Ultraleap extension measurements demonstrated acceptable-to-high reliability and moderate-to-strong correlations with manual goniometry in most joints, but Bland-Altman analyses revealed systematic and proportional biases indicating limited agreement for absolute ROM values.
Key Findings
Results
Intra-rater reliability for finger flexion measured by Ultraleap 3D motion capture showed variable intraclass correlation coefficients across joints.
Flexion intra-rater ICCs ranged from 0.576 to 0.896 across finger joints
Extension intra-rater ICCs were ≥0.81 for all joints except the thumb interphalangeal (IP) joint
Measurements were conducted in stroke patients with upper limb hemiplegia over a 3-day period
The system used was the Ultraleap Stereo IR 170 (Ultraleap Ltd., Bristol, UK)
Results
Inter-rater reliability for finger flexion measured by Ultraleap also showed variability, while extension demonstrated more consistent reliability.
Flexion inter-rater ICCs ranged from 0.426 to 0.841
Extension inter-rater ICCs were ≥0.81 for all joints except the thumb IP joint
The thumb IP joint was a consistent exception to acceptable reliability thresholds for both intra- and inter-rater analyses
Results
Criterion-related validity of Ultraleap compared to manual goniometry was substantially different for flexion versus extension measurements.
Correlation coefficients for flexion ranged from -0.0706 to 0.493, indicating poor to weak validity
Correlation coefficients for extension ranged from 0.698 to 0.814 (excluding the thumb IP joint), indicating moderate-to-strong correlations
The thumb IP joint was excluded from the extension validity summary due to poor performance
Goniometer measurements served as the criterion standard for validity comparisons
Results
Bland-Altman analyses revealed systematic and proportional biases between Ultraleap and manual goniometry measurements, indicating limited agreement for absolute ROM values.
Systematic bias was identified, meaning Ultraleap measurements consistently differed from goniometer measurements in a directional manner
Proportional bias was also present, meaning the disagreement between methods varied with the magnitude of the measurement
These biases indicate the two measurement systems cannot be directly interchanged for absolute ROM assessment
The Bland-Altman method was used to assess agreement beyond correlation coefficients
Background
The Fugl-Meyer Assessment (FMA) finger subscale and manual goniometry were identified as having limitations that motivated investigation of 3D motion capture alternatives.
The finger FMA has a narrow scoring range, limiting its sensitivity for detecting change
Goniometer ROM measurements were noted to lack reliability
These limitations prompted the investigation of the Ultraleap 3D motion capture system as an alternative measurement tool for stroke patients
Conclusions
The Ultraleap system was assessed as a potentially useful supplementary tool for repeated finger extension measurements in clinical settings despite its inability to replace goniometry for absolute values.
The system may serve as 'an efficient supplementary tool for repeated measurements of finger extension in clinical settings'
Its use is considered appropriate for tracking relative changes rather than absolute ROM values
Clinical applicability is recommended to be limited to extension measurements, given poor flexion validity and reliability
What This Means
This research suggests that a commercial 3D motion capture device (Ultraleap Stereo IR 170) can reliably measure finger extension movements in stroke patients with arm weakness, but is less reliable and valid for measuring finger flexion. The study tested the device by having multiple raters measure finger range of motion in stroke patients over three days and comparing those results to measurements taken with a standard handheld goniometer (a protractor-like device used in clinics). For extension movements, the device showed good consistency across repeated measurements and reasonable agreement with the goniometer. For flexion movements, consistency and agreement were much weaker and variable across finger joints.
However, even where the two methods correlated reasonably well, detailed statistical analyses (Bland-Altman plots) revealed that the Ultraleap system does not give the same absolute numerical values as the goniometer — it shows both consistent offsets and errors that grow with larger measurements. This means a clinician could not simply swap one method for the other and expect to get equivalent numbers.
This research suggests that the Ultraleap 3D sensor could be a practical supplementary tool in rehabilitation settings — particularly for tracking changes in finger extension over time without needing a clinician to manually apply a goniometer for every measurement. This could make repeated assessments faster and easier in busy clinical environments, though it should not replace standard goniometry when precise absolute angle values are needed, and its use for flexion measurement requires further improvement.
Moriguchi S, Takebayashi T, Arata J, Bandara D, Matsumoto R, Shimada S, et al.. (2026). The Reliability and Validity of Active Range of Motion Measurement for Fingers Using Three-Dimensional Motion Analysis.. Sensors (Basel, Switzerland). https://doi.org/10.3390/s26196162