Available Projects
On-line tracking of C. elegans: the development of an automated real-time tracking platform
C. elegans, a sub-millimetre nematode (worm), is a so-called model organism, that is used by biologists for a variety of use-cases - chief amongst them is to observe their behaviour changes under genetic modification and/or the effects of drugs. Changes in behaviour are used to assess the expected effects of drug(s) and/ or genetic modification(s) - and are usually manually observed and calculated. Automating this process is hugely beneficial to the community.
Primary supervisor: Professor Christopher James - Email: c.james@warwick.ac.uk
Project detail:
TrakBox is a commercially available platform that uses servo motors to move a low-cost video microscope across a 2D plane. This allows for recording close-up images of C. elegans in order to observe sub-behaviours related to body shape and movement, whilst also being able to track C. elagans across the wider 2D space. This automated setup allows for the worm to be tracked for much longer periods of time than is possible with manual observation - resulting in much richer datasets of behaviour for C. elegans under various mutations / effects of drugs.
This platform depends on the accurate analysis of each frame of the sequence of images in realtime, where the worm needs to be isolated in the presence of other (non-worm) material in the field of view, and this needs be done in the order of 5 to 10 frames/ second. This means that methods that are quick to operate are paramount here - such as the use of feedforward neural networks.
This project concerns the use of artificial neural networks trained with various machine learning techniques to automate the efficient and accurate detection of worms - in fine detail - from a sequence of video images and over long periods of time. The image recognition system needs. to be robust to noisy images, debris, different lighting conditions and differing worm sizes - based on both age of the worm and/ or current levels of magnification.
Once an effective ML based method is achieved, the worm "centre of mass" location can be extracted and used to effectively move the microscope the keeping the worm always within the field of view wherever it is within the 2D space.
The University of Warwick provides an inclusive working and learning environment, recognising and respecting every individual’s differences. We welcome applications from individuals who identify with any of the protected characteristics defined by the Equality Act 2010.