Evaluation of ensemble strategy on the development of multiple view ankle fracture detection algorithm.

OBJECTIVE: To identify the feasibility and efficiency of deep convolutional neural networks (DCNNs) in the detection of ankle fractures and to explore ensemble strategies that applied multiple projections of radiographs.Ankle radiographs (AXRs) are the primary tool used to diagnose ankle fractures. Applying DCNN algorithms on AXRs can potentially improve the diagnostic accuracy and efficiency of detecting ankle fractures. METHODS: A DCNN was trained using a trauma image registry, including 3102 AXRs. We separately trained the DCNN on anteroposterior (AP) and lateral (Lat) AXRs. Different ensemble methods, such as "sum-up," "severance-OR," and "severance-Both," were evaluated to incorporate the results of the model using different projections of view. RESULTS: The AP/Lat model's individual sensitivity, specificity, positive-predictive value, accuracy, and F1 score were 79%/84%, 90%/86%, 88%/86%, 83%/85%, and 0.816/0.850, respectively. Furthermore, the area under the receiver operating characteristic curve (AUROC) of the AP/Lat model was 0.890/0.894 (95% CI: 0.826-0.954/0.831-0.953). The sum-up method generated balanced results by applying both models and obtained an AUROC of 0.917 (95% CI: 0.863-0.972) with 87% accuracy. The severance-OR method resulted in a better sensitivity of 90%, and the severance-Both method obtained a high specificity of 94%. CONCLUSION: Ankle fracture in the AXR could be identified by the trained DCNN algorithm. The selection of ensemble methods can depend on the clinical situation which might help clinicians detect ankle fractures efficiently without interrupting the current clinical pathway. ADVANCES IN KNOWLEDGE: This study demonstrated different ensemble strategies of AI algorithms on multiple view AXRs to optimize the performance in various clinical needs.

Methicillin-resistant Staphylococcus aureus nasal carriage and infection among patients with diabetic foot ulcer.

PURPOSE: To evaluate the prevalence of methicillin-resistant Staphylococcus aureus (MRSA) nasal carriage in patients with diabetic foot ulcer (DFU) in Taiwan, and to assess the concordance between colonizing and clinical MRSA isolates from the patients. METHOD: A total of 354 nasal specimens were collected from 112 to 242 diabetic patients with and without foot ulcer, respectively. MRSA clinical isolates from DFU wound cultures were collected for comparison. RESULTS: Nasal carriage rate of S. aureus and MRSA was similar between diabetic patients with and without foot ulcer (15.2% vs. 16.9% for S. aureus and 5.4% vs. 1.7% for MRSA). Nasal S. aureus colonization was an independent predictor for wound S. aureus infection (Odds ratio [OR]: 5.33, 95% confidence interval [CI]: 1.61-17.59), so did nasal MRSA colonization (OR: 19.09, 95% CI: 2.12-171.91). The levels of glycated hemoglobin, and the usage with immunosuppressant agent were associated with S. aureus nasal colonization while oral hypoglycemic agent usage a protective factor. Sequence type 59/staphylococcal chromosome cassette mec IV or V, the local endemic community-associated clone, accounted for 42% and 70% of the clinical and colonizing isolates, respectively. Six of 10 patients with paired colonizing and clinical isolates, either MRSA or methicillin-sensitive S. aureus, had a genetically identical strain from a single patient. CONCLUSION: Less than one-fifth of patients with DFU have nasal S. aureus, including MRSA, colonization; however, the colonization is significantly associated with S. aureus diabetic foot infection. Screening for S. aureus colonizing status in DFU patients might have a potential clinical implication.