I have worked extensively on developing novel artificial intelligence (AI) and machine learning (ML) methods and on their application to problems in healthcare and medicine. A particular focus of my research has been biomedical image computing, covering all aspects from image acquisition to image analysis and interpretation.

My current research interests include:

- Combining AI with inverse modelling approaches to achieve the next level of performance in biomedical imaging, enabling higher resolution, greater sensing sensitivity, faster imaging, and multi-dimensional, quantitative imaging.
- Establishing foundational AI models that integrate multimodal patient data (imaging, sensing, omics, clinical and lifestyle information as well as environmental factors) to create a comprehensive representation of patients' health and disease status.
- Combining these AI models with biophysical and physiological models as well as domain-specific clinical knowledge to enable more accurate and interpretable modelling and understanding of health and disease characteristics in patients and populations.
- Developing reliable and privacy-preserving AI methods that are safe (e.g., ensuring no harm to patients), secure (e.g., resilient against manipulation), fair and responsible (e.g., following ethical norms and focusing on people’s needs).

A strong emphasis of my research has been the translation of the AI and ML methods for novel application of imaging and computing technology to improve the diagnosis and stratification of patients with dementia, stroke and traumatic brain injury, improve the understanding of brain development (in-utero and ex-utero) as well as for the comprehensive diagnosis of and management of patients with cardiovascular disease.

To facilitate translation of research to healthcare applications, I have co-founded the Imperial College spin-off company IXICO (http://www.ixico.com) to commercialize the image analysis techniques developed in our research group. IXICO provides imaging analysis solutions and imaging biomarkers for clinical trials in the pharmaceutical industry and healthcare diagnostics. Since 2003, IXICO has rapidly grown and now employs 50 people. IXICO’s technologies have been used in more than 60 large-scale clinical trials involving more than 20.000 patients. The imaging biomarkers based on our image analysis technologies have recommended by the European Medicines Agency as a tool to enrich recruitment into regulated clinical trials in Alzheimer’s disease (AD).