Home-based rehabilitation training with human key point detection for chronic low back pain patients: a randomized controlled trial protocol

BACKGROUND: Core stability exercise (CSE) is a globally acknowledged intervention for managing chronic low back pain. However, the sustained adherence of patients with chronic low back pain to CSE can be challenging, mainly due to the absence of supervision and guidance from physical therapists during home-based exercise sessions. Consequently, exercise compliance tends to decline, resulting in suboptimal long-term effectiveness of the intervention. In this trial, our primary aim is to evaluate the potential therapeutic equivalence between home-based rehabilitation training employing key point identification technology and exercise guidance administered in a hospital setting. METHODS: In this trial, we will randomly assign 104 adults with chronic low back pain (CLBP) to either an intervention or control group, with 52 participants in each group. Both interventions will consist of three weekly 0.5-h sessions of core stability exercise (CSE). The intervention group will engage in home rehabilitation training utilizing key identification technology for movement, while the control group will perform supervised exercises in a hospital setting. Outcome assessments will be conducted at 4 weeks and 16 weeks after randomization. The primary outcome measure will be the change in pain intensity based on numeric rating scale (NRS scores) from baseline to 4 weeks. Secondary outcomes will include changes in physical function (measured by the Oswestry Disability Index (ODI)) and lumbar spine mobility as well as activity participation and treatment satisfaction. DISCUSSION: If home-based rehabilitation method is demonstrated to be non-inferior or even superior to traditional face-to-face exercise guidance, it could significantly advance the adoption of digital medical care and contribute to improving the overall health of the population. TRIAL REGISTRATION: NCT05998434. Registered on 16 August 2023.