2.3 - Mobility Monitoring for Geriatrics: Gait Detection using e-Textile
- Event
- iCCC2026 - iCampus Cottbus Conference
2026-05-05 - 2026-05-07
Cottbus - Band
- Vorträge
- Chapter
- Gesundheit und Sport
- Author(s)
- T. Steinmetzer, S. Michel - BTU Cottbus-Senftenberg, Cottbus
- Pages
- 55 - 58
- DOI
- 10.5162/iCCC2026/2.3
- ISBN
- 978-3-910600-10-2
- Price
- free
Abstract
This study introduces and validates a practical, twostage pipeline for the unobtrusive detection and analysis of gait using a single Inertial Measurement Unit (IMU) integrated into a smart undershirt, positioned at the abdomen, aimed at enabling reliable long-term geriatric monitoring. Data were collected from 15 healthy young adults performing free movement and quiet standing. The pipeline first employs a 1D Convolutional Recurrent Neural Network (1D-CRNN) classifier using six IMU signals and derived features to segment gait versus standing activity. Subsequently, a routine utilizing bandpass filtering and peak detection on the Euler angle magnitude is used to count steps and derive cadence and step variability. The 1D-CRNN achieved excellent classification performance with an Accuracy of 0.981 ± 0.005. The step detection successfully isolated continuous walking segments, yielding plausible cadence estimates (e.g., 94.6 steps/min) and measures of step variability (0.236 s). The smart-textile-based system provides a robust and lightweight solution for reliable gait detection and the extraction of fundamental mobility metrics from an unobtrusive trunk location. This foundation is crucial for developing practical wearable technologies for assessing functional health in older adults, with future work focusing on clinical validation in geriatric cohorts and expanding the activity classification spectrum.