About

Hi — I’m Yusuf, a PhD student in the Department of Biostatistics and Health Informatics at the Institute of Psychiatry, Psychology & Neuroscience (IoPPN), King’s College London. I’m part of the EPSRC Centre for Doctoral Training in Data-Driven Health, co-funded by LifeArc, and supervised by Prof. Richard Dobson, Prof. Alfredo Iacoangeli, Prof. Honghan Wu, Dr Chris Tomlinson and Dr Zhi Yao.

Research

I build machine-learning systems for rare and neurodegenerative disease, with a focus on Motor Neuron Disease (MND / ALS). The central problem I work on is that rare diseases are where prediction matters most — early diagnosis, progression forecasting, subtype discovery — and also where patient cohorts are smallest, so standard ML tends to fail.

My thesis develops an end-to-end pipeline that fuses multimodal electronic health records, biomedical knowledge graphs, genomics, and clinical foundation models to make prediction work in the small-cohort regime, applied to MND for early risk prediction, progression forecasting, and patient stratification.

More broadly, I’m interested in the methodology that makes small-cohort rare-disease AI trustworthy: knowledge-graph reasoning with LLMs, multimodal representation learning and uncertainty quantification.

News

• [April 2026] - Presented at AI meets Knowledge Graphs & Memory on knowledge graphs and LLMs in rare diseases.
• [January 2026] - Featured in an ALS-TDI blog post.
• [December 2025] - Presented at the International Symposium for ALS/MND: “Characteristics and Early Diagnosis of Motor Neuron Disease (MND) in 67 million individuals in England”.
• [July 2025] - “Characteristics and Early Diagnosis of Motor Neuron Disease (MND) in 67 million individuals in England: a comparative study on phenotyping models derived by AI, Knowledge Graphs and the MND Association” preprinted on medRxiv.
• [May 2025] - “BioHopR: A Benchmark for Multi-Hop, Multi-Answer Reasoning in Biomedical Domain” accepted to ACL Findings.