I enjoy reading, cooking, and DIY projects, and I've been combining these activities to spend quality time with my family. I have a deep appreciation for culture and history, reflecting my interest in the past, while my research focuses on the future with artificial intelligence and medical imaging.

I started working in neuroimaging analysis engineering after my master's degree in artificial intelligence and discovered its value in quantitative analysis for brain research and disease diagnosis. Realising the potential to apply my expertise in AI to this field, I saw an opportunity to make a real-world impact and help people. Growing up in a family of doctors, researching MS and other neurodegenerative diseases helped bridge my knowledge with my family's dedication to healthcare.

One of the most exciting developments in MS research is the application of AI in assisted diagnosis and patient monitoring. Recent breakthroughs have shown that monitoring brain and lesion changes can significantly improve patient status assessment and treatment decisions. The next step, which aligns with my current research project, is to predict disease progression using patient records. This advancement holds great potential for early intervention and personalised treatment strategies, important to MS patient care

My current research project aims to utilise the recently developed and popular AI technique, large language models (LLMs), to advance the prediction of disease progression in MS, focusing on electronic health records (EHR) and neuroimaging. This study seeks to explore, validate, and apply LLMs in a clinical setting to evaluate their predictive performance and analyse findings to identify novel biomarkers or clinical indicators of rapid disease progression.

Recent progress in deep learning and AI has already offered promising pathways to better understand MS, such as through quantitative analysis of MS lesions and their activities in longitudinal studies. My research builds on this by leveraging cutting-edge AI techniques like LLMs to identify additional factors or biomarkers relevant to disease progression. This could contribute to early intervention strategies and personalised treatments, ultimately enhancing patient outcomes and advancing our overall understanding of MS.

  I enjoy working on algorithms and pushing their performance forward, especially when achieving state-of-the-art results in classification, segmentation, and prediction tasks. The challenges can be daunting, as algorithm performance is influenced by multiple factors such as imaging quality, hyperparameters, data bias, and pre- and post-processing. However, my sense of accomplishment upon solving these complex issues is immensely rewarding.