Investigating the use of artificial intelligence to detect scaphoid fractures from a single incidence radiograph

Article Article scientifique Article Investigating the use of artificial intelligence to detect scaphoid fractures from a single incidence radiograph eng 2026-02-22 author Schmid Jérôme Frédéric 0000-0003-2464-8971 662359 author Al-Musibli Azal 630905 author Cagdas Onur 671727 author Foufi Vasiliki 0000-0002-8625-0734 848485 author Bjelogrlic Mina 0000-0002-6922-3283 918489 Département de radiologie et informatique médicale (RADIO) author Lovis Christian 0000-0002-2681-8076 29016 Département de radiologie et informatique médicale (RADIO) author Poletti Pierre-Alexandre Alois 0000-0001-9214-5546 34563 Département de radiologie et informatique médicale (RADIO) author Montet Xavier Cédric Rodolphe 0000-0003-2442-5866 58743 author Bouredoucen Hicham 0009-0003-1826-6827 419059 author Boudabbous Sana 0000-0003-0516-4613 800627 Département de radiologie et informatique médicale (RADIO) 2634726 10.1080/21681163.2026.2634726 2168-1163 We developed a deep learning (DL) algorithm to segment scaphoids and detect scaphoid fractures on single-incidence wrist radiographs, the most frequent carpal injuries whose early diagnosis is crucial for wrist function. We exploited a dataset of 1477 wrist radiographs and investigated strategies to mitigate data imbalance caused by low fracture prevalence (9%). The segmentation model achieved excellent precision (mAP@0.75 of 0.98), localising the scaphoid in all but one of 1141 test radiographs. However, severe class imbalance posed challenges in fracture detection. Our best fracture detection model achieved 74% sensitivity and 76% specificity on a balanced test dataset, surpassing an expert musculoskeletal radiologist’s sensitivity of 48% but falling short of their 94% specificity. This study demonstrates the potential of DL to detect scaphoid fractures from single-incidence radiographs, even in highly imbalanced datasets, potentially avoiding misdiagnosis and improving accuracy in emergency settings and non-specialised centres while reducing reliance on additional ionising imaging. Artificial intelligence Deep learning Scaphoid bone Fractures Radiography Dewey 616.0757 62 Département de radiologie et informatique médicale 227 false 469 Radiologie des urgences 800 344 Interfaces Homme-machine en milieu clinique 610 Computer methods in biomechanics and biomedical engineering. Imaging & visualization 14 1 51a09561-b7d6-4f69-a916-f79b283a5e4d Article (Published version) Public Investigating_the_use_of_artificial1.pdf 5604063 application/pdf CC BY-4.0 ca9a0e906afd46449130022aa7b7a8de0dd754590e7ae84c496dd3984127e14e openAccess Faculté de médecine / Section de médecine clinique / Département de radiologie et informatique médicale

Title	Published in	Access level	OA Policy	Year	Views	Downloads

Investigating the use of artificial intelligence to detect scaphoid fractures from a single incidence radiograph	Computer methods in biomechanics and biomedical engineering. Imaging & visualization			2026	23	234

Archive ouverte UNIGE

Cagdas, Onur