About the Research Centre
In collaboration with the CityU, the University of Oxford and the Karolinska Institutet, the Hong Kong Centre for Cerebro-cardiovascular Health Engineering (COCHE) was established with the support of the ITC-InnoHK Programme to develop new ways to tackle cardiovascular disease (CVD), which is the world's leading cause of morbidity and mortality. To address the most pressing health challenges caused by CVD, COCHE focuses on developing innovative technologies such as flexible and wearable devices, biomedical and molecular imaging, nano-biosensing, micro-robotics, healthcare AI systems, and other groundbreaking methods to enable early prevention, diagnosis and intervention of the disease. These technologies can reduce human costs while alleviating the healthcare system's financial and social burdens.
Job Description
Focus: Federated Learning, Privacy-Preserving AI, & Real-World Clinical Application
________________________________________
Position Overview
We are seeking highly motivated Postdoctoral Research Fellows and Research Assistants/Engineers (Bachelor’s/Master’s level) to join an international research initiative focused on Federated Learning (FL) for medical data. This project is a unique collaboration among leading medical and technical teams from the UK, China, and Australia.
The successful candidates will conduct research into Federated Learning methodologies and contribute to their deployment in real-world clinical settings. This role is ideal for researchers who possess a deep mastery of Deep Learning and a passion for high-quality software engineering, offering the opportunity to publish in top-tier venues while seeing their work translated into tangible medical applications.
________________________________________
Key Responsibilities
1. For Postdoctoral Research Fellows
• Conduct research to develop and optimize novel federated learning algorithms, with emphasis on data heterogeneity, communication efficiency, and privacy preservation.
• Adapt and implement advanced Federated Learning architectures for large-scale medical datasets and Large Language Models.
• Facilitate the integration of research models into real-world clinical applications, ensuring seamless interoperability across diverse institutional environments.
2. For Research Engineers
• Develop robust backend infrastructure for medical federated systems, prioritizing stability, scalability, and security.
• Create clean, maintainable, and efficient codebases to enable the seamless deployment of federated learning models in diverse institutional settings.
• Drive system interoperability by designing and managing data pipelines and APIs, ensuring smooth integration into clinical workflows.
General Qualification
Essential Qualifications:
• A Ph.D. in Computer Science, Software Engineering, AI, or a related quantitative discipline for postdoctoral applicants.
• Robust experience in training and fine-tuning state-of-the-art LLM architectures.
• Proficient in common deep learning frameworks (e.g., PyTorch).
Preferred Qualifications:
• Prior research or implementation experience in Federated Learning frameworks and privacy-preserving protocols.
• A documented history of successfully migrating theoretical models into practical, deployable systems (e.g., active GitHub profile).
• Experience handling specialized medical data formats (e.g., DICOM, EHR) and knowledge of clinical data privacy standards.
________________________________________
Benefits of the Position
• Direct engagement with a global consortium of world-leading academic institutions.
• The opportunity to observe the direct clinical application of your research, bridging the gap between theory and patient care.
• Access to high-performance computing (HPC) resources and a vibrant, interdisciplinary research environment in COCHE.
Preferred Hiring Qualifications:
Remuneration & Duration of Appointment
The application Materials that Should be Included
Application Closing Date:
Additional Information/links:
Application Procedures:
Applicants should submit their Curriculum Vitae to COCHE HR at HR@hkcoche.org.
Note: We encourage candidates to include links to GitHub repositories, technical portfolios, or summaries of significant software implementations to demonstrate coding proficiency.