This paper proposes a new approach to automatically quantify the severity of knee osteoarthritis (OA) from radiographs using deep convolutional neural networks (CNN). Clinically, knee OA severity is assessed using Kellgren & Lawrence (KL) grades, a five point scale. Previous work on automatically predicting KL grades from radiograph images were based on training shallow classifiers using a variety of hand engineered features. We demonstrate that classification accuracy can be significantly improved using deep convolutional neural network models pre-trained on ImageNet and fine-tuned on knee OA images. Furthermore, we argue that it is more appropriate to assess the accuracy of automatic knee OA severity predictions using a continuous distance-based evaluation metric like mean squared error than it is to use classification accuracy. This leads to the formulation of the prediction of KL grades as a regression problem and further improves accuracy. Results on a dataset of X-ray images and KL grades from the Osteoarthritis Initiative (OAI) show a sizable improvement over the current state-of-the-art.

Cutting ability is an important aspect of many team sports, however, the biomechanical determinants of cutting performance are not well understood. This study aimed to address this issue by identifying the kinetic and kinematic factors correlated with the time to complete a cutting maneuver. In addition, an analysis of the test-retest reliability of all biomechanical measures was performed. Fifteen (n = 15) elite multidirectional sports players (Gaelic hurling) were recruited, and a 3-dimensional motion capture analysis of a 75° cut was undertaken. The factors associated with cutting time were determined using bivariate Pearson's correlations. Intraclass correlation coefficients (ICCs) were used to examine the test-retest reliability of biomechanical measures. Five biomechanical factors were associated with cutting time (2.28 ± 0.11 seconds): peak ankle power (r = 0.77), peak ankle plantar flexor moment (r = 0.65), range of pelvis lateral tilt (r = -0.54), maximum thorax lateral rotation angle (r = 0.51), and total ground contact time (r = -0.48). Intraclass correlation coefficient scores for these 5 factors, and indeed for the majority of the other biomechanical measures, ranged from good to excellent (ICC >0.60). Explosive force production about the ankle, pelvic control during single-limb support, and torso rotation toward the desired direction of travel were all key factors associated with cutting time. These findings should assist in the development of more effective training programs aimed at improving similar cutting performances. In addition, test-retest reliability scores were generally strong, therefore, motion capture techniques seem well placed to further investigate the determinants of cutting ability.

Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet

Motion analysis technologies have been widely used to monitor the potential for injury and enhance athlete performance. However, most of these technologies are expensive, can only be used in laboratory environments, and examine only a few trials of each movement action. In this paper, we present a novel ambulatory motion analysis framework using wearable inertial sensors to accurately assess all of an athlete’s activities in real training environment. We first present a system that automatically classifies a large range of training activities using the discrete wavelet transform (DWT) in conjunction with a random forest classifier. The classifier is capable of successfully classifying various activities with up to 98% accuracy. Second, a computationally efficient gradient descent algorithm is used to estimate the relative orientations of the wearable inertial sensors mounted on the shank, thigh, and pelvis of a subject, from which the flexion–extension knee and hip angles are calculated. These angles, along with sacrum impact accelerations, are automatically extracted for each stride during jogging. Finally, normative data are generated and used to determine if a subject’s movement technique differed to the normative data in order to identify potential injury-related factors. For the joint angle data, this is achieved using a curve-shift registration technique. It is envisaged that the proposed framework could be utilized for accurate and automatic sports activity classification and reliable movement technique evaluation in various unconstrained environments for both injury management and performance enhancement.