AMACS  Automatic monitoring of Actitiviies of Daily Living using Contactless Sensors
Research project in collaboration between the following knowledge centers: K.H.Kempen - K.U.Leuven - CETIC - Lessius - Fontys


Overview AMACS project
Overview of the AMACS system.

The general aim of this project is to develop and evaluate a system that can automatically detect the activities of daily living of older persons living alone at home. This automatic detection is based on measurements of contactless sensors that are installed in the home environment of the older persons. In this project we use sensors that can measure the consumption of public utilities (water, electricity and gas), sensors used in security applications (such as IR-switches), and video cameras. The result of the project is a prototype that will be evaluated in a real-life situation.


The ongoing aging in our modern society leads to the tendency to allow older persons to stay as long as possible in their home environment. Although they are mainly able to independently organize themselves, it is for a certain group nevertheless necessary to observe their activities of daily living (ADL). Examples of such activities include sleeping, cooking, making a phone call, visiting the toilet, washing etc. Based on this automatic detection of ADL, we want to be able to detect changes in the behavior, i.e. changes in these ADL patterns. These changes include both acute and gradual changes. Acute changes are abnormal events that are critical and require an immediate alarm. Examples of such abnormal events include for instance fall incidents; water or gas that keeps running or sudden general absence of activity. On the other hand, we also want to detect gradual changes. These changes are important for an early detection of problems such as (early stage) dementia. Examples of such changes are sleeping disorders, ADL decline and behavioral disturbances. The information about these activities and changes in behavior can then be presented to the caregivers (including family members) to adapt older people's care plans, and as a consequence, increase their quality of care and quality of life. Hence allowing them to stay longer at their homes.

This project is partly triggered by an opportunity generated by the ongoing IWT-TETRA FallCam project (funded by the Flemish Government) in which we want to detect fall incidents from video data. Although the elderly population is prawn to higher risks of falling, it nevertheless takes multiple weeks (sometimes months) of video recordings before an actual fall incident occurs. Our clinical partners indicated that there is much more useful information available in these video recordings. By applying state-of-the-art image processing techniques we want to capture these relevant activities from the video recordings. Some recordings are readily available from the FallCam Tetra project.

Another trigger for setting up this project is the availability of sensors to measure public utilities. Colleagues that are active in Energy Management have pointed this out. The advantage is that these sensors can typically be installed in the basement or storage area of the house and are hence non-intrusive. The original aspect of this project is that we combine multiple sensor modalities to increase the performance of detecting ADLs.


The detected ADL should also be accurately notified to the relevant stakeholders - the family and relatives, the visiting nurses, the general practitioner, etc. The project will study the corresponding interfaces according to the efficiency and their impact on the medical process and by respecting their acceptance by the different parties and the privacy of personal medical data.

Other target groups

Although the goal of this project is to develop a system that helps community dwelling elderly, the results are generic of nature. At the final stage of the project, we will, as an example, study the usability of the system for people with Chronic Fatigue Syndrome (CFS). But other target groups can be considered as well. The system might be useful in the monitoring and/or treatment of other groups such as patients suffering from Chronic Obstructive Pulmonary Disease (COPD), depression or non-congenital brain disorders (post trauma or post CVA patients). Diminished activity in the home environment might indicate a decrease in the health status of those people.