Knee Osteoarthritis Classification Using 3D CNN and MRI

Carmine Guida,Ming Zhang,Juan Shan

doi:10.3390/app11115196

Abstract

Osteoarthritis (OA) is the most common form of arthritis and can often occur in the knee. While convolutional neural networks (CNNs) have been widely used to study medical images, the application of a 3-dimensional (3D) CNN in knee OA diagnosis is limited. This study utilizes a 3D CNN model to analyze sequences of knee magnetic resonance (MR) images to perform knee OA classification. An advantage of using 3D CNNs is the ability to analyze the whole sequence of 3D MR images as a single unit as opposed to a traditional 2D CNN, which examines one image at a time. Therefore, 3D features could be extracted from adjacent slices, which may not be detectable from a single 2D image. The input data for each knee were a sequence of double-echo steady-state (DESS) MR images, and each knee was labeled by the Kellgren and Lawrence (KL) grade of severity at levels 0–4. In addition to the 5-category KL grade classification, we further examined a 2-category classification that distinguishes non-OA (KL ≤ 1) from OA (KL ≥ 2) knees. Clinically, diagnosing a patient with knee OA is the ultimate goal of assigning a KL grade. On a dataset with 1100 knees, the 3D CNN model that classifies knees with and without OA achieved an accuracy of 86.5% on the validation set and 83.0% on the testing set. We further conducted a comparative study between MRI and X-ray. Compared with a CNN model using X-ray images trained from the same group of patients, the proposed 3D model with MR images achieved higher accuracy in both the 5-category classification (54.0% vs. 50.0%) and the 2-category classification (83.0% vs. 77.0%). The result indicates that MRI, with the application of a 3D CNN model, has greater potential to improve diagnosis accuracy for knee OA clinically than the currently used X-ray methods.

Highlights

The most common form of joint disorder in the United States is osteoarthritis (OA) [1].Knee OA can cause pain and is the number one disease at causing loss of ability to perform daily activities such as walking and stair climbing [2]
The fully convolutional neural network (FCN) method was found to be highly accurate in determining regions of interest (ROI), and when combined with a convolutional neural networks (CNNs) for classification, the method achieved an accuracy of 61.9% [16]
In addition to magnetic resonance imaging (MRI), we studied traditional X-ray images, with an interest of finding out which imaging modality coupled with the modern CNNs can achieve better accuracy for knee OA diagnosis

Summary

Introduction

Knee OA can cause pain and is the number one disease at causing loss of ability to perform daily activities such as walking and stair climbing [2]. Knee OA is associated with age [3]. Is characterized by the loss of articular cartilage volume [4]. OA is viewed as a “wholeorgan” disorder, manifesting damage to a range of articular structures, especially the hyaline cartilage, meniscus, periarticular bone, ligaments, and tendons [5]. Despite its importance for public health, we have no interventions that effectively modify the OA disease process [6]. The absence of useful biomarkers to detect OA progression is a major technological obstacle to the development of treatment and prevention of knee OA [7]

Methods

Results

Discussion

Conclusion