Automated identification of arthroplasty implants could aid in pre-operative planning and is a task which could be facilitated through artificial intelligence (AI) and deep learning. The purpose of this study was to develop and test the performance of a deep learning system (DLS) for automated identification and classification of knee arthroplasty (KA) on radiographs. We collected 237 AP knee radiographs with equal proportions of native knees, total KA (TKA), and unicompartmental KA (UKA), as well as 274 radiographs with equal proportions of Smith & Nephew Journey and Zimmer NexGen TKAs. Data augmentation was used to increase the number of images available for DLS development. These images were used to train, validate, and test deep convolutional neural networks (DCNN) to 1) detect the presence of TKA; 2) differentiate between TKA and UKA; and 3) differentiate between the 2 TKA models. Receiver operating characteristic (ROC) curves were generated with area under the curve (AUC) calculated to assess test performance.Introduction
Methods
Several new studies have shown, that the defect size plays a major role in the clinical outcome of the different procedures. Thus, it makes sense to measure the size of a cartilage defect before indicating a certain method for biological repair.
In order to proof the reliability and the usefulness of this device, we carried out following study: in each of 6 cadaver-knees we performed 2 full-thickness cartilage defects (MFC and LFC) of different size. In a first run 3 surgeons had to scope the joint and estimate the defect size with means of a scaled probe-hook. In a second run we performed a measurement of the defect with the Orthopilotâ„¢; finally an open measurement after arthrotomy was done to act as reference.