Artificial Intelligence for Automated Identification of Total Shoulder Arthroplasty Implants.

Overview

abstract

BACKGROUND: Accurate and rapid identification of implant manufacturer and model is critical in the evaluation and management of patients requiring revision total shoulder arthroplasty (TSA). Failure to correctly identify implant designs in these circumstances may lead to delay in care, unexpected intraoperative challenges, increased morbidity, and excess healthcare costs. Deep learning (DL) permits automated image processing and holds the potential to mitigate such challenges while improving the value of care rendered. The purpose of the current study was to develop an automated DL algorithm to identify shoulder arthroplasty implants from plain radiographs. METHODS: A total of 3,060 postoperative images from patients who underwent TSA between 2011-2021 by 26 fellowship-trained surgeons at two independent tertiary academic hospitals in the Pacific Northwest and Mid-Atlantic Northeast were included. A DL algorithm was trained using transfer learning and data augmentation to classify 22 different reverse (rTSA) and anatomic (aTSA) prostheses from eight implant manufacturers. Images were split into training and testing cohorts (2,448 training, 612 testing). Optimized model performance was assessed using standardized metrics including area under the multi-class receiver-operator characteristic curve (AUROC) and compared with a reference standard of implant data from operative reports. RESULTS: The algorithm classified implants at a mean speed of 0.079 (±0.002) seconds per image. The optimized model discriminated between eight manufactures (22 unique implants) with an AUROC of 0.994-1.000, accuracy of 97.1%, and sensitivities between 0.80-1.00 on the independent testing set. In the subset of single-institution implant predictions, a DL model identified 6-specific implants with AUROC of 0.999-1.000, accuracy of 99.4%, and sensitivity >0.97 for all implants. Saliency maps revealed key differentiating features across implant manufacturers and designs recognized by the algorithm for classification. CONCLUSION: A DL model demonstrated excellent accuracy in identifying 22 unique TSA implants from eight manufacturers. This algorithm may provide a clinically meaningful adjunct in assisting with preoperative planning for the failed TSA and allows for scalable expansion with additional radiographic data and validation efforts.