Temporal Alignment of Dual Monitor Accelerometry Recordings

Abstract

Combining accelerometry from multiple independent activity monitors worn by the same subject have gained widespread interest with the assessment of physical activity behavior. However, a difference in the real time clock accuracy of the activity monitor introduces a substantial temporal misalignment with long duration recordings which is commonly not considered. In this study, a novel method not requiring human interaction is described for the temporal alignment of triaxial acceleration measured with two independent activity monitors and evaluating the performance with the misalignment manually identified. The method was evaluated with free-living recordings using both combined wrist/hip (n = 9) and thigh/hip device (n = 30) wear locations, and descriptive data on initial offset and accumulated day 7 drift in a large-scale population-based study (n = 2513) were calculated. The results from the Bland-Altman analysis show good agreement between the proposed algorithm and the reference suggesting that the described method is valid for reducing the temporal misalignment and thus reduce the measurement error with aggregated data. Applying the algorithm to the n = 2513 samples worn for 7-days suggest a wide and substantial issue with drift over time when each subject wears two independent activity monitors.

Keywords: clock drift; measurement bias; method; sensor fusion.

Figure 1

Overview of the three stages and individual processing steps used in the temporal alignment method.

Figure 2

Subject three correlogram for hip and wrist pre-processed acceleration for days 1, 3 and 5. The data block for day 1 and 3 is from 02.00–03.00 PM, whereas for day 2 it is from 03.00–04.00 PM. The arrows identify the peak correlation coefficient which gives the lag and thus time shift between the two 1-h data blocks. The hip-worn monitor is the reference and thus the lag is the number of samples that is required to offset the acceleration of the wrist to align it with the hip acceleration.

Figure 3

Individual data points, confidence bounds and regression line for the association between the estimated lag for the one-hour data blocks and time. This is data for subject three in the WH group. Initial offset was estimated to −12.8 samples and the drift to 30.2 samples per day. Time is the hip-worn reference monitor time and lag is the number of samples required to offset the wrist acceleration to align with the hip acceleration.

Figure 4

Bland-Altman plots of the initial offset (A) and the accumulated drift at day 7 (B) for the WH group.

Figure 5

Bland-Altman plots of the initial offset (A) and the accumulated drift at day 7 (B) for the TH group. Subject nine was excluded from the plots.

Figure 6

The number of subjects with an accumulated day 7 drift in seconds divided into six categories (0–5, >5–10, >10–15, >15–20, >20–30 and >30 s).