Brain development in infants and its disruption by preterm birth or perinatal injury, can be measured with functional MRI (fMRI). Unfortunately, infants move in the scanner and half the images are discarded, precluding clinical application. In recent years, deep neural networks (DNNs) have led to breakthroughs in artificial intelligence and are finding growing application in biomedical imaging. DNNs have considerable potential to correct head motion in fMRI, as they can learn complex mappings, and exploit knowledge of brain structure. The FreezeMotion project will develop DNNs to motion correct fMRI data.
Candidates must have expertise in at least one of the two following areas and must be willing to develop skills in the other:
- design and optimisation of deep neural networks
- neuroimaging with fMRI
Candidates must have a strong level of expertise in programming in python or another language.
FreezeMotion is a collaboration between Rhodri Cusack (Trinity College Institute of Neuroscience, Trinity College Dublin), Chen Qin (Department of Electrical and Electronic Engineering and Imperial-X, Imperial College London), Mark Chiew (Medical Biophysics, University of Toronto) and Alex Bronstein (Computer Science, Technion).
The post will be based in Dublin, Ireland. Appointment will be on SFI Experienced Post Doctoral Research (2B) or Research Fellow (Level 3) scales, full time, €51,677-€68,318 per annum. Benefits include a pension contribution and PRSI social insurance.
Theses posts will be for two years but there may arise the opportunity to extend them.
See further details and application procedure
Deadline 12 noon on Sept 16, 2024