Publications

One of the effects of late-stage dementia is the loss of the ability to communicate verbally. Patients become unable to call for help if they feel uncomfortable. The first objective of this article was to record facial expressions of bedridden demented elderly. For this purpose, we developed a video acquisition system (ViAS) that records synchronized video coming from two cameras. Each camera delivers uncompressed color images of 1,024 x 768 pixels, up to 30 frames per second. It is the first time that such a system has been placed in a patient's room. The second objective was to simultaneously label these video recordings with respect to discomfort expressions of the patients. Therefore, we developed a Digital Discomfort Labeling Tool (DDLT). This tool provides an easy-to-use software representation on a tablet PC of validated "paper" discomfort scales. With ViAS and DDLT, 80 different datasets were obtained of about 15 minutes of recordings. Approximately 80% of the recorded datasets delivered the labeled video recordings. The remainder were not usable due to under-or overexposed images and due to the patients being out of view as the system was not properly replaced after care. In one of 6 observed patients, nurses recognized a higher discomfort level that would not have been observed without the DDLT.

Pen-paper observational scales have proven to be useful to monitor pain, but have also a lot of disadvantages. The aim of this study is to present and examine the advantages of an electronic observational pain scale.
The pilot study (approved by a medical ethical committee) took place in a specific geriatric centre. Patients are included when they’re bedridden and have limited ability to communicate verbally. The caregiver has to objectify pain by means of a behavioural observation scale. Their usability is often limited, they are time-consuming because of their length, difficulties in calculating scores, and post processing required to evaluate the pain evolution. Also the exact timing of the indicators can’t be recorded in real-time, and thus complex timing patterns can’t be evaluated. Each patient was tracked for 6 random days, in which 6 assessment sessions were performed at clinically interesting moments, after an informed consent was signed by a relative. As input for the electronic scale, 3 valid and reliable scales are chosen: the Pain Assessment Checklist for Seniors with Limited Ability to Communicate, the Discomfort Scale - Dementia of Alzheimer Type, and the Faces Pain Scale Revised. The electronic device (a tablet PC with a touch screen) allows the caregiver to record indicator events, such as frequency and duration, as they occur in real time. The scores are calculated automatically. The digital information is stored in a database, improving administration and allowing database applications. With the advantages of the electronic device, nurses will be stimulated to perform more frequent pain measurements. A more adequate treatment can be provided with the knowledge of a more accurate pain level. Advantages of this electronic observational pain scale are an increased usability, less time-consuming and inclusion of complex timing patterns. This will lead to a more accurate measurement of pain, a more adequate treatment and thus improving quality of life.

Epidemiological research shows that 5% up to 10% of the population older than 65 years is suffering from dementia. With the increased live-expectation, the chance of obtaining dementia, is increasing too. One important and underestimated aspect is discomfort (pain) in our target group. Recognition of pain is therefore an important issue for the caregivers. The purpose of our project is to present a way of monitoring the non-verbal expressions of patients with severe dementia by means of a vision system and to identify discomfort in real-time by means of pattern recognition techniques. This system can constantly monitor the patient and alarm a caregiver when the patient reaches a certain critical level of pain.

This paper describes the setup of the acquisition system to obtain high quality stereo images of the patient’s face together with pain indicator information provided by a pain expert. This acquisition system must meet the requirements enforced from technical point of view (frame rate, resolution, etc.) as well as from medical point of view (simple to use, do not disturb the natural look of a patient room, etc).

Epidemiological research shows that 5% up to 10% of the population older than 65 years is suffering from dementia. With the increased liveexpectation, the chance of obtaining dementia, is increasing too. One important and underestimated aspect is discomfort (pain) in our target group. Recognition of pain is therefore an important issue for the caregivers. The purpose of our project is to present a way of monitoring the non-verbal expressions of patients with severe dementia by means of a vision system and to identify discomfort in real-time by means of pattern recognition techniques. This system
can constantly monitor the patient and alarm a caregiver when the patient reaches a certain critical level of pain.
This paper describes the setup of the acquisition system to obtain high quality stereo images of the patient’s face together with pain indicator information provided by a pain expert. This acquisition system must meet the requirements enforced from technical point of view (frame rate, resolution, etc.) as well as from medical point of view (simple to use, do not disturb the natural look of a patient room, etc).

Epidemiological research shows that 5% up to 10% of the population older than 65 years is suffering from dementia. With the increased live-expectation, the chance of obtaining dementia, is increasing too. One important and underestimated aspect is discomfort (pain) in our target group. Recognition of pain is therefore an important issue for the caregivers. The purpose of our project is to present a way of monitoring the non-verbal expressions of patients with severe dementia by means of a vision system and to identify discomfort in real-time.

This abstract describes the setup of the acquisition system to obtain high quality stereo images of the patient’s face together with pain indicator information provided by a pain expert.

Epidemiological research shows that 5% up to 10% of the population older than 65 years is suffering from dementia. With the increased live-expectation, the chance of obtaining dementia, is increasing too. One important and underestimated aspect is discomfort (pain) in our target group. Recognition of pain is therefore an important issue for the caregivers. The purpose of our project is to present a way of monitoring the nonverbal expressions of patients with severe dementia by means of a vision system and to identify discomfort in real-time by means of pattern recognition techniques. This system can constantly monitor the patient and alarm a caregiver when the patient reaches a certain critical level of pain. This paper describes the setup of the acquisition system to obtain high quality stereo images of the patient’s face together with pain indicator information provided by a pain expert. This acquisition system must meet the requirements enforced from technical point of view (frame rate, resolution, etc.) as well as from medical point of view (simple to use, do not disturb the natural look of a patient room, etc).