Document resource
Introduction Adolescent idiopathic scoliosis (AIS) requires long-term conservative management to prevent curve progression. While physiotherapeutic scoliosis-specific exercises, specifically the Schroth method, are considered the gold standard for conservative treatment, their clinical efficacy is often limited by accessibility barriers, high costs and suboptimal treatment adherence. This study aims to evaluate the efficacy of a novel artificial intelligence (AI)-based digital therapeutic system, which uses computer vision for remote, personalised posture analysis and adaptive exercise prescription compared with traditional outpatient Schroth therapy.Methods and analysis This parallel-group randomised controlled trial will be conducted at Guangzhou Women and Children’s Medical Center (Guangzhou, China). 300 adolescents aged 10–18 years with AIS who present with a Cobb angle between 10° and 30° and a Risser sign of 0–2 will be recruited and randomised in a 1:1 ratio into an intervention group and a control group. The intervention group will use a smartphone application to capture standardised bi-weekly images. These images will be processed by an AI algorithm to classify curve patterns and assign personalised exercise modules with adaptive dosing ranging from maintenance to high-intensity levels. The control group will receive standard outpatient Schroth care. The primary outcome is the absolute change in the major curve Cobb angle from baseline to 6 months. Secondary outcomes include the angle of trunk rotation, trunk appearance perception, Scoliosis Research Society-22 Revised (SRS-22r) quality-of-life scores and adherence rates. Statistical analysis will follow the intention-to-treat principle using linear mixed models to account for repeated measures.Ethics and dissemination Ethical approval has been obtained from the Medical Ethics Committee of Guangzhou Women and Children’s Medical Center (Guangzhou, China) (approval no. [2025]497A01). Written informed assent and consent will be obtained from participants and their legal guardians respectively. Results will be disseminated through peer-reviewed journals and international conferences.Trial registration number ClinicalTrials.gov, NCT07341633.