FSI-AIBR
Start a research conversation

FEMI SAMSON INSTITUTE OF AI & BIOMEDICAL RESEARCH

Building intelligence for a healthier future.

FSI-AIBR develops AI-powered biomedical and public-health intelligence systems that transform complex health, climate and biological data into prediction and decision support starting with malaria in Nigeria.

SIGNAL → SCIENCE → INTELLIGENCE → ACTION

DATA SCIENCE AI + ANALYTICS INTELLIGENCE DECISION IMPACT

ABOUT FSI-AIBR

Where biomedical science meets artificial intelligence.

FSI-AIBR is an independent African research institute building intelligence systems for public health bringing together field epidemiology, spatial modelling and agentic AI around real disease-surveillance problems.

Science

Biomedical & public-health research

Field epidemiology and physiological science grounded in real surveillance data, not simulated scenarios.

Data

Health, climate & scientific datasets

DHS/MIS biomarker surveys integrated with multi-decade CHIRPS-NDVI climate and remote-sensing records.

Intelligence

AI, analytics & decision support

Machine learning and explainability methods that turn integrated data into spatial and temporal risk intelligence.

OBSERVEUNDERSTANDPREDICTACT

WHO WE ARE

An independent African research institute — not a consultancy or a startup — working at the intersection of field epidemiology, spatial modelling, malaria surveillance and agentic AI.

WHY WE EXIST

Public-health decisions in Nigeria and across Sub-Saharan Africa deserve intelligence systems built from the region's own data, not adapted from elsewhere after the fact.

WHAT WE BUILD

Disease-intelligence platforms, physiological AI research tools, and the data infrastructure that connects surveillance, climate and biomedical signal into decision support.

INSTITUTIONAL ROADMAP

Where FSI-AIBR is in its own build-out.

Stages, not dates — each one reflects where the institute's work actually stands today.

Foundation

Established

Institute formed; initial research direction and leadership in place.

Research

Underway

Malaria digital-twin methodology, preprint and physiology research active.

Platform

Underway

Migrating research prototypes from Streamlit into reusable web infrastructure.

Network

Underway

Partnerships including the Physiological Society of Nigeria collaboration.

Scale

Upcoming

Expansion across additional disease domains and research partners.

OUR PEOPLE

Who's behind FSI-AIBR.

Daniel Onimisi, Founder and Executive Director of FSI-AIBR

Daniel Onimisi

FOUNDER & EXECUTIVE DIRECTOR · FIMC, CMC, CMS

Works at the intersection of field epidemiology, spatial modelling, malaria surveillance and agentic AI for public-health decision intelligence.

FSI-AIBR is early-stage hence this section will grow as the institute's research staff, fellows and collaborators expand. We list people here as their roles are confirmed, not before.

RESEARCH

Six domains, one question: what can data reveal about health before a crisis begins?

01

Biomedical AI

Applying machine learning to biomedical research and physiological science.

Explore →

02

Disease Intelligence

Surveillance, risk prediction and early-warning systems — spanning the Malaria Digital Twin and EPIDEXA.

Explore →

03

Environmental Health

How climate and ecological conditions shape disease risk across Nigeria's 774 LGAs.

Explore →

04

Digital Health

Data systems and dashboards that make population-health information usable by decision-makers.

Explore →

05

Agentic Research

AI agents that collaborate across literature, data and analysis as part of the research process.

Explore →

06

Geospatial Intelligence

Spatial modelling and remote sensing to understand where and when health risk concentrates.

Explore →

FLAGSHIP INTELLIGENCE SYSTEM

Nigeria Malaria Digital Twin

A Random Forest and SHAP-based disease-intelligence platform integrating decades of climate signal with biomarker surveillance to explore malaria risk across every Nigerian LGA.

PILOT · LIVE ON STREAMLIT
34+
Years of CHIRPS–NDVI climate data
774
Nigerian LGAs covered
Multi-source
Climate + biomarker survey data
ML
Random Forest + SHAP modelling

THE INTELLIGENCE LAB

FSI-AIBR's projects, as one connected research ecosystem.

Hover, tap, or tab through the nodes to see what each project is, what it runs on, and where it stands.

RESEARCH

Select a node

Hover, tap, or focus a node in the network to see the project behind it.

AGENTIC RESEARCH

AI agents as part of the research team, not a replacement for it.

PhysioLab's research pipeline — literature through to a human physiologist who retains scientific judgment and final approval.

Literature Agent
Surveys existing research on heat-stress cardiovascular adaptation
WAITING
Cardiovascular Agent
Models skin vs. muscle/cardiac blood-flow balance
WAITING
Thermoregulation Agent
Tracks repeated heat-exposure adaptation over time
WAITING
Autonomic Agent
Assesses autonomic nervous system response
WAITING
Data & Statistics Agent
Runs the quantitative analysis
WAITING
Pathophysiology & Critic Agent
Flags where the balance becomes cardiovascular risk
WAITING
Human Physiologist
Reviews findings and retains final scientific judgment
WAITING

PhysioLab is in development, built for a Physiological Society of Nigeria presentation — not yet an operational system.

PHYSIOLAB AI

The future of physiology is computational.

AI doesn't replace physiology, it gives the discipline new instruments: digital biomarkers, physiological modelling, and agentic research pipelines that let physiologists ask bigger questions of their data.

"How does repeated heat exposure shift the balance between skin blood flow and muscle or cardiac blood flow and where does that balance become cardiovascular risk?"

The research question behind PhysioLab's first demonstration, built for the FSI-AIBR × Physiological Society of Nigeria partnership.

RESEARCH IN MOTION

Research stories

CLIMATE & HEALTH

When climate signals become health signals

How three decades of rainfall and vegetation data feed a malaria risk model.

DISEASE INTELLIGENCE

What a malaria digital twin can tell us about early warning

Inside the Random Forest model behind Nigeria's LGA-level risk estimates.

AGENTIC AI

Can AI become part of the biomedical research team?

What PhysioLab's agent pipeline is designed to do — and what it isn't.

METHODOLOGY

From data to decisions.

Every FSI-AIBR system follows the same discipline: know what the data can and can't say before turning it into a recommendation.

Fragmented data
Integrated data
Analytics
Predictive models
Intelligence
Decision support

Scientific integrity

Evidence before claims.

Human oversight

AI supports judgment, never replaces it.

Transparency

Methods and assumptions are documented.

Privacy

Health data is protected by design.

PROGRAMS

Who we're developing.

SPARTAN Applied Intelligence Fellowship

FSI-AIBR's applied-AI training program, from Python foundations through applied engineering thinking.

Physiology Hub

A capacity-building partnership with the Physiological Society of Nigeria and SPAN.

Fellowships

Research and technology fellowships under FSI-AIBR.

Student programs

Early exposure to applied research and technology.

PARTNERSHIPS

Who can build with us.

Research partners

Universities and research institutions.

Public health partners

Government agencies and health organisations.

Technology partners

AI, cloud, data and infrastructure organisations.

Corporate partners

Organisations investing in health intelligence.

Training partners

Institutions developing biomedical and AI talent.

PUBLICATIONS

Selected research output.

Publication records populate here as they're finalized including the malaria digital-twin preprint. This section is intentionally left as a live feed, not a static list.

START A RESEARCH CONVERSATION

Tell us what you're trying to understand and we'll tell you what the data can say.