Objective monitoring of home-based flexibility and strength training using wearable inertial sensors: validity, reliability, inter-device agreement and feasibility – a study protocol

Abstract

Introduction & Purpose

Training interventions targeting physical capacities, such as muscle strength and joint flexibility, require adequate execution and training-load monitoring to enable individualized adjustments and accurate interpretation of adaptations. Short-term interventions (< 6 weeks) can often be performed under direct supervision, but this approach becomes impractical for long-term interventions. Home-based training is therefore the most realistic solution for studying long-term adaptations, yet maintaining motivation, adherence, and exercise quality over months of training remains challenging. This is important because exercise adherence depends not only on completing prescribed sessions, but also on performing the required quantity and quality of exercise (Argent et al., 2018). Although digital interventions may improve adherence more consistently in short-term home-exercise interventions, longer-term effects remain uncertain (Lang et al., 2022). Objective remote monitoring is therefore a key methodological requirement for studying long-term training adaptations, not merely a digital convenience. The aim of the planned research is to develop and validate an IMU-based system to objectively monitor home-based flexibility and strength training.

Methods

The system includes two exercise-monitoring modules. The stretching module quantifies stretch angle from pedal rotation (stretching device), defined as the change in pedal orientation from the unloaded resting position. The system automatically extracts imposed stretch angle, median and peak hold angle, hold-angle variability, valid hold duration, and repetition count. The eccentric plantarflexion module automatically quantifies repetition count within prescribed sets, eccentric and concentric phase durations, rest intervals, while external resistance is entered by the participant and recorded alongside the IMU-derived metrics. The experimental protocol will include 15 healthy adults of both sexes completing standardized stretching and eccentric plantarflexion protocols during two identical laboratory sessions. Criterion validity will be examined by comparing IMU-derived outcomes with 3D motion capture during both exercise tasks. Reliability will be assessed by repeating the protocol across two days. A mechanical validation block, in which the pedal is manually moved to reproducible positions, will isolate sensor and algorithm reliability from biological variability. Inter-device agreement will be evaluated by mounting two IMUs simultaneously on the same pedal and comparing both sensors against each other and against motion capture. Finally, participants will complete 3-5 unsupervised home sessions with different combinations of repetitions, sets and stretch hold durations, reporting the completed training via a training log to compare self-reported and IMU-derived exercise metrics and evaluate real-world feasibility. Agreement and reliability will be assessed using bias, root mean square error, mean absolute error, Bland–Altman limits of agreement, and intraclass correlation coefficients.

Results

Preliminary 2D video analysis showed close agreement between IMU-derived and video-derived pedal angles in rest and stretched positions, supporting feasibility before full motion-capture validation.

Discussion

This framework addresses a central limitation of long-term home-based interventions: the lack of objective information about exercise dose and execution quality.

Conclusion

This open-source exercise-tracking framework provides a methodological foundation for future long-term home-based intervention studies on strength and flexibility adaptations.

References

Argent, R., Daly, A., & Caulfield, B. (2018). Patient involvement with home-based exercise programs: Can connected health interventions influence adherence? JMIR mHealth and uHealth, 6(3), Article e47.https://doi.org/10.2196/mhealth.8518

Lang, S., McLelland, C., MacDonald, D., & Hamilton, D. F. (2022). Do digital interventions increase adherence to home exercise rehabilitation? A systematic review of randomised controlled trials. Archives of Physiotherapy, 12, Article 24. https://doi.org/10.1186/s40945-022-00148-z