Post Doctoral Research Associate in deep learning for neuroimage analysis

About us:

We are seeking experts in medical image deep learning to join our team and help develop novel computationally efficient segmentation algorithms.

We welcome application from individual with experience in:

• Deep learning
• Medical imaging computing (preferably neuroimaging)
• Computationally efficient deep learning
• Deep learning model generalisation techniques.
• Translating deep learning models to the clinic

The post holder will be based in the Department of Biomedical Computing as part of the School of Biomedical Engineering & Imaging Sciences, King’s College London, a vibrant community of engineers designing & translating technology into the clinical. This role will be supported by the Quality Management Systems team at London Institute for Health Care Engineering who will collaborate closely with the group to ensure internal & regulatory requirements are met for the developed quality management system.

About the role:

We are looking for someone with a passion for translating deep learning models into the clinical as part of a team.  This role is focused on developing computationally efficient deep learning models to perform neuroimaging segmentation tasks to help inform treatment decisions within a busy neurosurgical centre.

In this role the successful candidate will develop deep learning algorithms & validate models on diverse clinical data which include magnetic resonance imaging (MRI) and computed tomography (CT). In this role the successful candidate will write and seek to publish the developed state-of-the-art algorithms in high quality medical image computing journals & conferences. The successful candidate will work with a three-person team of software engineers to integrated algorithms into the Epilepsy Navigation (EpiNav™) software platform for stereotactic neurosurgery planning. The successful candidate will help to support the team to formally document algorithms within a quality management system.

EpiNav™ provides state-of-the-art computer-assisted support for the planning of stereotactic neurosurgery procedures. It has been in daily clinical use at the National Hospital for Neurology & Neurosurgery for over 7 years aiding neurologist & neurosurgeons in the surgical treatment of over 150 patients with medically refractory epilepsy and brain tumour biopsy. This software enables clinicians to use complex multi-modal imaging to inform surgical decisions. Based on the early success of EpiNav™ to help improve patient care we are refining and streamlining the software so that it is suitable for use outside of a research setting.  The team is multidisciplinary consisting of experts in medical image computing, medical regulation, neuroimaging and neurosurgery.

This is a full-time post (35 Hours per week), and you will be offered a fixed-term contract until 31/7/2028 

About you:
(the candidate)

To be successful in this role, we are looking for candidates to have the following skills & experience:

Essential criteria

1. PhD qualified in relevant subject area*
2. Experience developing deep learning segmentation models
3. Experience with Pytorch, MONAI, CUDA or equivalent software libraries for developing deep learning models.
4. Familiarity with medical image such as MRI, CT, or volumetric ultrasound.
5. Knowledge on common medical imaging data augmentation techniques for robust model design.
6. Strong written & oral communication skills including evidence of publishing peer reviewed research articles.

Desirable criteria

1. Experience working with neuroimaging
2. Experience deploying deep learning models into clinical settings
3. Experience working within a development team & associated tools (git, svn or equivalent version control)
4. Experience working with multidisciplinary teams including people from non-technical backgrounds
5. Experience with developing or deploying light weight models in environments with restricted resources (e.g. outside of HPC environments)

This post is subject to Disclosure & Barring Service and/or Occupational Health clearances.

Advert information