Skip to content

Prediction of Lower Extremity Injuries Using Lower Limb-worn Inertial Measurement Units

Prediction of Pre-existing Lower Extremity Injuries Using Lower Limb-worn Inertial Measurement Units

Status
Completed
Phases
Unknown
Study type
Observational
Source
ClinicalTrials.gov
Registry ID
NCT07289828
Acronym
PredKnee
Enrollment
108
Registered
2025-12-17
Start date
2024-01-29
Completion date
2024-06-04
Last updated
2025-12-17

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

Conditions

Lower Extremity Injuries

Keywords

prediction, ankle injury, knee injury, inertial sensor, artificial intelligence

Brief summary

This study analyses questionnaires and inertial sensor data from 108 sports science students regarding previous lower extremity injuries, sports activity, and preventive measures, combined with the prospective development of an AI-based prediction algorithm. Inertial sensor data were collected during walking and running on a standard 400 m track, with sensors placed on the thighs and ankles, and heart rate recorded via smartwatch. Participants also completed questionnaires on previous injuries, comorbidities, sports activity, and prevention. The aim is to use the anonymized data to identify gait and running patterns associated with prior knee and ankle injuries using AI analysis, and to correlate these findings with sports activity and preventive measures. Hypothesis: Prior lower extremity injuries leave specific gait and running patterns detectable by inertial sensors and AI-based analysis.

Detailed description

In this study, analysis of questionnaires and inertial sensor data from 108 sports science students is conducted with regard to previous injuries of the lower extremities, their sports activities, and a possible association with performed preventive measures, along with the prospective development of an AI-based prediction algorithm to detect prior injuries of the lower extremities. In all participants, inertial sensor data were collected during walking and running on a defined track (5 minutes walking, 5 minutes running, 5 minutes walking on a standard 400 m oval tartan track). Sensors were placed on the lateral aspects of both thighs above the knee joint and on the lateral aspects of both ankles above the lateral malleolus. In addition, participants wore a smartwatch on the left wrist to record heart rate. Furthermore, participants completed questionnaires regarding previous injuries, comorbidities, sports activity, and preventive measures undertaken. The aim of the current analysis is to utilize the anonymized data from questionnaires and inertial sensors to identify gait and running patterns indicative of previous injuries of the lower extremities (knee and ankle) by means of an AI algorithm, and to correlate these findings with reported sports activities and preventive measures. Hypothesis: Previous injuries of the lower extremities (particularly of the knee and ankle) result in specific gait and running patterns measurable by inertial sensors, which can be identified through AI-based analysis.

Interventions

OTHERIMU Data collection

Participants were walking and running while wearing inertial measurement units (IMU) on both legs. The IMUs (MetaMotionS sensor by Mbientlab) where recording at 100Hz (accelerometer and gyroscope) and 25Hz (magnetometer).

OTHERQuestionnaire

On the day of the examination, the test subjects completed a standardized questionnaire on previous injuries, type of sport, sporting activity, and preventive measures.

Sponsors

Technical University of Munich
Lead SponsorOTHER

Study design

Observational model
COHORT
Time perspective
RETROSPECTIVE

Eligibility

Sex/Gender
ALL
Age
18 Years to 60 Years
Healthy volunteers
Yes

Inclusion criteria

* Subjectively healthy participants * Age: \>18 years and under 60 years * German language skills sufficient to follow the exercise instructions and complete the questionnaires

Exclusion criteria

* Age \<18 years or \>60 years * Recent injuries and trauma to the lower extremities (less than 6 months ago) * Acute malignant disease * Acute inflammatory disease * Lack of German language skills * Lack of cardiopulmonary endurance for testing

Design outcomes

Primary

MeasureTime frameDescription
IMU dataat baselineTime-stamped, unfiltered, device-coordinate-based 3-axis IMU data (Ax, Ay, Az) from four IMUs, placed laterally on both thighs (above the knee joint) and on both ankles (above the lateral malleolus).

Secondary

MeasureTime frameDescription
questionnaire injuries lower extremityBaselineParticipants reported previous injuries and illnesses affecting the lower extremities on a standardized questionnaire.
questionnaire sports activityBaselineParticipants indicated their sport and intensity level on a standardized questionnaire.
questionnaire preventionbaselineParticipants indicated on a standardized questionnaire whether preventive measures to avoid sports injuries were being implemented.

Countries

Germany

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026