Postdoc multimodal cardiovascular AI research

In this postdoctoral project, you will develop innovative multimodal AI frameworks to integrate diverse cardiovascular data sources. Your research will focus on combining imaging data (e.g., cardiac MRI, echocardiography), signals (e.g. electrocardiography, intracardiac signals), clinical measurements, molecular and genomic profiles to improve diagnostic accuracy and patient stratification in cardiovascular diseases. You will work on synthetic data generation techniques to improve multi-centre training while maintaining privacy, and develop methods to optimize cross-modal representations for enhanced predictive modeling.

As a postdoctoral researcher, you will:

• Design and implement novel multimodal AI architectures for cardiovascular applications;
• Develop synthetic data generation methods to augment sparse modalities;
• Create frameworks for joint optimization of cross-modal representations;
• Validate models using clinical datasets in collaboration with cardiologists;
• Mentor PhD and MSc students and present findings at international conferences;
• Intention to apply for grants to expand our group, and potentially start your own research group.

• You have a PhD in Computer Science, AI/ML, Biomedical Engineering, or related field;
• You demonstrate strong programming skills in Python and have experience with deep learning frameworks such as PyTorch (experience with high performance computing infrastructures is a plus);
• You have experience in medical image analysis (multimodal learning experience is a plus);
• You have a publication track record in top AI/ML or medical informatics venues such as NeurIPS, MICCAI, or SPIE Medical Imaging and/or in translational venues such as Nature portfolio journals (Nature Medicine, Nature Cardiovascular Medicine, npj Digital Medicine) ;
• Experience with clinical healthcare data systems is a plus.