Portfolio

Real-world evidence

cost-utility

cost-effectiveness

health data science

cdisc

sdtm

adam

define.xml

SAS

machine learning

Ousmane Diallo - Portfolio | Clinical Programming, HEOR, RWE & Machine Learning Projects

Author

Ousmane DIALLO

Published

Invalid Date

Welcome to my professional portfolio

A curated showcase of clinical research, real-world evidence (RWE), and health analytics projects developed using
SAS, R, and Python — bridging statistical rigor, reproducibility, and regulatory science.

Areas of focus

Clinical Trials Programming (CDISC, SAS)
Health Economics & Outcomes Research (HEOR)
Real-World Evidence (RWE) Analytics
Machine Learning & Advanced Statistical Modeling
Geospatial Epidemiology

Explore each section for methodology, R/SAS code, and key results.

🧬 Clinical Trials Programming (CDISC / SAS)

Regulatory-compliant workflows for clinical data analysis and submission-ready deliverables.

My clinical programming work focuses on end-to-end pipelines for clinical trial data using SAS and CDISC standards.
These projects replicate real-world regulatory workflows, from raw data cleaning to submission-ready outputs.

Highlighted Capabilities

SDTM and ADaM dataset derivation
Tables, Listings, and Figures (TLFs) programming
Define-XML and annotated CRF support
QC checks and validation workflows

Featured Projects

🔹 End-to-End Study Data Tabulation Model (SDTM)
🔹 ADaM BDS Derivations for Efficacy & Safety
🔹 Tables, Listings, and Figures (TLFs)
🔹 define.xml Generation & Validation Workflow

💰 Health Economics & Outcomes Research (HEOR)

Cost-effectiveness and cost-utility analyses supporting value-based healthcare decisions.

These projects apply economic evaluation and outcomes research methods to clinical and public health programs, highlighting the impact of interventions on cost and health outcomes.

Featured Projects

🔹 Cost-Effectiveness of the REPPOP Pediatric Obesity Management Program
🔹 Cost-Utility of Lung Transplantation

They demonstrate experience with:

Cost-effectiveness and cost-utility models
Decision-analytic frameworks and sensitivity analyses
Communication of results for policymakers and healthcare stakeholders

📊 Real-World Evidence (RWE) & Statistical Modeling

Observational analytics and evidence generation for clinical and public health impact.

This section covers observational research and real-world data analyses that inform clinical and policy decisions.
The projects integrate data from surveys, hospitals, and surveillance systems.

Featured Projects

🔹 Patient-level Risk Factors for Clinical Malaria in Children (2–60 months)
🔹 Hospital Mortality and Positive Blood Culture with Difficult-to-Treat Resistance – Marseille, France
🔹 Major Discrepancy Between Factual Antibiotic Resistance and Consumption in Southern France
🔹 Early Arterial Catheter Use and 28-Day Mortality in Mechanically Ventilated ICU Patients (MIMIC)

These projects highlight:

Study design for observational data
Regression modeling and risk factor analysis
Translation of findings into actionable recommendations

🗺️ Geospatial Analysis & Epidemiology

Spatial and spatio-temporal modeling for malaria surveillance and intervention planning.

Here, I use geospatial methods and epidemiological modeling to understand how disease risk and intervention coverage vary across space and time.

Featured Projects

🔹 Risk Factors – ITN Use in Guinea (DHS 2018)
🔹 Spatio-temporal Modeling of Treatment-Seeking Coverage Using DHS Data and INLA
🔹 Retrospective Analysis of Malaria Trends in Burkina Faso

Key themes include:

Spatial risk mapping and cluster detection
Linking survey, facility, and environmental data
Supporting targeted intervention strategies

🧩 Health Data Engineering & Multi-Source Data Harmonization

Data extraction, cleaning, transformation, harmonization, and automation for clinical, survey, and health information system datasets.

This section highlights my work in data engineering across multiple health data sources, including SQL-based analytics, survey data harmonization (DHS/MIS), and health management information systems (HMIS). The projects demonstrate end-to-end pipelines from raw data to analysis-ready datasets.

Featured Projects

🔹 Antimicrobial Resistance Surveillance: Analysis-Ready Dataset Using SQL

🔹 Multi-Country DHS/MIS Data Harmonization for Malaria Analytics

🔹 Data Quality Pipeline for Routine Surveillance Data from HMIS

The project demonstrates:

SQL data extraction and transformation across large datasets
Cross-country survey data harmonization (DHS/MIS)
Health facility data standardization (HMIS/DHIS2)
Dataset cleaning and missingness handling
Variable standardization across heterogeneous data structures
Survey-weighted estimation workflows
Time-series construction and anomaly detection
Surveillance logic for early-warning signals
Automated pipelines for multi-source data integration

🤖 Machine Learning & Predictive Analytics

Advanced modeling and classification for outcome prediction and health system performance.

These projects apply machine learning and multivariate methods to extract patterns and build predictive models from complex health data.

Featured Projects

🔹 Seasonality Archetypes Using Machine Learning
🔹 Predictors of Insecticide-Treated Net Ownership and Use in Guinea

Methods used include:

Clustering and dimensionality reduction
Supervised learning for classification and prediction
Interpretation of models in a public health context

📞 Let’s Collaborate

If you’re seeking expertise in data science, biostatistics, epidemiology, or clinical data programming, I’m open to:

Collaboration on research projects
Consulting on clinical trials programming and CDISC workflows
RWE and HEOR analytics in partnership with clinical or industry teams

👉 Contact Me to discuss your project or proposal.

Ousmane Diallo, MPH-PhD – Biostatistician, Data Scientist & Epidemiologist based in Chicago, Illinois, USA. Specializing in SAS programming, CDISC standards, and real-world evidence for clinical research.