I started my career where data has real stakes: clinical diagnostics and molecular genetics, where a misread result affects a patient, not just a metric.

That foundation shaped everything about how I work: structured, precise, and outcome-focused. Today I build data pipelines, machine learning models, and analytical systems that support better clinical and business decisions.

I bring something most data analysts don't have, deep domain knowledge in healthcare, paired with hands-on experience building the data systems that support it.

Clinical Analytics       ██████████████████░░   Healthcare data, patient risk, diagnostics
Machine Learning         ████████████████░░░░   Predictive modelling, classification, NLP
Business Intelligence    ██████████████░░░░░░   Power BI dashboards, KPI reporting, ETL
Forecasting              ████████████░░░░░░░░   Time-series models, inventory planning

Zion Tech Hub Data Challenge — May 2026

Analyzed 350 hypertension patients to predict blood pressure control failure and identify the most effective antihypertensive drug by comorbidity profile.

• Built a logistic regression model achieving 100% recall at threshold 0.3 — catching every at-risk patient before the 3-month review
• Identified combination therapy as the top-performing treatment across diabetes, dyslipidemia, and obesity cohorts (25% control rate vs 3–11% for monotherapy)
• Uncovered that ACE inhibitors produced 0% control rate across all 4 comorbidity groups — a clinically significant prescribing finding
• Delivered full analytical report with EDA, ML modelling, threshold tuning, and evidence-based clinical recommendations

Python Scikit-learn Pandas Matplotlib Seaborn Logistic Regression Random Forest

Developed an interpretable clinical risk model to predict prolonged hospital stay using logistic regression and probability-based patient stratification.

• Built for clinical interpretability — designed so results can be acted on by medical teams, not just data scientists
• Used probability-based stratification to rank patients by risk level, enabling prioritised early intervention

Python Scikit-learn Logistic Regression Clinical Risk Modelling

Live deployment — Blunce Medical Diagnostics, Lagos

Built and deployed an ML model to predict which discharged patients require proactive follow-up — calls, home visits, or early appointments — directly improving patient retention and care continuity.

Python Machine Learning Healthcare Analytics Patient Retention

Built a predictive churn model to identify high-risk customers and simulate revenue retention strategies.

• Applied logistic regression to flag churners before they disengaged
• Simulated revenue impact of targeted retention interventions

Python Scikit-learn Logistic Regression Business Analytics

I am open to healthcare analytics roles, clinical data positions, and remote opportunities globally.

"I started in healthcare because data there has real consequences. I stayed in data because the right analysis can change outcomes, for patients and for businesses."