Published December 31, 2020 | Version v1.0.0

Robust step detection from different waist-worn sensor positions – implications for clinical studies

  • 1. Novartis Phrama
  • 2. Novartis Pharma

Description

The dataset contains tri-axial acceleration and gyroscope data (100 Hz sampling) from walks from 19 healthy volunteers, each walking up to three times a parcours of 20 meters with self-selected speed, slow speed or with five soft turns at self-selected speed. Each participant wore 11 time-synchronized sensors during these tests: left/right foot, 5 around waist, non-dominant wrist and upper arm and collar and pocket. In addition to the sensor recordings each 20 meter walk was timed with a stop-watch.  

Also see: https://doi.org/10.1159/000511611

Notes

Parts of this dataset was analysed and published in Digital Biomarkers, special issue "The Future of Digital Health" (Tietsch et al., 2020, https://doi.org/10.1159/000511611).

Files

README.txt

Files (549.3 MB)

Name Size
md5:b5ab61604d5e9c47f27821cd91bcfcfd
2.7 kB Preview Download
md5:036b73c56acc68ca3358b3225c0daf8b
549.3 MB Preview Download