability of ultrasonographic assessment of neonatal hips, J. Bone Joint Surg. Br., 75, 479–482, 1993.
• 12. Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, Gradient-based learning applied to document recognition, Proc. IEEE, 86, 2278–2324, 1998.
• 13. D. Golan, Y. Donner, C. Mansi, J. Jaremko, and M. Ramachandran, Fully automating Graf's method for DDH diagnosis using deep convolutional neural networks, in Lecture Notes in Computer Science, G. Carneiro, Ed., 10008, Springer, 130–141, 2016.
• 14. A. R. Hareendranathan, D. Zonoobi, M. Mabee, D. Cobzas, K. Punithakumar, M. Noga, and J. L. Jaremko, Toward automatic diagnosis of hip dysplasia from 2D ultrasound, in Proc. IEEE 14th Int. Symp. Biomed. Imaging (ISBI), 982–985, 2017.
• 15. H. Den, J. Ito, and A. Kokaze, Diagnostic accuracy of a deep learning model using YOLOv5 for detecting developmental dysplasia of the hip on radiography images, Sci. Rep., 13, article 6693, 2023.
• 16. S.-W. Lee, H.-U. Ye, K.-J. Lee, W.-Y. Jang, J.-H. Lee, S.-M. Hwang, and Y.-R. Merhaba, Ultrasonografik Gelişimsel Kalça Displazisi (GKD) Taramasında Yeni Derin Öğrenme Modeline Dayalı Segmentasyon ve Anahtar Noktalı Çoklu Tespit Yönteminin Doğruluğu, Teşhis, 11, article 1174, 2021.
• 17. M. Kinugasa, A. Inui, S. Satsuma, D. Kobayashi, R. Sakata, M. Morishita, I. Komoto, and R. Kuroda, Derin Öğrenmeyi Kullanarak Ultrason Görüntüleme ile Gelişimsel Kalça Displazisinin Tanısı, Pediatrik Ortopedi Dergisi, 43 (7), e538-e544, 2023.
• 18. A. R. Hareendranathan, B. Chahal, S. Ghasseminia, et al., Impact of scan quality on AI assessment of hip dysplasia ultrasound, J. Ultrasound, 25, 145–153, 2022.
• 19. S.-C. Zhang, H.-L. Ma, X.-W. Sun, et al., Clinical study on ultrasonic artificial intelligence-assisted diagnosis of developmental hip dysplasia in children, Preprint, Research Square, May 8, 2023.
• 20. R. Gandhi, R-CNN, Fast R-CNN, Faster R-CNN, YOLO — Object Detection Algorithms: Understanding object detection algorithms, Towards Data Science, Jul. 9, 2018. [Online]. Available: https://towardsdatascience.com/r-cnn-fast-r-cnn-faster-r-cnn-yolo-object-detection-algorithms-36d53571365e. [Accessed: Apr. 6, 2023].
• 21. S. Ren, K. He, R. Girshick, and J. Sun, Faster R-CNN: Towards real-time object detection with region proposal networks, IEEE Trans. Pattern Anal. Mach. Intell., 39, 1137–1149, 2017. 22. Kadiroğlu Z., Deniz E., Şenyiğit A., A comparison of deep learning models for pneumonia detection from chest x-ray images, Journal of the Faculty of Engineering and Architecture of Gazi University, 39 (2), 729-740, 2024.
• 23. Üzülmez S., Çifçi M.A., Early diagnosis of lung cancer using deep learning and uncertainty measures, Journal of the Faculty of Engineering and Architecture of Gazi University, 39 (1), 385-400, 2024.
• 24. Yılmaz A., Diagnosing COVID-19 from X-Ray images with using multi-channel CNN architecture, Journal of the Faculty of Engineering and Architecture of Gazi University, 36 (4), 1761-1774, 2021.
• 25. İsmail, T., "Evaluation of Ultrasonography of the Neonatal Hip," Journal of the Turkish Orthopedics and Traumatology Association, 3, no. 3-4, 2004.
• 26. Ronneberger, O., Fischer, P., and Brox, T., U-NET: Convolutional Networks for Biomedical Image Segmentation, in Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, N. Navab, J. Hornegger, W. Wells, and A. Frangi, Eds., Springer, 9351, 2015.
• 27. He, K., Gkioxari, G., Dollar, P., and Girshick, R., Mask R-CNN, in Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2961–2969, 2017.
• 28. Lee, S.-W., Ye, H.-U., Lee, K.-J., Jang, W.-Y., Lee, J.-H., Hwang, S.-M., and Heo, Y.-R., Accuracy of New Deep Learning Model-Based Segmentation and Key-Point Multi-Detection Method for Ultrasonographic Developmental Dysplasia of the Hip (DDH) Screening, Diagnostics, 11 (7), 1174, 2021.
• 29. Anand, V., Gupta, S., Koundal, D., Nayak, S. R., Barsocchi, P., and Bhoi, A. K., Modified U-NET Architecture for Segmentation of Skin Lesion, 22, 867, 2022.
• 30. Rosebrock, A., Intersection over Union (IoU) for Object Detection, PyImageSearch, Nov. 7, 2016. [Online]. Available: https://www.pyimagesearch.com/2016/11/07/intersection-over-union-iou-for-object-detection/. [Accessed: Apr. 6, 2023].
• 31. Tiu, E., Metrics to Evaluate Your Semantic Segmentation Model, Towards Data Science, Aug. 10, 2019. [Online]. Available: https://towardsdatascience.com/metrics-to-evaluate-your-semantic-segmentation-model-6bcb99639aa2. [Accessed: Apr. 6, 2023].
• 32. Tomà, P., Valle, M., Rossi, U., and Brunenghi, G. M., Pediatric Hip—Ultrasound Screening for Developmental Dysplasia of the Hip: A Review, European Journal of Ultrasound, 14, 45-55, 2001.
• 33. Hareendranathan, A. R., Chahal, B. S., Zonoobi, D., et al., Artificial Intelligence to Automatically Evaluate Scan Quality in Hip Ultrasound, Journal of the Indian Orthopaedic Association, 55, 1535–1542, 2021.
• 34. Shin, Y., et al., Artificial Intelligence in Musculoskeletal Ultrasound Imaging, Ultrasonography (Seoul, Korea), 40 (1), 30-44, 2021.
• 35. Chen, T., Zhang, Y., Wang, B., Wang, J., Cui, L., He, J., and Cong, L., Development of a Fully Automated Graf Standard Plane and Angle Evaluation Method for Infant Hip Ultrasound Scans, Diagnostics, 12, 1423, 2022.
• 36. Zhang, S.-C., Sun, J., Liu, C.-B., Fang, J.-H., Xie, H.-T., and Ning, B., The Bone & Joint Journal, The Bone & Joint Journal, 102-B, no. 11, 1574-1581, 2020.
• 37. Li, Q., Zhong, L., Huang, H., Liu, H., Qin, Y., Wang, Y., Zhou, Z., Liu, H., Yang, W., Qin, M., Wang, J., Wang, Y., Zhou, T., Wang, D., Wang, J., Xu, M., and Huang, Y., Auxiliary Diagnosis of Developmental Dysplasia of the Hip by Automated Detection of Sharp's Angle on Standardized Anteroposterior Pelvic Radiographs, Medicine, 98, no. 52, article e18500, 2019.
• 38. Lee, S.-W., Ye, H.-U., Lee, K.-J., Jang, W.-Y., Lee, J.-H., Hwang, S.-M., and Heo, Y.-R., Accuracy of New Deep Learning Model-Based Segmentation and Key-Point Multi-Detection Method for Ultrasonographic Developmental Dysplasia of the Hip (DDH) Screening, Diagnostics, 11 (7), 1174, 2021.
• 39. Hareendranathan, A. R., Mabee, M., Chahal, B. S., Dulai, S. K., and Jaremko, J. L., Can AI Automatically Assess Scan Quality of Hip Ultrasound? Applied Sciences, 12 (8), 4072, 2022.
• 40. Jejurikar, N., et al., What Is the Interobserver Reliability of an Ultrasound-Enhanced Physical Examination of the Hip in Infants? A Prospective Study on the Ease of Acquiring Skills to Diagnose Hip Dysplasia, Clinical Orthopaedics and Related Research, 479 (9), 1889-1896, 2021.
• 41. Patil, S., Varadarajan, V., Mahadevkar, S., Athawade, R., Maheshwari, L., Kumbhare, S., Garg, Y., Dharrao, D., Kamat, P., and Kotecha, K., Enhancing Optical Character Recognition on Images with Mixed Text Using Semantic Segmentation, Journal of Sensor and Actuator Networks, 11, 63, 2022.