Supervisor: Prof. dr. Lucas Noldus, Section Neurophysics, Donders Institute for Brain, Cognition and Behavior, Radboud University
Co-supervisor: Prof. dr. Judith Homberg, Department of Cognitive Neuroscience, Donders Institute for Brain, Cognition and Behavior, Radboud University Medical Center
For the foreseeable future, the discovery and development of new therapies for neurological disorders will require research on animals, in particular mice and rats. Full understanding of an animal’s response to internal or external stimuli requires measurement of multiple biomarkers: visible (behavior), acoustic (audible sounds, ultrasonic vocalizations) and physiological parameters (cardiovascular, respiratory, brain activity and other signals). In line with the principles of Refinement, Reduction and Replacement (3Rs) of animal testing, we strive to collect this data in a continuous fashion, in an environment that mimics the natural habitat of rodents (such as social housing in an enriched cage), with minimal discomfort for the animals. This project will focus on two modalities: facial expression and heart rate, with the aim to measure both using computer vision.
Automated facial expression analysis using computer vision is a mature technology in studies with human subjects. Commercially available software tools can detect different basic expressions, valence, arousal and facial action units. For mice and rats, ‘grimace scales’ have been developed, based on different facial features, to detect levels of distress experienced by the animal. The next step is to automate this scoring in a continuous manner using computer vision.
Heart rate measurement in freely moving rodents is traditionally performed using a radio transmitter implanted in the abdomen with electrical leads running to the heart, which requires surgery under full anesthesia, a highly invasive procedure. A contact-free method, if technically feasible, would be highly preferred. In humans, this can be achieved using remote photoplethysmography (RPPG), i.e. detection of color changes of the skin – resulting from pulsation of the blood through the hair vessels – using a video camera. Recent studies have demonstrated the feasibility of RPPG in several animal species, including laboratory rats.
If facial expression as well as heart rate can be derived from a frontal camera, this device must be positioned in such a way that the chance of successful image capture is maximized. Eventually, the new tool must be integrated into a multimodal test environment, in which rodent behavior and physiology can be studied in a non-invasive manner.
This internship will encompass the following phases:
You will perform the tests on rats that are subjects in ongoing animal studies, under daily supervision of a PhD student or postdoctoral researcher. Hardware design and engineering will be supported by mechanical and electronic engineers of DCN’s technical services group.
As a research intern in the Donders Center for Neuroscience (DCN), you join a vibrant community of more than 200 researchers, technical and support staff. DCN is part of the world-renowned Donders Institute for Brain, Cognition and Behavior. With more than 20 nationalities represented in our center, our common language is English. You will become a member of the local research community “Rodent behavior and bioacoustics” consisting of Prof. Lucas Noldus, Prof. Judith Homberg, Dr. Bernhard Englitz, postdoctoral researchers, PhD candidates and master students, some 15 in total. The project will be carried out in collaboration with Noldus Information Technology in Wageningen, a leading developer of research tools for behavioral neuroscience. During the internship you will be able to visit the company and get a sense of a work environment outside academia. If your project is successful, the results may be published in a journal paper, of which you will become a co-author.
Prof. dr. Lucas P.J.J. Noldus, Section Neurophysics, Donders Center for Neuroscience, Radboud University
Address: Huygens Building, Room 00.136, Heyendaalseweg 135, 6525 AJ Nijmegen
Phone: +31-6-53425199
Email: lucas.noldus@donders.ru.nl
RU-homepage: https://www.ru.nl/en/people/noldus-l
LinkedIn: http://www.linkedin.com/in/lucasnoldus