Development and validation of an artificial intelligence model for the classification of hip fractures using the AO-OTA framework.

BACKGROUND AND PURPOSE: Artificial intelligence (AI) has the potential to aid in the accurate diagnosis of hip fractures and reduce the workload of clinicians. We primarily aimed to develop and validate a convolutional neural network (CNN) for the automated classification of hip fractures based on the 2018 AO-OTA classification system. The secondary aim was to incorporate the model's assessment of additional radiographic findings that often accompany such injuries. METHODS: 6,361 plain radiographs of the hip taken between 2002 and 2016 at Danderyd University Hospital were used to train the CNN. A separate set of 343 radiographs representing 324 unique patients was used to test the performance of the network. Performance was evaluated using area under the curve (AUC), sensitivity, specificity, and Youden's index. RESULTS: The CNN demonstrated high performance in identifying and classifying hip fracture, with AUCs ranging from 0.76 to 0.99 for different fracture categories. The AUC for hip fractures ranged from 0.86 to 0.99, for distal femur fractures from 0.76 to 0.99, and for pelvic fractures from 0.91 to 0.94. For 29 of 39 fracture categories, the AUC was ≥ 0.95. CONCLUSION: We found that AI has the potential for accurate and automated classification of hip fractures based on the AO-OTA classification system. Further training and modification of the CNN may enable its use in clinical settings.

Cause of death and significant disease found at autopsy.

The use of clinical autopsy has been in decline for many years throughout healthcare systems of developed countries despite studies showing substantial discrepancies between autopsy results and pre-mortal clinical diagnoses. We conducted a study to evaluate over time the use and results of clinical autopsies in Sweden. We reviewed the autopsy reports and autopsy referrals of 2410 adult (age > 17) deceased patients referred to two University hospitals in Sweden during two plus two years, a decade apart. There was a decline in the number of autopsies performed over time, however, mainly in one of the two hospitals. The proportion of autopsy referrals from the emergency department increased from 9 to 16%, while the proportion of referrals from regular hospital wards was almost halved. The autopsies revealed a high prevalence of cardiovascular disease, with myocardial infarction and cerebrovascular lesion found in 40% and 19% of all cases, respectively. In a large proportion of cases (> 30%), significant findings of disease were not anticipated before autopsy, as judged from the referral document and additional data obtained in some but not all cases. In accordance with previous research, our study confirms a declining rate of autopsy even at tertiary, academic hospitals and points out factors possibly involved in the decline.