Skip to content

AI-based gait analysis: Development of personalized models based on kinematic data in patients with orthopedic disorders

AI-based gait analysis: Development of personalized models based on kinematic data in patients with orthopedic disorders

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
DRKS
Registry ID
DRKS00038078
Enrollment
300
Registered
2025-10-31
Start date
2025-11-26
Completion date
Unknown
Last updated
2026-04-27

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

Conditions

Other orthopedic conditions of the lower limb. M16 M17 M25.5

Interventions

Group 1: Observation of a cohort of patients with orthopedic complaints. Performing markerless motion capture recordings of gait and force plate measurements. No therapeutic or drug intervention.

Sponsors

OTH Regensburg
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: Adults (= 18 years) attending the orthopedic outpatient clinic. Ability to walk independently without walking aids. Understanding of study information and provision of written informed consent. For non-German speakers: sufficient English comprehension (translated documents available).

Exclusion criteria

Exclusion criteria: Neurological or other non-orthopedic disorders affecting gait. Inability to walk independently. Lack of consent or insufficient understanding of study information.

Design outcomes

Primary

MeasureTime frame
The primary outcome is the validity and plausibility of the developed personalized gait score and the predictive accuracy of the AI-based model for ground-reaction forces. Measurements are performed once during the study visit using a markerless motion-capture system (multi-camera video capture) in combination with force plates. Spatio-temporal and kinematic gait parameters (e.g., joint angle trajectories, step time, stride length, gait asymmetry, ground-reaction forces) are recorded. Model performance will be assessed by comparing predicted and measured values using mean absolute error (MAE), mean absolute percentage error (MAPE), and correlation coefficients (r).

Secondary

MeasureTime frame
Correlation between kinematic parameters, patient-reported outcome measures (KOOS, EQ-5D), and clinical diagnosis; technical feasibility and user acceptance; agreement between motion-capture and wearable data.

Countries

Germany

Contacts

Public ContactStefano Pagano

Orthopädische Klinik für die Universität Regensburg

stefano.pagano@ukr.de09405180

Outcome results

None listed

Source: DRKS (via WHO ICTRP) · Data processed: May 1, 2026