Table 2.
Descriptions for each type of wearable sensor feature
| Feature level | Feature(s) name | N | Description |
|---|---|---|---|
| Activity index | AI mean | 1 | Activity index32 was computed for each 1-s window of triaxial accelerometer data over the recording period. ‘Activity intensity (AI) mean’ is the mean activity index value over all daytime activity over the week-long period. ‘Periods of inactivity are excluded from the calculation of AI mean, AI median, AI mode and AI entropy’.30 |
| AI median | 1 | Median activity intensity over all daytime activity. | |
| AI mode | 1 | The most common value (mode) of activity intensity over all daytime activity. | |
| AI entropy | 1 | The entropy of the distribution of daytime activity intensity. | |
| % daytime with low AI | 1 | The percentage of daytime that is spent performing low-intensity movements as previously defined.30 | |
| % daytime with moderate AI | 1 | The percentage of daytime that is spent performing moderate-intensity movements. | |
| % daytime with high AI | 1 | The percentage of daytime that is spent performing high-intensity movements. | |
| % accel in single direction | 3 | For each 1-s window of movement, principal component analysis was performed on the triaxial accelerometer data to identify the principal direction of acceleration. This feature is the percentage of accelerometer data variance explained by the first principal component direction, averaged over 1-s windows. This measure was computed separately for low AI, moderate AI and high AI 1-s windows resulting in three features. | |
| Spectral | Total power | 1 | Cumulative power in the 0.1- to 5-Hz frequency band. |
| Activity bout | Bout acceleration | 2 | ‘Activity bouts’ are continuous periods of activity with durations between 4 and 18 s long based on an activity index threshold.19 Bout acceleration is the maximum acceleration in m/s2 during an activity bout. ‘M and SD are computed over a participant’s activity bouts resulting in two features (applies to bout acceleration and bout jerk)’. |
| Bout jerk | 2 | Bout jerk is the mean jerk (derivative of acceleration) in m/s3 during an activity bout. | |
| SM | SM distance | 8 | The distance in meters traveled during a submovement (SM). ‘Mean and standard deviation are computed over a participant’s SMs for short-duration and long-duration SMs in the primary and secondary directions of planar movement resulting in 2 ∗ 2 ∗ 2 = 8 total features (applies to SM distance, velocity, acceleration, jerk and duration)’. |
| SM velocity | 8 | The maximum velocity in m/s during a SM. | |
| SM acceleration | 8 | The maximum acceleration in m/s2 during a SM. | |
| SM jerk | 8 | The normalized jerk of a SM. This measure is dimensionless and is scaled based on SM duration and SM peak velocity.20,33,34 | |
| SM duration | 8 | The duration of a SM in seconds. | |
| SM PC1 score | 6 | The principal component 1 (PC1) score for a submovement. PC1 captures low-frequency characteristics of the SM velocity–time curve (e.g. the SM ‘shape’). The PC1 ‘basis function’ is a single sinusoidal waveform with the peak present in the first half of the submovement.19 ‘Mean absolute value, standard deviation and kurtosis are computed for long-duration SMs in the primary and secondary directions of movement resulting in 3 ∗ 2 = 6 total features (applies to SM PC1–5 scores)’. | |
| SM PC2 score | 6 | The principal component 2 score for a submovement. Similar to PC1, PC2 captures low-frequency characteristics of the SM velocity–time curve. The PC2 basis function is a single sinusoidal waveform with the peak present in the second half of the submovement.19 | |
| SM PC3–5 scores | 18 | The principal component 3–5 scores for a submovement. PC3–5 scores capture higher frequency characteristics of the SM velocity–time curve. The PC3, PC4 and PC5 basis functions consist of 1.5, 2 and 2.5 sinusoidal cycles, respectively.19 |
Bolded features were preselected for individual feature analysis. AI, activity intensity; N, number of features; M, mean; PC, principal component; SD, standard deviation; s, seconds; SM, submovement.