Search

Abstract Background The development of computer-assisted technologies to diagnose anterior cruciate ligament (ACL) injury by analyzing knee magnetic resonance images (MRI) would be beneficial, and convolutional neural network (CNN)-based deep learning approaches may offer a solution. This study aimed to evaluate the accuracy of a CNN system in diagnosing ACL ruptures by a single slice from a knee MRI and to compare the results with that of experienced human readers. Methods One hundred sagittal MR images from patients with and without ACL injuries, confirmed by arthroscopy, were cropped and used for the CNN training. The final decision by the CNN for intact or torn ACL was based on the probability of ACL tear on a single MRI slice. Twelve board-certified physicians reviewed the same images used by CNN. Results The sensitivity, specificity, accuracy, positive predictive value and negative predictive value of the CNN classification was 91.0%, 86.0%, 88.5%, 87.0%, and 91.0%, respectively. The overall values of the physicians’ readings were similar, but the specificity was lower than the CNN classification for some of the physicians, thus resulting in lower accuracy for the human readers. Conclusions The trained CNN automatically detected the ACL tears with acceptable accuracy comparable to that of human readers.

Objective: To elucidate the possible role of MRI-detected osteophytes as a predictive imaging biomarker for knee osteoarthritis (KOA). Design: Subjects (n ​= ​303) were selected according to the following inclusion criteria from the Osteoarthritis Initiative (OAI) data set: (1) ​

Background: Older adults often overestimate their motor performance, which represents a serious safety hazard. The cause of this self-overestimation is to date, not yet fully established. Thus, the present study aimed to reveal the factors associated with self-overestimation by focusing on motor function.Methods: This study included 105 community-dwelling older adults [20 males, median (25, 75 percentile) age: 73.00 (69.50, 77.50)]. Participants were assessed for errors in their self-estimation using a two-step test. They estimated the two-step distance that could be reached with maximum effort. Thereafter, they performed the actual two-step action. Participants were comprehensively assessed for motor function by various tests (i.e., 10-meter Walking Test, Timed Up and Go Test, postural stability, and muscle strength). They were then divided into two groups (the self-underestimation or self-overestimation group) and their motor performances were compared. Multiple linear regression analysis was then utilized to investigate the relationship between self-estimation error and motor function.Results: Significant differences were found between the two groups regarding age, weight, actual two-step distance, and the time required for the Timed Up and Go Test and 10-meter Walking Test (p < 0.05). The regression analysis showed that self-estimation error was significantly related to the result of the 10-meter Walking Test (beta = 0.24, p = 0.011).Conclusions: The self-overestimation of motor performance, which is likely to lead to several dangers (i.e., falling or obstacle collision), was related to walking ability. Consequently, the results showed that the 10-meter Walking Test would assist in detecting the self-overestimation of motor performance.

[This corrects the article DOI: 10.1371/journal.pone.0202228.].

This study aimed to investigate (a) motor planning difficulty by using a two-step test in Parkinson's disease (PD) compared with age-matched healthy subjects and (b) the relationship between motor planning difficulty and clinical factors. The two-step test was performed by 58 patients with PD with Hoehn & Yahr (H&Y) stage I-IV and 110 age-matched healthy older adult controls. In the two-step test, the participants estimated the two-step distance with maximum effort. Subsequently, they performed the actual two-step trial to measure the actual maximum distance. We calculated the accuracy of the estimation (estimated distance minus actual distance). In both groups, subjects who estimated >5 cm were defined as the overestimation group, and those who estimated