sensor_EMA_PD_monitoring

Code to test technical validity of public data set containing objective wearable sensor (accelerometer and gyroscope) data and subjective EMA (ecological momentary assessment) data. READ ME

Data repository: Habets, Jeroen; Kubben, Pieter, 2020, "EMA and wearable sensor monitoring in PD", https://doi.org/10.34894/5HHK8H, DataverseNL. Published Data Descriptor in DATA, Feb 23, 2021: https://www.mdpi.com/2306-5729/6/2/22

Please cite data descriptor when using data.

WEARABLE SENSOR DATA

Folders per patient

Sensor files are organised in folders per patient. Folders have patient ID’s as name ’110001’, ’110002’, etc until ’110021’ (’110012’ is correctly missing).

Filenames for sensor location and timestamp

First part filename: Sensorlocation_

File names start with sensor code. left sensors = ['13797', '13799', '13794', '13806'] right sensors = ['13805', '13801', '13793', '13795'] chest sensors = ['13804', '13792', '13803', '13796']

Second part filename: date

Date in format yyyymmdd

Last part filename: time

Time in format hhmmss.

Explanation file structure

Every filename is the starting timestamp of the file. Every file is recorded with 200 Hz sample frequency.

Sensors recorded when not charging. When recording period passed 00.00, a new file was created.

For example code how to extract and analyse the elf-files, see GitHub / jgvhabets / sensor_EMA_PD_monitoring.

Explanation about EMA data xlsx file attached to Dataverse repository.

For further questions: j.habets@maastrichtuniversity.nl, or jgvhabets@gmail.com