AŞİKAR OLMAYAN RADİUS VE KARPAL KEMİK KIRIKLARININ TESPİTİ İÇİN DERİN ÖĞRENME SİNİR AĞLARI MODELİNİN İNSAN GÖZLEMLERİYLE KARŞILAŞTIRILMASI

Year 2025, Volume: 6 Issue: 3, 304 - 310, 25.11.2025

Ayça Koca , Kaan Orhan , Ahmet Burak Oğuz , Yusuf Kahya , Atilla Elhan Elhan , Müge Günalp , Huan Yan , Pengfei Liu , Emre Can Çelebioğlu , Ayten Kayi Cangır

Abstract

Giriş: Acil serviste sıklıkla el ve bilek travması olan hastalara tanı konur. Derin öğrenme algoritmaları, X-ışını bilek görüntülerinden kırıkları teşhis etmek için güçlü araçlar haline gelebilir. Bu çalışma, radyografilerle tespit edilmesi zor olan bilek kırıklarını tespit etmede derin öğrenme algoritmasının tanı performansını değerlendirmeyi amaçlamaktadır.

Yöntemler: Bu retrospektif çalışma, BT görüntülemesi yapılan el/bilek travması olan yetişkin hastaları içermektedir. Uzman bir radyolog tarafından yorumlanan yaralı bölgelerin BT görüntüleri "temel gerçek" (TG) olarak kabul edildi. 313 vaka çalışmaya dahil edildi, toplam 121 kırık (82 radius 39 karpal kemik) BT görüntülerinden TG olarak tanımlandı. Algoritma kullanılarak, el ve bilek X-ışını görüntülerinden oluşan veri setinde kırık tespit prosedürü gerçekleştirildi. Aynı veri setleri dört acil tıp doktoru tarafından değerlendirildi. Doğruluk, eğri altında kalan alan, duyarlılık, kesinlik ve F1 skoru gibi tanı performansları hesaplandı. TG, gözlemciler ve derin öğrenme algoritması arasındaki uyum (Kappa katsayısı (κ)) belirlendi.

Bulgular: Algoritma %69,6 doğruluk, %57 duyarlılık ve %61,6 kesinlik gösterdi. Acil tıp doktorları daha yüksek doğruluk, duyarlılık ve kesinlik ve AUC değerleriyle daha iyi tanı performansı gösterdi. Dört acil tıp doktoru arasındaki gözlemciler arası uyum orta düzeydeyken algoritmayla uyum yalnızca orta düzeydeydi.

Sonuç: Derin öğrenme algoritması, bilek röntgenlerinde kırıkları doğru bir şekilde tespit etti ve acil tıp doktorlarınınkine benzer yeteneklere sahipti, ancak daha iyi sonuçlar elde etmek için bahsedilen algoritmanın daha da iyileştirilmesi gerekiyor.

Keywords

deep learning , neural networks , fractures , carpal bones , radius fractures

A COMPARISON OF A DEEP LEARNING NEURAL NETWORKS MODEL WITH HUMAN OBSERVATIONS FOR DETECTING NON-OBVIOUS RADIUS AND CARPAL BONE FRACTURES

Abstract

Introduction: Patients with hand and wrist trauma are frequently diagnosed in the emergency department. Deep learning algorithms could potentially become powerful tools to diagnose fractures from X-ray wrist images. This study aims to assess the diagnostic performance of a deep learning algorithm in detecting wrist fractures that are difficult to detect through radiographs.

Methods: This retrospective study included adult patients with hand/wrist trauma who undergo CT imaging. CT imaging of injured areas, interpreted by an expert radiologist were considered as “ground truth” (GT). There were 313 cases, a total of 121 fractures (82 radius, 39 carpal bones) were identified as GT from CT images. Using the algorithm, fracture detection procedure was performed on dataset of hand and wrist X-ray images. The same datasets were evaluated by four emergency medicine doctors. Diagnostic performances such as accuracy, area under curve, sensitivity, precision and F1 score were calculated. Agreement (Kappa coefficient (κ)) between GT, observers and deep learning algorithm was determined.

Results: The algorithm showed 69.6% accuracy, 57% sensitivity and 61.6% precision. Emergency medicine doctors showed better diagnostic performance with higher accuracy, sensitivity and precision and AUC values. The interobserver agreement among four EM doctors was moderate whereas the agreement with the algorithm was only fair.

Conclusions: The Deep learning algorithm demonstrated an accurate detection of fractures in wrist X-rays and it had capabilities that were comparable to those of emergency medicine physicians, but the algorithm mentioned needs to be further improved to produce better outcome.

Keywords

deep learning , neural networks , fractures , carpal bones , radius fractures

Ethical Statement

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics committee of Ankara University School of Medicine (Ethics approval number: 2021000367).

References

• Eschweiler J, Li J, Quack V et al. Anatomy, Biomechanics, and Loads of the Wrist Joint. Life 2022;12:188.
• Shahabpour M, Abid W, Van Overstraeten L, De Maeseneer M. Wrist Trauma: More Than Bones. Journal of the Belgian Society of Radiology 2021;105:90.
• Obert L, Loisel F, Jardin E, Gasse N, Lepage D. High-energy injuries of the wrist. Orthopaedics & traumatology, surgery & research 2016;102:81-93.
• Ozkaya E, Topal FE, Bulut T, Gursoy M, Ozuysal M, Karakaya Z. Evaluation of an artificial intelligence system for diagnosing scaphoid fracture on direct radiography. European journal of trauma and emergency surgery 2022;48:585–592.
• Bruno F, Arrigoni, Palumbo P, et al. The Acutely Injured Wrist. Radiologic clinics of North America 2019;57:943–955.
• Pinto A, Reginelli A, Pinto F, et al. Errors in imaging patients in the emergency setting. The British journal of radiology 2016;89:20150914.
• Elzinga JL, Dunne C L, Vorobeichik A, et al. A Systematic Review Protocol to Determine the Most Effective Strategies to Reduce Computed Tomography Usage in the Emergency Department. Cureus 2020;12: e9509.
• Ma Y, Lin C, Liu S, et al. Radiomics features based on internal and marginal areas of the tumor for the preoperative prediction of microsatellite instability status in colorectal cancer. Frontiers in Oncology 2022;12:1020349.
• Wu J, Fang Q, Yao J, et al. Integration of ultrasound radiomics features and clinical factors: A nomogram model for identifying the Ki-67 status in patients with breast carcinoma. Frontiers in Oncology 2022;12:979358.
• Xie D, Xu F, Zhu W, et al. Delta radiomics model for the prediction of progression-free survival time in advanced non-small-cell lung cancer patients after immunotherapy. Frontiers in Oncology 2022;12:990608.
• Cangir AK, Orhan K, Kahya Y, et al. A CT-Based Radiomic Signature for the Differentiation of Pulmonary Hamartomas from Carcinoid Tumors. Diagnostics 2022;12:416.
• Langerhuizen DWG, Bulstra AEJ, Janssen SJ, et al. Is Deep Learning On Par with Human Observers for Detection of Radiographically Visible and Occult Fractures of the Scaphoid?. Clinical orthopaedics and related research 2020;478:2653-2659.
• Long, J., E. Shelhamer, T. Darrell. Fully convolutional networks for semantic segmentation. In:2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2015.
• Ronneberger O, Fischer P, Brox T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation. In: Medical Image Computing and Computer-Assisted Intervention MICCAI 2015, Lecture Notes in Computer Science 2015; 9351.
• Ren S, He K, Girshick R, Sun J. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017;39:137-1149.
• Lin TY, Dollár P, Girshick R, He K, Hariharan B, Belongie S. Feature pyramid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition 2017;2117-2125.
• Zech JR, Santomartino SM, Yi PH. Artificial Intelligence (AI) for Fracture Diagnosis: An Overview of Current Products and Considerations for Clinical Adoption, From the AJR Special Series on AI Applications. American journal of roentgenology 2022;219:869-878.
• Lamb L, Kashani P, Ryan J, et al. Impact of an in-house emergency radiologist on report turnaround time. Canadian journal of emergency medicine 2015;17:21-26.
• Duron L, Ducarouge A, Gillibert A, et al. Assessment of an AI Aid in Detection of Adult Appendicular Skeletal Fractures by Emergency Physicians and Radiologists: A Multicenter Cross-sectional Diagnostic Study. Radiology 2021;300:120-129.
• Guermazi A, Tannoury C, Kompel AJ, et al. Improving radiographic fracture recognition performance and efficiency using artificial intelligence. Radiology 2022;302:627-36.
• Nehrer S, Ljuhar R, Steindl P, et al. Automated Knee Osteoarthritis Assessment Increases Physicians' Agreement Rate and Accuracy: Data from the Osteoarthritis Initiative. Cartilage 2021;13:957S-965S.
• Cohen M, Puntonet J, Sanchez J, et al. Artificial intelligence vs. radiologist: accuracy of wrist fracture detection on radiographs. European radiology 2023;33:3974-3983.
• Heo YM, Kim SB, Yi JW, et al. Evaluation of associated carpal bone fractures in distal radial fractures. Clinics in orthopedic surgery 2013;5:98-104.
• Nguyen T, Maarek R, Hermann AL, et al. Assessment of an artificial intelligence aid for the detection of appendicular skeletal fractures in children and young adults by senior and junior radiologists. Pediatric Radiology. 2022;52:2215-2226.
• Herpe G, Nelken H, Vendeuvre T, et al. Effectiveness of an artificial intelligence software for limb radiographic fracture recognition in an emergency department. Journal of Clinical Medicine 2024;13:5575.
• Canoni-Meynet L, Verdot P, Danner A, Calame P, Aubry S. Added value of an artificial intelligence solution for fracture detection in the radiologist's daily trauma emergencies workflow. Diagnostic and interventional imaging 2022;103:594-600.
• Çalışkan SA, Demir K, Karaca O. Artificial intelligence in medical education curriculum: an e-Delphi study for competencies. PLoS One 2022;17:e0271872.
• Lindsey R, Daluiski A, Chopra S, et al. Deep neural network improves fracture detection by clinicians. Proceedings of the National Academy of Sciences 2018;115:11591-11596.
• Snaith BA, Lancaster A. Clinical history and physical examination skills–a requirement for radiographers ?. Radiography 2008;14150-153.
• Huang X, Han F, Chen YF, et al. Bibliometric analysis of the application of artificial intelligence in orthopedic imaging. Quantitative Imaging in Medicine and Surger 2025;15:3993
• Alowais SA, Alghamdi SS, Alsuhebany N, et al. Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC medical education 2023;23:689.