evidoria

← Back to browse

Good practice Imported

3S — Senegal's AI-Assisted One Health Platform Cuts Disease-Alert Time by 39%

Senegal · Podor · See the Senegal profile

Evidence: Observational / pre–post Top 62% 53/100 · Ask Evidence Copilot about this practice

Senegal's Ministry of Health ran the 3S One Health platform — chatbot Taggàt, alert system Jottali, dashboard Gëstu — in three pilot districts, issuing 330 zoonotic-disease alerts in 2025 and cutting detection time from 4.6 to 2.8 days, a peer-reviewed study found.

330
Alerts issued (Apr-Oct 2025)
144 (43.6%)
Alerts validated (Apr-Oct 2025)
-39 %
Reduction in average detection time (Apr-Oct 2025)
4.6 to 2.8 days
Average detection time, before vs after
3S — Senegal's AI-Assisted One Health Platform Cuts Disease-Alert Time by 39%

Details

Maturity
Pilot
Promoter
Ministry of Health and Social Action, Vaccine Surveillance and Response Division (with AI4PEP)
Period
2025
Keywords
public health, One Health, epidemic surveillance, community health

Context

Senegal faces recurring zoonotic-disease outbreaks such as Rift Valley fever that spread silently before official surveillance detects them. The Ministry of Health and Social Action, with the AI4PEP research network, built the 3S One Health platform.

Activities

Three linked modules were piloted in Podor, Kedougou and Pikine health districts: Taggat (AI chatbot for symptom identification), Jottali (community-reporting alerts) and Gestu (monitoring dashboard).

Results

330 alerts were issued between April-October 2025, of which 144 (43.6%) were validated. Digital reporting cut average detection time by 39%, from 4.6 to 2.8 days, with some signals flagged up to ten days before official confirmation.

Conclusions

Real operational limits remain: an average 34.1-hour alert-validation delay and 48-72+ hour rural investigation delays, with lab confirmation still taking four to five days — the AI layer speeds up detection, but the response chain around it is a bottleneck.

Implementation
Indicative cost
Low (< €50k) — Not disclosed; a three-district pilot deployment.
Time to results
Short (< 1 year) — Piloted April-October 2025 (7 months).
Staffing & skills
Ministry of Health Vaccine Surveillance and Response Division, AI4PEP research network, Community reporters

Conditions for success

  • Community-level chatbot access lowering the barrier to reporting suspected cases
  • Cross-border district selection covering high-risk zones (Mauritania border, mining border region, urban Dakar)

Common failure modes

  • 34.1-hour average alert-validation delay and multi-day rural investigation delays remain bottlenecks
  • Laboratory confirmation still takes four to five days

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful