PhD Scholarship in Multimodal Federated Learning and Medical Image Analysis

Monash•Clayton, NC

6d•Onsite

About The Position

Expressions of interest are sought from outstanding candidates for PhD study in the Department of Data Science and Artificial Intelligence at the Faculty of IT, Monash University. As part of this scholarship, the successful candidate will develop novel federated learning and multimodal deep learning models for healthcare. The project will focus on enabling privacy-preserving learning from distributed healthcare data sources, including longitudinal medical imaging, electronic health records, pathology, and other clinical data. The candidate will investigate novel approaches for multimodal representation learning, foundation model adaptation, and federated learning to improve disease diagnosis, risk prediction, and clinical decision support. Applications will include chronic diseases such as cancer, diabetes, and rheumatoid arthritis. This research forms part of the National Infrastructure for federated learNing in DigitAl health (NINA), a national initiative aimed at advancing privacy-preserving AI infrastructure and analytics for healthcare across Australia. The candidate will be supervised by Dr Yasmeen George ([email protected]), and will work closely with collaborators across medicine, healthcare organisations, and industry partners.

Requirements

A relevant Honours or Masters degree with H1 or equivalent.
Meet the eligibility criteria for PhD candidature at Monash University.
Satisfy Monash’s English Language Proficiency requirements.
Demonstrate knowledge of machine learning, medical image analysis, computer vision, federated learning, foundation models, adaptation techniques, multimodal learning, longitudinal image analysis or related areas, evidenced through coursework, research projects or publications.
Experience with Python programming and the use of high-performance computing infrastructure for AI research.
Excellent written and verbal communication skills.
Ability to work independently, as well as part of a team.
Ability to plan, organise, manage multiple tasks and meet deadlines.
Analytical thinking, data analysis and critical problem-solving skills.
Be enrolled full time and on campus.
Applicants who already hold a PhD will not be considered.

Responsibilities

Develop novel federated learning and multimodal deep learning models for healthcare.
Focus on enabling privacy-preserving learning from distributed healthcare data sources, including longitudinal medical imaging, electronic health records, pathology, and other clinical data.
Investigate novel approaches for multimodal representation learning, foundation model adaptation, and federated learning to improve disease diagnosis, risk prediction, and clinical decision support.