About | Me
I'm Judah Akinlaja, a health data scientist working at the intersection of clinical knowledge and applied analytics. My work is about one thing: turning health data into decisions that actually improve how care is delivered.
My path into data science started in clinical science. I studied Basic Medical and Clinical Sciences at the University of Ilorin, Nigeria, where I built a deep understanding of human physiology, disease pathways, and the scientific method. That foundation matters. It means I don't just run models on health data, I understand what the variables represent, why certain patterns emerge, and what clinicians need from the numbers.
I then moved to the University of Manchester for an MSc in Health Data Science, where I trained in machine learning, statistical modelling, medical image analysis, and clinical bioinformatics. Manchester gave me the technical toolkit. Nigeria gave me the clinical instinct. The combination is what shapes how I approach every project.
What drives me is a specific frustration: the gap between what health data could do and what it actually does. Hospitals sit on years of operational data that never gets analysed. Clinical teams make decisions based on intuition when the evidence is right there in the system. I build the analytical bridges that close that gap, not with flashy AI, but with rigorous, interpretable solutions that people actually trust and use.
Clinical Foundations
BSc in Basic Medical/Clinical Sciences, understanding the biology behind the data
Technical Training
MSc Health Data Science, Manchester: ML, deep learning, medical imaging, bioinformatics
Applied Focus
Building for adoption: models clinicians trust, dashboards managers actually open
Next Chapter
Joining NHS England's Early Talent programme in applied health analytics
What I've | Built
Every project I take on starts with a real problem, not a Kaggle dataset or a tutorial exercise. These are the health data bottlenecks I've chosen to tackle, each one grounded in a system that affects real patients and real budgets.
DNA Prediction Engine
Predicting missed NHS appointments, a problem costing the health system an estimated £1.2 billion annually. Built a classification model that identifies high-risk no-shows before they happen, enabling targeted intervention.
Cancer Pathology Automation
Automated detection of cancerous tissue in histopathology slides using deep learning. Designed to support pathologists with early screening, not replace clinical judgement.
Disease Surveillance Dashboard
Real-time epidemiological tracking across population-level datasets. Built to give public health teams faster access to outbreak patterns and intervention metrics.
Resource Allocation Models
Analytical frameworks for NHS commissioning decisions, helping trusts understand where money should go based on demand forecasting and outcome data.
How I | Think
Start with the decision, not the dataset
The first question is never "what does the data say?" It's "what decision needs to be made, and what would change someone's mind?" Every project I take on begins with the stakeholder's problem. The model is just the vehicle.
A model nobody trusts is a model nobody uses
Interpretability is not optional. It's the difference between a dashboard that gathers dust and an insight that changes how a ward operates. I build for adoption, not accuracy leaderboards. If the clinical team can't explain why the model flagged a patient, the work isn't finished.
Understand the money, not just the maths
Who commissions the work, who funds the intervention, and who carries the risk if it fails. These questions shape every recommendation. Ignoring procurement, budgets, and political context produces technically correct analysis that never reaches a decision.
Build for systems that don't exist yet
The long game: bringing rigorous health data science to markets where the infrastructure is still being built. Nigeria and West Africa first. Not charity, not parachute work. Sustainable, locally-owned analytical capacity that compounds over decades.