Paul H Yi1, Jinchi Wei2, Tae Kyung Kim3, Haris I Sair4, Ferdinand K Hui5, Gregory D Hager6, Jan Fritz7, Julius K Oni8. 1. The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, 601 N Caroline St, Room 4223, Baltimore, MD 21287, United States of America; Radiology Artificial Intelligence Lab (RAIL), Malone Center for Engineering in Healthcare, Johns Hopkins University Whiting School of Engineering, 3400 N Charles St, Baltimore, MD 21218, United States of America. Electronic address: pyi10@jhmi.edu. 2. Radiology Artificial Intelligence Lab (RAIL), Malone Center for Engineering in Healthcare, Johns Hopkins University Whiting School of Engineering, 3400 N Charles St, Baltimore, MD 21218, United States of America. Electronic address: jwei9@jhu.edu. 3. The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, 601 N Caroline St, Room 4223, Baltimore, MD 21287, United States of America; Radiology Artificial Intelligence Lab (RAIL), Malone Center for Engineering in Healthcare, Johns Hopkins University Whiting School of Engineering, 3400 N Charles St, Baltimore, MD 21218, United States of America. Electronic address: tonykim@jhmi.edu. 4. The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, 601 N Caroline St, Room 4223, Baltimore, MD 21287, United States of America; Radiology Artificial Intelligence Lab (RAIL), Malone Center for Engineering in Healthcare, Johns Hopkins University Whiting School of Engineering, 3400 N Charles St, Baltimore, MD 21218, United States of America. Electronic address: hsair1@jhmi.edu. 5. The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, 601 N Caroline St, Room 4223, Baltimore, MD 21287, United States of America; Radiology Artificial Intelligence Lab (RAIL), Malone Center for Engineering in Healthcare, Johns Hopkins University Whiting School of Engineering, 3400 N Charles St, Baltimore, MD 21218, United States of America. Electronic address: fhui2@jhmi.edu. 6. Radiology Artificial Intelligence Lab (RAIL), Malone Center for Engineering in Healthcare, Johns Hopkins University Whiting School of Engineering, 3400 N Charles St, Baltimore, MD 21218, United States of America. Electronic address: Hager@cs.jhu.edu. 7. The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, 601 N Caroline St, Room 4223, Baltimore, MD 21287, United States of America; Radiology Artificial Intelligence Lab (RAIL), Malone Center for Engineering in Healthcare, Johns Hopkins University Whiting School of Engineering, 3400 N Charles St, Baltimore, MD 21218, United States of America. Electronic address: jfritz9@jhmi.edu. 8. Department of Orthopaedic Surgery, Johns Hopkins University School of Medicine, 4940 Eastern Avenue Building A 6th Floor, Baltimore, MD 21224, United States of America.
Abstract
BACKGROUND: Preoperative identification of knee arthroplasty is important for planning revision surgery. However, up to 10% of implants are not identified prior to surgery. The purposes of this study were to develop and test the performance of a deep learning system (DLS) for the automated radiographic 1) identification of the presence or absence of a total knee arthroplasty (TKA); 2) classification of TKA vs. unicompartmental knee arthroplasty (UKA); and 3) differentiation between two different primary TKA models. METHOD: We collected 237 anteroposterior (AP) knee radiographs with equal proportions of native knees, TKA, and UKA and 274 AP knee radiographs with equal proportions of two TKA models. Data augmentation was used to increase the number of images for deep convolutional neural network (DCNN) training. A DLS based on DCNNs was trained on these images. Receiver operating characteristic (ROC) curves with area under the curve (AUC) were generated. Heatmaps were created using class activation mapping (CAM) to identify image features most important for DCNN decision-making. RESULTS: DCNNs trained to detect TKA and distinguish between TKA and UKA both achieved AUC of 1. Heatmaps demonstrated appropriate emphasis of arthroplasty components in decision-making. The DCNN trained to distinguish between the two TKA models achieved AUC of 1. Heatmaps showed emphasis of specific unique features of the TKA model designs, such as the femoral component anterior flange shape. CONCLUSIONS: DCNNs can accurately identify presence of TKA and distinguish between specific arthroplasty designs. This proof-of-concept could be applied towards identifying other prosthesis models and prosthesis-related complications.
BACKGROUND: Preoperative identification of knee arthroplasty is important for planning revision surgery. However, up to 10% of implants are not identified prior to surgery. The purposes of this study were to develop and test the performance of a deep learning system (DLS) for the automated radiographic 1) identification of the presence or absence of a total knee arthroplasty (TKA); 2) classification of TKA vs. unicompartmental knee arthroplasty (UKA); and 3) differentiation between two different primary TKA models. METHOD: We collected 237 anteroposterior (AP) knee radiographs with equal proportions of native knees, TKA, and UKA and 274 AP knee radiographs with equal proportions of two TKA models. Data augmentation was used to increase the number of images for deep convolutional neural network (DCNN) training. A DLS based on DCNNs was trained on these images. Receiver operating characteristic (ROC) curves with area under the curve (AUC) were generated. Heatmaps were created using class activation mapping (CAM) to identify image features most important for DCNN decision-making. RESULTS: DCNNs trained to detect TKA and distinguish between TKA and UKA both achieved AUC of 1. Heatmaps demonstrated appropriate emphasis of arthroplasty components in decision-making. The DCNN trained to distinguish between the two TKA models achieved AUC of 1. Heatmaps showed emphasis of specific unique features of the TKA model designs, such as the femoral component anterior flange shape. CONCLUSIONS: DCNNs can accurately identify presence of TKA and distinguish between specific arthroplasty designs. This proof-of-concept could be applied towards identifying other prosthesis models and prosthesis-related complications.
Authors: Ravi Patel; Elizabeth H E Thong; Vineet Batta; Anil Anthony Bharath; Darrel Francis; James Howard Journal: Radiol Artif Intell Date: 2021-03-17
Authors: Cesar D Lopez; Anastasia Gazgalis; Venkat Boddapati; Roshan P Shah; H John Cooper; Jeffrey A Geller Journal: Arthroplast Today Date: 2021-09-03