Research
SmartCare Unit – Mimikmonitoring
Motivation
Pain is not only an unpleasant experience, it also puts the human body under stress. Post-operative pain frequently occurs after surgery and affects both the well-being of patients and the duration and quality of their recovery. For this reason one of the tasks of clinical staff is to keep patients as free of pain as possible. While awake and responsive patients are able to draw staff's attention to their pain, sleeping or sedated people cannot do so. Nevertheless, sedated patients can also suffer pain. Unresponsive patients in intensive care units and recovery rooms are therefore monitored by staff and given analgesics when needed. At present, however, there is no method or measuring device that can reliably detect pain by measuring physical quantities. For this reason, staff on their rounds pay particular attention to patients' facial expressions, which are an indication of perceived pain even in sedated people.
Goals
The Facial Expression Monitoring project aims to reproduce this assessment of pain levels from patients' facial expressions, as practised by staff, with a technical setup. Patients' faces are to be recorded with an infrared thermography (IRT) camera and the pain level is to be determined continuously from these images. This requires extensive research in digital image processing at the LfB. The first step is the development of a reliable method for detecting and tracking faces in IRT videos. The particular challenge here is that thermographic images show partly very different facial features from videos recorded with conventional cameras. In addition, the contrast in IRT data is low, which further reduces the accuracy that tracking can achieve. There is also a lack of methods for the reliable and accurate detection of important facial regions such as the eyes or the nose, which can provide a great deal of potential information about a person's pain level. Nevertheless, the method to be developed must achieve high accuracy and robustness in detection and tracking in order to provide location information about different facial regions as precisely as possible. In a further step, a classification method must be investigated for estimating perceived pain from the images. Machine learning algorithms are trained with reference data and are then to assign a pain index to the faces autonomously and in real time. The difficulty here is the large variance in the IRT images, which makes it harder to develop a robust classifier. In a final step all the sub-components are combined into a system that allows medical staff to use it in a real intensive care setting.
Facial Expression Monitoring is a project of the SmartCareUnit-Netzwerks. The core idea of this consortium of industrial, academic and medical partners is to develop innovative and sustainable concepts in intensive care medicine. It is a publicly funded project which, by involving research institutions as well as small and medium-sized enterprises, aims to secure the competitiveness of Germany as a location for research and industry.
Thesis Topics
New theses in the area of facial expression monitoring are advertised regularly. In addition to the Overview , there are also numerous topics that have not yet been advertised, which will be gladly presented in a personal conversation.
Partners
- Prof. Dr. Steffen Leonhardt, Medical Information Technology at the Helmholtz Institute for Biomedical Engineering (MedIT), RWTH Aachen University
- Prof. Dr. Vladimir Blazek, Medical Information Technology at the Helmholtz Institute for Biomedical Engineering (MedIT), RWTH Aachen University
- Prof. Dr. Rolf Rossaint, Department of Anesthesiology, University Hospital Aachen
- Prof. Dr. Michael Czaplik, Department of Anesthesiology, University Hospital Aachen
- Prof. Dr. Bettina Wiese, Institute of Psychology, Work and Organisational Psychology, RWTH Aachen
- Dr. Christian Burk, Institute of Psychology, Work and Organisational Psychology, RWTH Aachen
- Netzwerkpartner: www.smart-care-unit.de/partner.html
External Funding
- BMWi/ZIM, „SmartCareUnit– facial expression analysis using geometric and dynamic IRT features and development of tracking algorithms”, project no. 16KN035923
- OPSF645 – Cooling Upper Lip = Agony? (CUL=A) – Detecting Differential Responses to Psychosocial Stress through Thermal Infrared Tomography, RWTH Aachen
Contact
Univ.-Prof. Prof. h.c. (Amity University, India) Dr.-Ing.
