TOTBİD Dergisi

TOTBİD Dergisi

2022, Cilt 21, Sayı, 1     (Sayfalar: 032-039)

Automatic hip fracture diagnosis with deep learning

Salih Beyaz 1, Şahika Betül Yaylı 2, Uğur Doktur 2

1 Başkent Üniversitesi, Adana Dr. Turgut Noyan Uygulama ve Araştırma Merkezi, Ortopedi ve Travmatoloji Ana Bilim Dalı, Adana
2 Turkcell Teknoloji, Yapay Zeka & Dijital Analitik Çözümler, İstanbul

DOI: /10.5578/totbid.dergisi.2022.07
Görüntüleme: 672
İndirme : 665

Although more than a century has passed since the discovery of X-ray, X-ray imaging is the gold standard diagnostic method for fracture detection. Image processing is the field where artificial intelligence applications are most successful. The increase in the incidence of hip fractures due to the aging population increases the interest in diagnosis and classification systems. Obtaining a large number of homogeneously distributed data and classification by experienced experts is essential for the success of the algorithm, which is planned to be developed in real life. When we look at the literature, it is seen that besides diagnostic systems, classification systems are also being developed. In the studies conducted, there are studies not only on the X-ray image, but also on the use of the personal health records of the patient (radiology report, drugs used, etc.) for diagnostic purposes. Studies evaluating the risk of fracture, on the other hand, are promising for more effective use of preventive health services. As a result, artificial intelligence- based hip fracture recognition systems will contribute to reducing the mistake arising from experience differences.

Anahtar Kelimeler : deep learning; artificial intelligence; hip fracture; X-ray