GaiaLab · Houston, TX, USA
LassaAI
Machine Learning Outbreak Prediction and Clinical Decision Support for Lassa Fever in Nigeria
Oluwafemi Idiakhoa
Founder, GaiaLab
partnerships@gailabai.com
gailabai.com/lassa
Version 1.0 · May 2026

Model Performance (Test Set 2024–2025, Prospective Holdout)

0.9994
AUROC
1.000
Recall
0.910
Precision
0.953
F1 Score
0
Missed outbreaks

Training used strict temporal cross-validation (train ≤2021, validate 2022–23, test 2024–25) to prevent data leakage. All metrics are retrospective. Prospective validation is the immediate next step and the subject of a NIH Fogarty R21 application.

What LassaAI Does

Outbreak Prediction Model

  • Generates weekly outbreak probability for all 37 Nigerian states
  • 4–8 week forward prediction horizon
  • Trained on 27,417 state-week observations, 2011–2026
  • Updates automatically when new NCDC sitrep data are available
  • Live dashboard: gailabai.com/lassa

Clinical Copilot

  • Bayesian probability calculator for healthcare workers at point of care
  • 23 clinical and epidemiological features; literature-derived weights
  • Calibrated to 7/7 published clinical scenarios (Bausch 2001; Okokhere 2018; WHO 2017)
  • Freely accessible: gailabai.com/lassa-copilot
  • Always recommends RT-PCR confirmation

Dataset

ItemValue
Time period2011–2026 (Week 22)
States covered37 (all states + FCT)
Observations27,417 state-weeks
Confirmed cases8,138
Deaths910
Case sourceNCDC Weekly Situation Reports
Weather sourceOpen-Meteo ERA5 reanalysis
Outbreak rate6.1% (1,684 positive / 25,733 negative)

Top Feature Importance

8-wk rolling cases
49.6%
Cases lag 1 week
26.1%
4-wk rolling cases
13.6%
Cases lag 2 weeks
2.2%
Weeks into dry season
1.8%
Calendar month
1.7%
Weather (combined)
~1%

Recent case history accounts for 89% of predictive power. Meteorological features (temperature, rainfall, humidity) contribute ~1%, consistent with published Lassa ecology literature.

Proposed Research Partnership

We are preparing a NIH Fogarty International Center R21 application (up to $275,000 / 2 years) with two aims:

  • Aim 1 (Year 1): Prospective validation with NCDC — 12-month blinded prediction logging vs. confirmed case outcomes
  • Aim 2 (Year 2): Clinical validation of the Copilot at a Nigerian Lassa fever treatment centre (target: ISTH Irrua or FMC Owo)

Application deadline: October 16, 2026 (LOI due September 16, 2026)

Top 5 States by Confirmed Cases (2011–2026)

StateConfirmed CasesNote
Edo2,574ISTH Irrua referral base
Ondo1,638FMC Owo / ILFRC site
Bauchi1,040
Taraba884
Ebonyi728

Key Limitations (Disclosed)

(1) Retrospective only — prospective validation not yet complete. (2) Confirmed cases only — true Lassa incidence substantially higher. (3) State-capital weather — rural microclimates may differ. (4) Reporting variability — testing capacity varies by state and year. (5) Clinical Copilot — not validated in a prospective clinical study; not a diagnostic device.

Not a medical device. LassaAI is a research and clinical decision support tool. It has not been approved by any regulatory authority. All clinical decisions must be made by qualified healthcare professionals with RT-PCR laboratory confirmation. For research and public health purposes only.