Principal Scientist, Imaging Biomarkers — Boehringer Ingelheim
Imaging pipelines · Quantitative biomarkers · Reproducible analysis at scale

I build end-to-end pipelines for processing and analysing medical imaging data — taking MRI, fMRI, PET and DTI from raw acquisition through to the quantitative measures a study actually reports. Neuroimaging is the core of the work, increasingly alongside retinal imaging and abdominal MR.

• Processing and QC pipelines across MRI, fMRI, DWI/DTI and PET — automated, containerised, and built to be handed over
• Quantitative imaging biomarkers supporting diagnosis, therapeutic target identification and drug development
• Reproducible workflows at scale on HPC (SLURM + Singularity), so multi-site studies run consistently
• Neurodegenerative and psychiatric disease — Alzheimer's, Parkinson's, schizophrenia, depression — with the same approach now extending across organ systems

Earlier in my career the work was more squarely in machine learning and deep learning: evolutionary-algorithm classifiers for resting-state fMRI in my PhD, white-box models separating depression from controls in UK Biobank data, deep learning for A/T/N staging from tau-PET, and the ML behind diagnostic devices at ClearSky. That background still shapes how I build and evaluate analyses.

Imaging biomarkers across organ systems — Boehringer Ingelheim, current
Building and running the pipelines and analyses that turn imaging data into quantitative biomarkers for drug development. Neuroimaging is the core of the work, with the remit extending to retinal imaging and abdominal MR.

ADPrep — end-to-end multimodal imaging pipeline
An automated pipeline standardising MRI, fMRI, PET and DTI for large multi-site studies, from raw acquisition through quality control to analysable measures. Set for integration into the GRIP platform, a Gates Ventures initiative.

ClearSky diagnostic device portfolio
Seven years as CTO of ClearSky Medical Diagnostics, developing ML-based devices for neurodegenerative diagnosis and monitoring: PD-Monitor, LID-Monitor and MCI-Monitor.

Imaging and data coordination for multi-site trials
Technical and data coordination across Parkinson's disease trials — PD-STAT (235 participants, 23 NHS hospitals), the Phase III Exenatide-PD3 study and the Phase II AZA-PD study.

Peer-reviewed work appears in Nature Communications, Brain, JAMA Neurology, Science Translational Medicine, EMBO Molecular Medicine and Alzheimer's & Dementia — selected list on my website or Google Scholar.

Pipelines & infrastructure

Neuroimaging

Languages