Abstract
To assess the ability of imaging-based deep learning to detect radiographic patellofemoral osteoarthritis (PFOA) from knee lateral view radiographs. Knee lateral view radiographs were extracted from The Multicenter Osteoarthritis Study (MOST) public use datasets (n=18,436 knees). Patellar region-of-interest (ROI) was first automatically detected, and subsequently, end-to-end deep convolutional neural networks (CNNs) were trained and validated to detect the status of patellofemoral OA. Patellar ROI was detected using deep-learning-based object detection method. Atlas-guided visual assessment of PFOA status by expert readers provided in the MOST public use datasets was used as a classification outcome for the models. Performance of classification models was assessed using the area under the receiver operating characteristic curve (ROC AUC) and the average precision (AP) obtained from the Precision-Recall (PR) curve in the stratified 5-fold cross validation setting. Of the 18,436 knees, 3,425 (19%) had PFOA. AUC and AP for the reference model including age, sex, body mass index (BMI), the total Western Ontario and McMaster Universities Arthritis Index (WOMAC) score, and tibiofemoral Kellgren-Lawrence (KL) grade to detect PFOA were 0.806 and 0.478, respectively. The CNN model that used only image data significantly improved the classifier performance (ROC AUC=0.958, AP=0.862). We present the first machine learning based automatic PFOA detection method. Furthermore, our deep learning based model trained on patella region from knee lateral view radiographs performs better at detecting PFOA than models based on patient characteristics and clinical assessments.
Highlights
Plain radiography is commonly used in diagnostics of osteoarthritis (OA) because it is cheap, fast, and widely available
Of the 18,436 knees in 2,803 persons, 3,425 (19%) had patellofemoral osteoarthritis (PFOA) based on the metadata and the PFOA status assessment provided in the Multicenter Osteoarthritis Study (MOST) public use dataset as described in the methods sections
Our Convolutional Neural Networks (CNNs) model for predicting the radiographic PFOA in the 5fold cross validation setting showed the best performance with Receiver Operating Characteristics (ROC) area under the ROC curve (AUC) 0.958 [95% CI: 0.954e0.961] and average precision (AP) 0.862 [95% CI: 0.852e0.871]
Summary
Plain radiography is commonly used in diagnostics of osteoarthritis (OA) because it is cheap, fast, and widely available. Both clinical practice and the majority of research studies in OA have traditionally concentrated on the tibiofemoral (TF) joint. Frontal plane radiography (postero-anterior (PA) view) is routinely used to evaluate the tibiofemoral joint. Patellofemoral (PF) joint is the most frequently affected compartment by OA and yet it has not been studied much compared to tibiofemoral osteoarthritis (TFOA)[1]. Patellofemoral osteoarthritis (PFOA) is both highly prevalent[2,3] and clinically important because it is more strongly associated with knee OA symptoms than tibiofemoral OA4. The patellofemoral joint cannot be evaluated from the most commonly used frontal plane radiography. Previous studies suggested that PF joint should routinely be considered in knee OA studies by obtaining multiple radiographic views of the knee[13,14]; otherwise 4e7% of OA cases would be missed[15]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have