Title: Acquiring a dataset of labeled video images showing discomfort in demented elderly
Authors: Bonroy, Bert ×
Schiepers, Pieter
Leysens, Greet
Miljkovic, Dragana
Wils, Maartje
De Maesschalck, Lieven
Quanten, Stijn
Triau, Eric
Exadaktylos, Vasileios
Berckmans, Daniel
Vanrumste, Bart #
Issue Date: May-2009
Publisher: Mary Ann Liebert
Series Title: Telemedicine and E-health vol:15 issue:4 pages:370-378
Abstract: One of the effects of dementia in a late stage, is the loss of the ability to communicate verbally.
Patients become unable to call for help if they feel uncomfortable. The aim of this paper is to develop
a video acquisition system and a digital discomfort label tool for this patient group. These tools
deliver labelled video recordings which can in the future be used for developing an automatic
discomfort detection algorithm.
The first objective is to build a video acquisition system which can be placed in a patient’s room. This
acquisition system records facial expressions of demented elderly lying in bed. The developed system
can record synchronised video from the input of two cameras. These cameras deliver uncompressed
colour images in a format of 1024 x 768 pixels, with a speed up to 30 frames per second.
The second objective is to develop a tool for easily and consistent labelling of the video recordings
with respect to discomfort. Therefore a digital discomfort label tool is developed. This tool is based
on the PACSLAC, DS‐DAT and FPS‐R, which are validated pain / discomfort scales.
For six patients suffering from severe dementia (level 7 on the GDS scale) about 80 different datasets
are obtained. These datasets delivers the required information to develop the discomfort recognition
algorithm. The use of the digital discomfort label tool, which immediately returns the discomfort
score of the completed assessment session, is considered as an added value, and has already led to
an earlier detection of discomfort.
ISSN: 1530-5627
Publication status: published
KU Leuven publication type: IT
Appears in Collections:Division M3-BIORES: Measure, Model & Manage Bioresponses (-)
ESAT - STADIUS, Stadius Centre for Dynamical Systems, Signal Processing and Data Analytics
Health Care Department - Lier Campus - TM K
Technologiecluster ESAT Elektrotechnische Engineering
Electrical Engineering (ESAT) TC, Technology Campus Geel
Electrical Engineering (ESAT) TC, Technology Campus Diepenbeek
× corresponding author
# (joint) last author

Files in This Item:

There are no files associated with this item.

Request a copy


All items in Lirias are protected by copyright, with all rights reserved.

© Web of science