The aim of this study was to create artificial intelligence (AI) software with the purpose of providing a second opinion to physicians to support distal radius fracture (DRF) detection, and to compare the accuracy of fracture detection of physicians with and without software support. The dataset consisted of 26,121 anonymized anterior-posterior (AP) and lateral standard view radiographs of the wrist, with and without DRF. The convolutional neural network (CNN) model was trained to detect the presence of a DRF by comparing the radiographs containing a fracture to the inconspicuous ones. A total of 11 physicians (six surgeons in training and five hand surgeons) assessed 200 pairs of randomly selected digital radiographs of the wrist (AP and lateral) for the presence of a DRF. The same images were first evaluated without, and then with, the support of the CNN model, and the diagnostic accuracy of the two methods was compared.Aims
Methods
We evaluated 100 consecutive patients with a suspected scaphoid fracture but without evidence of a fracture on plain radiographs using MRI within 24 hours of injury, and bone scintigraphy three to five days after injury. The reference standard for a true radiologically-occult scaphoid fracture was either a diagnosis of fracture on both MRI and bone scintigraphy, or, in the case of discrepancy, clinical and/or radiological evidence of a fracture. MRI revealed 16 scaphoid and 24 other fractures. Bone scintigraphy showed 28 scaphoid and 40 other fractures. According to the reference standard there were 20 scaphoid fractures. MRI was falsely negative for scaphoid fracture in four patients and bone scintigraphy falsely positive in eight. MRI had a sensitivity of 80% and a specificity of 100%. Bone scintigraphy had a sensitivity of 100% and a specificity of 90%. This study did not confirm that early, short-sequence MRI was superior to bone scintigraphy for the diagnosis of a suspected scaphoid fracture. Bone scintigraphy remains a highly sensitive and reasonably specific investigation for the diagnosis of an occult scaphoid fracture.
