Reliability and validity of computer vision-based markerless human pose estimation for measuring hip and knee range of motion

Abstract

Telerehabilitation requires accurate joint range of motion (ROM) measurement methods. The aim of this study was to evaluate the reliability and validity of a computer vision (CV)-based markerless human pose estimation (HPE) application measuring active hip and knee ROMs. For this study, the joint ROM of 30 healthy young adults (10 females, 20 males) aged 20-33 years (mean: 22.9 years) was measured, and test-retests were assessed for reliability. For validity evaluation, the CV-based markerless HPE application used in this study was compared with an identical reference picture frame. The intraclass correlation coefficient (ICC) for the CV-based markerless HPE application was 0.93 for active hip inner rotation, 0.83 for outer rotation, 0.82 for flexion, 0.82 for extension, and 0.74 for knee flexion. Correlations (r) of the two measurement methods were 0.99 for hip-active inner rotation, 0.98 for outer rotation, 0.87 for flexion, 0.85 for extension, and 0.90 for knee flexion. This study highlights the potential of a CV-based markerless HPE application as a reliable and valid tool for measuring hip and knee joint ROM. It could offer an accessible solution for telerehabilitation, enabling ROM monitoring.

Keywords: computer vision; patient rehabilitation; reliability; telemedicine.

FIGURE 1

Measurement protocol: (a) Active flexion of hip and knee joint; (b) active extension of hip joint; (c) active inner‐ and (d) outer rotation of hip joint.

FIGURE 2

Study design, one participant was measured two times with an interval of 24 h. Joint angles were registered from the CV‐based markerless HPE application and an identical reference picture frame.

FIGURE 3

Key points and angles of interest for measuring hip (a) flexion and (b) extension. The angle is calculated as β = 180 − α. Blue dots show key points from a lateral view, the middle of the knee joint, hip joint, and shoulder joint.

FIGURE 4

Key points and angles of interest for measuring knee range of motion. The angle is calculated as β = 180 − α. Blue dots show key points from a lateral view: the middle of the ankle joint, knee joint, and hip joint.

FIGURE 5

Key points and angles of interest for measuring hip (a) inner and (b) outer rotation. Angle θ is calculated using trigonometrical functions. Blue dots show key points from the frontal view, the middle of the ankle joint, and the knee joint.

FIGURE 6

Photograph of (a) test setup when a participant performs active hip flexion and (b) computer screen view of our CV‐based markerless HPE application.

FIGURE 7

Reference picture with manually drawn markers for (a) active flexion of hip, (b) active flexion of knee joint, (c) active extension of hip joint, (d) active inner‐ and (e) outer rotation of hip joint.

FIGURE 8

Correlation between our CV‐based markerless HPE application and the reference picture for (a) hip active flexion, (b) hip active extension, (c) hip active inner rotation, (d) hip active outer rotation and (e) knee active flexion, r = Person's correlation coefficient.

FIGURE 9

Bland–Altman plots for (a) hip active flexion, (b) hip active extension, (c) hip active inner rotation, (d) hip active outer rotation and (e) knee active flexion. The outer lines represent 95% limits of agreement. The middle line represents the mean of the differences between the two measurement methods.