Related Experiment Video
Updated: Jul 11, 2025
05:51
Assessing the Accuracy of Fitness Smartwatch Data for Cardiovascular and Physical Activity Monitoring: A Validation Study in Digital Health
Published on: February 21, 2025
486
Open-Source, Step-Counting Algorithm for Smartphone Data Collected in Clinical and Nonclinical Settings: Algorithm
Marcin Straczkiewicz1, Nancy L Keating2,3, Embree Thompson4
1Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, United States.
JMIR Cancer
|November 15, 2023
Summary
This study validates an open-source smartphone step-counting method, showing it reliably estimates steps across various conditions and populations, offering a scalable alternative to commercial trackers.
Area of Science:
- Biomedical Engineering
- Digital Health
- Wearable Technology
Background:
- Step counts are vital for public health and clinical research.
- Commercial activity trackers have limitations in reproducibility and scalability.
- Smartphones offer a promising, accessible alternative for step-counting.
Purpose of the Study:
- To evaluate an open-source smartphone step-counting method.
- Validation was performed against cross-body, visually assessed, and commercial wearable data.
- The study assessed performance under diverse measurement conditions.
Main Methods:
- Utilized 8 independent datasets from smartphones and accelerometers.
- Employed a previously published smartphone step-counting algorithm using raw accelerometer data.
- Bland-Altman analysis calculated mean bias and limits of agreement.
Main Results:
- Cross-body validation showed a mean bias of -7.2 steps (-0.5%).
- Visually assessed validation yielded a mean bias of -0.4 steps (0.1%).
- Commercial wearable validation (Fitbit) demonstrated a 3.4% difference (-67.1 steps).
Conclusions:
- The open-source smartphone method provides reliable step counts.
- Accuracy is consistent across different sensor locations and measurement scenarios.
- The method is validated for both healthy adults and cancer patients.
