| 1. |
- Nguyen, Hanh H., et al.
(författare)
-
AFFnet - a deep convolutional neural network for the detection of atypical femur fractures from anteriorposterior radiographs
- 2024
-
Ingår i: Bone. - 8756-3282 .- 1873-2763. ; , s. 117215-117215
-
Tidskriftsartikel (refereegranskat)abstract
- Despite well-defined criteria for radiographic diagnosis of atypical femur fractures (AFFs), missed and delayed diagnosis is common. An AFF diagnostic software could provide timely AFF detection to prevent progression of incomplete or development of contralateral AFFs. In this study, we investigated the ability for an artificial intelligence (AI)-based application, using deep learning models (DLMs), particularly convolutional neural networks (CNNs), to detect AFFs from femoral radiographs. A labelled Australian dataset of pre-operative complete AFF (cAFF), incomplete AFF (iAFF), typical femoral shaft fracture (TFF), and non-fractured femoral (NFF) X-ray images in anterior-posterior view were used for training (N = 213, 49, 394, 1359, respectively). An AFFnet model was developed using a pretrained (ImageNet dataset) ResNet-50 backbone, and a novel Box Attention Guide (BAG) module to guide the model's scanning patterns to enhance its learning. All images were used to train and internally test the model using a 5-fold cross validation approach, and further validated by an external dataset. External validation of the model's performance was conducted on a Sweden dataset comprising 733 TFF and 290 AFF images. Precision, sensitivity, specificity, F1-score and AUC were measured and compared between AFFnet and a global approach with ResNet-50. Excellent diagnostic performance was recorded in both models (all AUC >0.97), however AFFnet recorded lower number of prediction errors, and improved sensitivity, F1-score and precision compared to ResNet-50 in both internal and external testing. Sensitivity in the detection of iAFF was higher for AFFnet than ResNet-50 (82 % vs 56 %). In conclusion, AFFnet achieved excellent diagnostic performance on internal and external validation, which was superior to a pre-existing model. Accurate AI-based AFF diagnostic software has the potential to improve AFF diagnosis, reduce radiologist error, and allow urgent intervention, thus improving patient outcomes.
|
|
| 2. |
- Zhou, Wei, et al.
(författare)
-
Gene-based association analysis of a large patient cohort identifies potential genecandidates for atypical femur fractures
-
Annan publikation (övrigt vetenskapligt/konstnärligt)abstract
- Background:Several small genetic association studies have been conducted for atypical femurfracture (AFF) without replication of results. We assessed previously implicated and novel genesassociated with AFFs in a larger set of unrelated AFF cases using whole exome sequencing (WES).Methods:We performed gene-based association analysis on 139 European AFF cases and 196 controlsmatched for bisphosphonate use. We tested all rare, protein-altering variants using both candidategene and hypothesis-free approaches. In the latter, genes suggestively associated with AFFs(uncorrected p-values < 0.01) were investigated in a Swedish whole-genome sequencing replicationstudy and assessed in 46 non-European cases.Results:In the candidate gene analysis, PLOD2 showed a suggestive signal. The hypothesis-freeapproach revealed 10 tentative associations, with XRN2, SORD, and PLOD2 being the most likelycandidates for AFF. XRN2 and PLOD2 showed consistent direction of effect estimates in thereplication analysis, albeit not statistically significant. Three SNPs associated with SORD expressionaccording to the GTEx portal, were in linkage disequilibrium (R2 ≥0.2) with a SNP previouslyreported in a genome-wide association study of AFF. The prevalence of carriers of variants for bothPLOD2 and SORD was higher in Asian versus European cases.Conclusions:While we did not identify genes enriched for damaging variants, we found suggestiveevidence of a role for XRN2, PLOD2 and SORD, which requires further investigation. Our findingsindicate that genetic factors responsible for AFFs are not widely shared among AFF cases. The studyprovides a stepping-stone for future larger genetic studies of AFF.
|
|
