Coventry University Featured PhD Programmes
University of Warwick Featured PhD Programmes
University of Edinburgh Featured PhD Programmes

Can deep learning algorithms automate the analysis of reach-and-grasp or walking by lab rodents when testing therapies for stroke?

Health Schools

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
Dr Lawrence Moon , Dr E Robinson No more applications being accepted Competition Funded PhD Project (Students Worldwide)

About the Project

Thousands of research labs worldwide use mice and rats to assess new therapies for CNS injury. However, behavioural assessment of rodents is very time consuming and expensive.

The PhD student will use machine learning (DeepLabCut) to automate the kinematic analysis of reach-and-grasp by mice after treatment using a novel therapy for brain injury (stroke) based on movies captured by a new device:

The “horizontal ladder task” is often used but it takes many hours to analyse accuracy of foot placement from videos played in slow motion. The PhD student will use machine learning (DeepLabCut) to automate the identification of paws and limb joints during walking on the ladder for marker-less kinematic analysis.

DeepLabCut was published in Nature Neuroscience in 2018 (Mathis et al 2018). We have begun using this method for tracking digits of mice. See for beautiful examples of this algorithm developed by others.

The successful student will have existing skills in Python and machine learning. See the DeepLabCut GitHub for technical information.

Funding Notes

This opportunity is only available for Chinese graduates starting in the academic year 2019/2020. These scholarships are open to graduate students who are citizens and permanent residents of the People’s Republic of China at the time of application.

Successful applicants will receive: Full International fees, airfare, visa application fees and annual stipend

See King's China Council Scholarships (K-CSC) at (scroll down to see it!)


FindAPhD. Copyright 2005-2021
All rights reserved.