The wrist and lower back IMU showed comparable RMSE values across all MET intensity levels, with both IMU positions showing lower RMSE values than the ankle position, and tremor affected RMSE negatively with the lower back position potentially favorable for those with tremor.
Key Findings
Results
Wrist and lower back IMU placements showed comparable accuracy for physical activity assessment across all MET intensity levels, both outperforming ankle placement.
Root mean square error (RMSE) values were calculated between scaled normalized IMU-derived ENMO and normalized MET values for each of nine IMU-MET combinations (three IMU positions × three MET intensity levels)
Both wrist and lower back positions showed lower RMSE values than the ankle position across all MET intensity levels
Study population consisted of 25 newly diagnosed persons with PD not yet taking disease-specific medication
A total of 8,494 15-minute epochs were used for the analysis
Results
Tremor negatively affected RMSE values in IMU-based physical activity assessment, with the lower back position potentially being slightly favorable for participants with tremor.
Tremor affected RMSE negatively across IMU positions
The lower back position may be slightly favorable for the assessment of physical activity in those with tremor compared to wrist or ankle placement
Participants wore IMUs on the most affected ankle, wrist, and lower back simultaneously for two weeks
The study focused on drug-naïve PD patients, representing the earliest clinically evident phase of the disease
Methods
The study recruited a rare cohort of 25 drug-naïve newly diagnosed persons with PD who wore IMUs at three body locations simultaneously for two weeks while maintaining a physical activity diary.
Participants were enrolled within the FAIRPARK-II trial
IMUs were worn on the most affected ankle, wrist, and the lower back
Participant-reported physical activity was transformed into Metabolic Equivalents of Tasks (METs) in 15-minute intervals using The Compendium of Physical Activities
Data inclusion required at least 3 days with at least four simultaneous 15-minute epochs of both valid IMU and diary data within the time window of 9:00 to 18:00 per participant
Euclidean Norm Minus One (ENMO) values were calculated and averaged over 15-minute intervals for the IMU data
Results
Physical activity assessment using IMUs in drug-naïve PD patients is feasible across three body locations, but placement location differentially affects measurement accuracy.
Nine IMU-MET combinations were examined (three IMU positions × three MET intensity levels)
The cohort represented persons with PD not taking disease-specific medication, reflecting the earliest clinically evident phase of PD
The authors state the dataset 'provides novel insights into the assessment of physical activity during the earliest clinically evident phase of Parkinson's disease without disease-specific medication'
Findings are noted to 'inform future clinical trials and observational studies'
What This Means
This research examined how the placement of wearable motion sensors (called inertial measurement units, or IMUs) on the body affects the accuracy of measuring physical activity in people newly diagnosed with Parkinson's disease who were not yet on any Parkinson's medication. Twenty-five participants wore sensors on their wrist, their most-affected ankle, and their lower back for two weeks, while also keeping a diary of their physical activities. The diary data was converted into a standard measure of physical activity intensity called Metabolic Equivalents of Tasks (METs), and the sensor data was compared to these diary records to see how well each sensor location tracked actual activity levels.
The study found that wrist and lower back sensors performed similarly well in capturing physical activity, and both outperformed the ankle sensor across different levels of activity intensity. Tremor — a common symptom of Parkinson's disease — made the measurements less accurate overall, but the lower back sensor appeared to be slightly better suited for people who experience tremor, likely because back movements are less directly affected by limb tremor than wrist or ankle movements.
This research suggests that for people in the earliest stages of Parkinson's disease who have not yet started medication, wrist or lower back placement of motion sensors is preferable to ankle placement when measuring physical activity. The findings are particularly relevant for designing future clinical trials and research studies in Parkinson's disease, where accurate measurement of physical activity is important since physical activity is known to help reduce the burden of the disease.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Maetzler C, Abedinifar M, Vinod V, Welzel J, Schaeffer E, Yilmaz R, et al.. (2026). Body placement of inertial measurement units differentially affects physical activity assessment accuracy in drug-naïve Parkinson's disease.. Scientific reports. https://doi.org/10.1038/s41598-026-55099-3