Effects of a Computer Vision-Based Exercise Application for People With Knee Osteoarthritis: Randomized Controlled...

Area of Science:

• Digital Health
• Rehabilitation Medicine
• Computer Vision in Healthcare

Background:

• Exercise is a key treatment for knee osteoarthritis (KOA), improving symptoms and function.
• Personalized exercise programs enhance physical fitness and well-being.
• Digital health apps with computer vision (CV) offer scalable, accessible, personalized KOA rehabilitation.

Purpose of the Study:

• To evaluate a CV-graded exercise app's impact on KOA clinical outcomes over 6 weeks.
• To compare the app's effectiveness against conventional video-based exercise education.

Main Methods:

• A 6-week randomized controlled trial with 60 KOA patients (aged 60-80).
• Two groups: graded exercise app (n=32) and video education brochure (n=28).
• Primary outcomes: WOMAC for pain, physical function, stiffness. Secondary: affect, self-efficacy, quality of life, user experience.

Main Results:

• The CV exercise app group showed significant improvements in physical function (P=.02) and self-efficacy (P=.04).
• Both groups had non-significant improvements in pain and stiffness.
• The app group reported positive user experience, usability, and engagement.

Conclusions:

• CV-based graded exercise app effectively enhances KOA patients' physical function and self-efficacy.
• This digital tool shows potential for superior personalized exercise rehabilitation compared to traditional methods.
• Further research should assess long-term efficacy and community-based replicability.