Senior Research Fellow

UCL Queen Square Institute of Neurology

Job Description

Senior Research Fellow in Machine Learning for Ultrasound Brain Imaging

Research Department

Translational Neuroscience & Stroke

Reports to

Director, Professor Mike Hanna

Head of Research Department, Professor David Werring Professor of Neurology, Professor Parashkev Nachev Senior Research Fellow

Line manager: Professor Parashkev Nachev

Context

Grade: 8

Location: Russell Square House, 10-12 Russell Square, London, WC1B 5EH / On-site First Structure of the UCL Queen Square Institute of Neurology The UCL Queen Square Institute of Neurology (IoN) is one of the world’s leading centres for research into neurological disease and its treatment.

The High-Dimensional Neurology Group, led by Professor Parashkev Nachev within the Department of Translational Neuroscience & Stroke, applies large-scale machine learning to routinely acquired clinical data, particularly brain imaging, to build generative and predictive models of the brain in health and disease and translate them into tools that improve the care of patients with acute neurological disorders such as stroke.

The group works closely with clinical services at the National Hospital for Neurology and Neurosurgery and with University College London Hospitals NHS Foundation Trust. It also has access to some of the largest curated clinical neuroimaging datasets in the UK.

Main purpose of the job

You will work on the Engineering and Physical Sciences Research Council (EPSRC) funded project that aims to create a radically new brain-imaging modality: low-frequency ultrasound transmitted across the head and reconstructed using full-waveform inversion (FWI), a computationally intensive technique originally developed in geophysics. The core objective is to generate the first highresolution ultrasound images of the living adult human brain, with the longer-term aim of developing a safe, fast, portable and universally applicable device for use in stroke and other acute neurological conditions, including outside hospital settings.

You will focus on making FWI faster and more robust by learning from large-scale clinical imaging data. This will involve building high-fidelity generative models of the human skull and brain that capture the full range of normal and pathological appearances, as well as variation across sex, ethnicity and age. You will use these models to provide anatomically informed priors that enable numerical inversion to start close to the final solution, quantify the trade-oƯs between computational speed and accuracy, and determine which combinations of recovered physical parameters have the greatest diagnostic value. You will also evaluate the resulting images against the clinical questions that matter most in acute stroke, working closely with clinical and research colleagues to ensure that the developed methods have meaningful potential for clinical application.

The post is available from 1 November 2026 and is funded by the EPSRC until 31 October 2027 in the first instance.

Duties and responsibilities

 Contribute to the design, development and validation of high-fidelity generative models of the human head (skull and brain), trained on large-scale clinical MRI and CT data, that capture the full range of normal and pathological anatomy and of variation across sex, ethnicity and age.

 Develop methods for deriving anatomical priors from these models and using them to initialise, regularise and guide ultrasound full-waveform inversion, so that inversion is faster, more robust to non-convexity and converges to accurate images.

 Design and run experiments quantifying the trade-oƯs between reconstruction speed and image accuracy and determining which combinations of recovered physical parameters carry the greatest diagnostic value.

 Contribute to the computing infrastructure (GPU workstations, national HPC facilities such as ARCHER2, and cloud platforms) needed to train generative models and run large-scale inversions across the consortium.

 Act as the day-to-day scientific lead for the UCL component, planning work to project milestones, reporting progress to the Principal Investigator and Co-Investigator, and contributing to project management, EPSRC reporting and stakeholder meetings.

 Work closely with engineering, physics, ultrasound and clinical members of the consortium at Imperial and UCL, including regular travel to Imperial College London for joint experiments and meetings.

 Prepare high-quality papers for publication in leading journals and conferences, and present the work at national and international meetings.

 Contribute to the translational and commercialisation pathway for the technology, including documentation of methods and software.

 Contribute to the supervision and mentoring of junior researchers and students working on related projects in the group.

 Maintain and develop well-documented, reproducible research software to the standards required for sharing across the consortium.

Teaching, research, and development

 Provide teaching on topics related to your work as requested by your line manager or Head of Research Department.

 Contribute to your Department’s multidisciplinary research projects as aligned with the departmental strategy.

 Prepare and analyse data for publications for the dissemination of research and for presentation at internal meetings at UCL, meetings with external collaborators, and national and international conferences.

General

 Attend and contribute to Departmental, Institutional, and university meetings, committees, and initiatives as appropriate.

 Always act in accordance with the highest professional standards and ensure that these are maintained in the delivery of all aspects of research.

 Actively promote equality of opportunity and inclusivity, supporting UCL’s Equal Opportunity policy within the remit of your role.

 Ensure the highest standard of record keeping, maintaining accurate, complete, and up-todate records.

 Ensure confidentiality is maintained as applicable.

 Follow and actively promote the UCL Ways of Working.

 Takes responsibility to carry out duties in a way that embodies UCL's environmental and social sustainability values, actively supporting UCL's Sustainability Strategy, policies and objectives within the remit of your role.

The above description is not exhaustive, and you will be required to undertake any other duties as may reasonably be requested within the scope, spirit, and purpose of the post.

Job descriptions are reviewed on a regular basis including at the annual appraisal. As duties and responsibilities change, the job description may be amended in consultation with the post-holder.

You will be expected to actively follow all UCL policies and procedures and complete UCL mandatory training, maintain an awareness of Fire and Health & Safety Regulations, attend meetings as required, and undertake such training and development as may be needed for the post.

All staƯ are required to act professionally, cooperatively, and flexibly in line with the requirements of the post and the Institute’s Code of Conduct.

Person Specification

Note to job applicant: copy and paste the following criteria into your “Statement in support of your application” and describe underneath each criteria how you meet it, giving examples. You will be scored on how you meet each criteria.

Essential Criteria

Qualifications, experience and knowledge

PhD in machine learning, computer science, computational physics, applied mathematics or a closely related quantitative discipline, or an equivalent combination of postgraduate qualification and substantial applied research experience Demonstrable track record of original research in machine learning for imaging or inverse problems, evidenced by peer-reviewed publications, theses, opensource software or deployed systems Substantial experience of designing, training and evaluating deep generative models, such as diƯusion models, variational autoencoders, generative adversarial networks (GANs) or normalising flows, using large-scale scientific or imaging datasets and high-performance computing resources Experience of full-waveform inversion or closely related model-based, physicsdriven inverse problems, including the use of learned priors or physics-informed methods to regularise or guide optimisation Experience of building robust, scalable machine learning systems in a research or production environment, including data pipelines, distributed or cloud computing and fault-tolerant workflows

A working knowledge of the physics of wave propagation relevant to ultrasound imaging

Skills and abilities

Advanced programming skills in Python and a modern deep learning framework, such as PyTorch, with proficiency in GPU computing, high-performance computing environments (e.g. Slurm) and cloud platforms, and experience of collaborative software development using version control Ability to plan and deliver an independent programme of research and development to agreed milestones within a multi-site collaborative project, and to take day-to-day ownership of a work package Excellent written and oral communication skills, including the ability to present complex technical work clearly to interdisciplinary audiences and to write highquality scientific papers and technical documentation Ability to work collaboratively and constructively across disciplines and institutions with engineers, physicists and clinicians, and to support and mentor more junior colleagues

Desirable Criteria

Interview / Practical Test)

Interview / Practical Test)

Experience of working with clinical neuroimaging data, particularly MRI and/or CT, or a good understanding of the clinical context of acute neurological disorders such as stroke Experience of working with patient data within NHS information governance and research ethics frameworks

Apply

To apply for this position visit: ucl.ac.uk/jobs Please complete the online application form and use the supporting statement section to outline your interest in joining the High-Dimensional Neurology Group and the Department of Translational Neuroscience & Stroke, and how you meet the essential and desirable criteria for the role. Please also upload a recent CV.

If you have any queries regarding the application process, please contact the Institute HR Team, UCL Queen Square Institute of Neurology, 23 Queen Square, London, WC1N 3BG (email: ion.hradmin@ucl.ac.uk).

For informal enquiries about the role please contact Professor Parashkev Nachev (email: p.nachev@ucl.ac.uk).

The post is available from 1 November 2026 and is funded by the EPSRC until 31 October 2027 in the first instance.

Salary

The post is graded as UCL Grade 8 with salary in the range £56,431 - £64,644 per annum, including

London Allowance.

Progression through the salary scale is incremental. Cost of living pay awards are negotiated nationally and are normally eƯective from 1 st August each year.

Probation

Appointments are subject to receipt of satisfactory references and a probationary period of nine months.

Hours of work

Full time, 36.5 hours per week; times of work are as determined by the line manager or Director.

Hybrid working arrangements

The nature of this role requires you to spend most of your time working on-site (On-site First) and you are expected to spend at least 80% of your working hours on campus.

Annual leave

Annual leave is 27 working days for a full-time member of staƯ + 6 UCL closure days in addition to 8 Bank Holidays.

Pension

Appointments are superannuable under the Universities Superannuation Scheme (USS).

Other benefits

UCL is a dynamic, global university based in one of the most exciting capital cities in the world. Not

only does working at UCL oƯer the opportunity to work with some of the greatest intellects in the world, it also oƯers competitive terms, conditions and benefits to its staƯ. In our 2025 staƯ survey, 95% of ION staƯ rate the Institute as an enjoyable place to work and over 90% feel that the Institute has a positive working culture.

As part of the UCL community you can access free lunch hour lectures, exhibitions and museums and collections. On campus UCL has the Bloomsbury theatre hosting a range of performances and a series of bars, cafes, and other facilities, which UCL staƯ can use.

In addition to 41 days annual leave (inclusive of public holidays and closure days) and generous pension schemes, UCL provides a number of other staƯ benefits.

UCL benefits and policies apply equally, whatever the sexual orientation and/or gender identity of employees. Benefits and policies relating to employees’ partners includes both diƯerent sex and same sex partners.