Application of deep learning algorithms in classification and localization of implant cutout for the postoperative hip.

OBJECTIVE: This study aims to explore the feasibility of employing convolutional neural networks for detecting and localizing implant cutouts on anteroposterior pelvic radiographs. MATERIALS AND METHODS: The research involves the development of two Deep Learning models. Initially, a model was created for image-level classification of implant cutouts using 40191 pelvic radiographs obtained from a single institution. The radiographs were partitioned into training, validation, and hold-out test datasets in a 6/2/2 ratio. Performance metrics including the area under the receiver operator characteristics curve (AUROC), sensitivity, and specificity were calculated using the test dataset. Additionally, a second object detection model was trained to localize implant cutouts within the same dataset. Bounding box visualizations were generated on images predicted as cutout-positive by the classification model in the test dataset, serving as an adjunct for assessing algorithm validity. RESULTS: The classification model had an accuracy of 99.7%, sensitivity of 84.6%, specificity of 99.8%, AUROC of 0.998 (95% CI: 0.996, 0.999) and AUPRC of 0.774 (95% CI: 0.646, 0.880). From the pelvic radiographs predicted as cutout-positive, the object detection model could achieve 95.5% localization accuracy on true positive images, but falsely generated 14 results from the 15 false-positive predictions. CONCLUSION: The classification model showed fair accuracy for detection of implant cutouts, while the object detection model effectively localized cutout. This serves as proof of concept of using a deep learning-based approach for classification and localization of implant cutouts from pelvic radiographs.

Concomitant Cement Spacer and Peri-spacer Fractures Seven Years After First-stage Revision Knee Arthroplasty: A Case Report.

Introduction: Prosthetic joint infections (PJIs) remain an undesirable complication after total knee arthroplasties. Two-stage revision arthroplasty is the current standard of care for treating PJIs. However, the incidence of spacer retention for prolonged periods is increasing, with little known about its potential complications. Case Report: We present a case of a 64-year-old female of Southeast Asian descent who had a cement spacer maintained in-situ for 7 years due to poor patient compliance with subsequent follow-up. Conclusion: While patients have satisfactory functional outcomes with the cement spacer, it is not meant for permanent weight bearing. Two-stage revision arthroplasties are only as effective as patients' compliance with subsequent follow-up and surgery. Clinicians must discourage patients from forgoing subsequent follow-up visits and surgery despite satisfactory function and quality of life with the cement spacer in situ to prevent complications related to prolonged retention of cement spacers.

Proximal Tibiofibular Joint Stabilization With Concurrent Posterolateral Corner Reconstruction in Multiligamentous Knee Injury.

Proximal tibiofibular joint (PTFJ) instability is a rare knee injury, accounting for less than 1% of knee injuries. It causes significant lateral sided knee pain and functional deficits and can be associated with up to 9% of multiligament knee injuries. Concurrent surgical treatment of posterolateral corner (PLC) and PTFJ instability poses technical challenges due to the limited working space of the fibula head and inherent risk of collision between grafts, bone tunnels, and implants. In this Technical Note, we detail our senior author's technique for PTFJ reconstruction without the use of additional bone tunnels or implants in the fibula head, to reduce the risk of overcrowding and tunnel collision. CLASSIFICATIONS: Level I: knee; Level II: other, proximal tibiofibular joint.

Predictors of failure following fixation of intertrochanteric fractures with proximal femoral nail antirotation.

INTRODUCTION: This retrospective matched case-control study aimed to identify predictors of cut-out following intramedullary nailing of intertrochanteric fractures with the 200-mm Synthes proximal femoral nail antirotation (PFNA). METHODS: 609 patients underwent intramedullary nailing for intertrochanteric fractures at our institution between January 2011 and December 2014. 370 patients satisfied the inclusion criteria. There were 20 cases of implant cut-out. Cases and controls were matched using a propensity score-matching method with an m:n ratio, matching the criteria of gender, age and side of operation. Radiographs were assessed to determine fracture classification, fracture reduction quality, tip-apex distance, calcar referenced tip-apex distance (CalTAD), anteroposterior (AP) Parker's ratio index, lateral Parker's ratio index and cervical angle difference. Conditional logistic regression analysis was performed to determine any association between potential predictors and cut-outs. RESULTS: The cut-out incidence was 5.4%. Of the 20 cut-outs, 16 were superior and four were cut-throughs. Univariate analysis only showed a significant association between unsatisfactory fracture reduction quality and cut-outs (odds ratio [OR] 10.1, 95% confidence interval [CI] 1.31-77.6, p = 0.027). This association remained significant with multivariate logistic regression analysis (OR 16.4, 95% CI 1.9-140.4, p = 0.011). Cut-throughs had significantly lower CalTAD (16.2 vs. 27.5, p = 0.016) and AP Parker's ratio index values (38.7 vs. 50.7, p = 0.007) than superior cut-outs. CONCLUSION: Unsatisfactory fracture reduction quality was a significant predictor of cut-out in intertrochanteric fractures treated with the 200-mm PFNA. Cut-outs had two distinct modes, with cut-throughs having a deeper and more inferior helical blade position in the femoral head compared to superior cut-outs.