Skip to content

Developing an AI tool to evaluate yoga poses and provide scoring

Design and validate an AI-enabled Yoga Posture Evaluation Platform - NIL

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
CTRI
Registry ID
CTRI/2025/12/098745
Enrollment
30
Registered
2025-12-10
Start date
Unknown
Completion date
Unknown
Last updated
2026-01-12

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

None listed

Interventions

Intervention1: NIL: NIL Intervention2: Nil: Nil Intervention3: Nil: Nil

Sponsors

School Of Physiotherapy - RK University
Lead Sponsor
Shivani Rajesh Sachdev
Collaborator

Eligibility

Inclusion criteria

Inclusion criteria: 1. Are volunteers aged 9 years and above. 2. Can independently perform Chakrasana pose without physical assistance for posture assessment. 3. Provide written privacy and image-use consent for using their yoga pose photographs/videos for image processing in an AI-enabled yoga posture evaluation platform. 4. In the case of minors, have written parental or legal guardian consent in addition to the child s assent wherever appropriate.

Exclusion criteria

Exclusion criteria: Not Applicable

Design outcomes

Primary

MeasureTime frame
Accuracy of the AI-enabled posture evaluation system for Chakrasana. Outcome Measures: Sensitivity, specificity, percentage accuracy, and error rate comparing AI-generated posture evaluation with expert instructor evaluation. Timepoint: Baseline (pre-test), Immediate post-test (same day after system training), and 2 weeks after baseline.

Secondary

MeasureTime frame
Reliability of the AI System (including both test retest reliability & inter-rater reliability.) Outcome Measures: Intraclass Correlation Coefficient (ICC) from repeated AI assessments of two consecutive still images (test retest reliability), & ICC with Kappa statistics comparing AI evaluations with two independent expert instructor assessments (inter-rater reliability).Timepoint: Baseline (within-session) & 2 weeks after baseline. ;Validity of the System (Construct & criterion validity of the AI system for posture assessment.) Outcome Measures: Construct validity: Ability of the AI to correctly identify alignment deviations. Criterion validity: Comparison of AI outputs with the gold-standard expert assessment. Timepoint: Baseline & 2 weeks after baseline.

Countries

India

Contacts

Public ContactShivani Rajesh Sachdev

School of Physiotherapy, RK University

priyanshu.rathod@rku.ac.in9426803108

Outcome results

None listed

Source: CTRI (via WHO ICTRP) · Data processed: Feb 4, 2026