Skip to main content

Table 2 Summary of parameters derived from the TUM:W smart-watch

From: Going beyond PA: Assessing sensorimotor capacity with wearables in multiple sclerosis—a cross-sectional study

Abbreviation

Sensor (dimension)

Assessment

Operationalization

Outcome

(thought) Target

CADENCE95

Pedometer

(PA)

60s every 60s

Steps/min

95th perc

Continuous walking capacity (to cover distance)

CADENCE

Pedometer

(PA)

60s every 60s

Steps/min if > 5 recognized steps

Mean

Walking fragmentation

MEANG

Accelerometer

(PA)

5s every 5s

Mean acceleration

90th perc

Energy expenditure

MAD

Accelerometer

(PA)

5s every 5s

Average deviation from mean of signal

90th perc

Energy expenditure

STEPFREQ

Pedometer

(PA)

5s every 5s

Steps/s

90th perc

Short distance walking (e.g., indoors)

ACTI

Gyroscope

(beyond PA)

5s every 5s

Mean angular vel

90th perc

Intensity of upper-limb use (e.g., during ADL)

AUTOCOR

Gyroscope

(beyond PA)

5s every 2.5s

Highest coefficient of autocorrelation

90th perc

Continuity of cyclic movements

AUTOCORFREQ

Gyroscope

(beyond PA)

5s every 2.5s

Frequency of highest autocorrelation (1.0–3.0Hz)

90th perc

Speed of cyclic movements

MAX5

Gyroscope

(beyond PA)

60s every 30s

5 highest vel. peaks

90th perc

Intensity of upper-limb use

STD

Gyroscope

(beyond PA)

60s every 30s

Standard deviation of signal

90th perc

Intensity of upper-limb use (by the variance of its movement)

STDMEAN

Gyroscope

(beyond PA)

60s every 30s

Standard deviation of signal divided by the mean of the signal

90th perc

Intensity of upper-limb use in relation to maximum capacity (within the 60s scenario)

PEAKRATIO

Gyroscope

(beyond PA)

60s every 30s

Ratio of delimited (min. distance to last minimum) and all vel. peaks

90th perc

Movement smoothness (signal-to-noise)

PEAKSPER360

Gyroscope

(beyond PA)

60s every 30s

Number of vel. peaks

per 360° summed rotation

90th perc

Movement smoothness

MEANPEAK

Gyroscope

(beyond PA)

60s every 30s

Average height of vel. peaks

90th perc

Intensity of upper-limb use

PEAKSTD

Gyroscope

(beyond PA)

60s every 30s

Standard deviation of the height of vel. peaks

90th perc

Intensity of upper-limb use

PEAKSTDMEAN

Gyroscope

(beyond PA)

60s every 30s

Standard deviation of the height of vel. peaks divided by the mean of vel. peaks

90th perc

Intensity of upper-limb use in relation to maximum capacity (within the 60s scenario)

FATREV

Accelerometer

(PA)

n.a

Ratio of ≥ 30s and ≥ 5s activity bouts (> 100 milli-g)

Complete dataset

Activity fragmentation (e.g., by fatigability)

RELPA

Accelerometer

(PA)

5s every 5s

Relative time with > 100 milli-g

Complete dataset

Physical activity

LOW

Gyroscope

(beyond PA)

5s every 2.5s

If > 5 steps/min, ratio of coeff. of autocor. < 0.30 and all > 5 steps/min data points

Complete dataset

Amount of gait with low autocorrelation (e.g., by changes of movement direction)

  1. The dimension was based on our a-priori assumption, with PA standing for physical activity / energy expenditure and beyond PA for sensorimotor capacity