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

Senegal faces recurring outbreaks of zoonotic diseases such as Rift Valley fever, which often spread silently between animals, humans and the environment before official surveillance systems detect them, delaying response.
The Ministry of Health and Social Action's Vaccine Surveillance and Response Division, working with the AI4PEP research network under Dr. Boly Diop and Prof. Sylvain Landry B. Faye, built the Surveillance-Santé-Sénégal (3S) One Health platform. It has three linked modules: Taggàt, an AI chatbot that helps community members identify possible zoonotic symptoms; Jottali, a community-reporting alert system; and Gëstu, an interactive monitoring dashboard for health officials. It was piloted in the Podor (Saint-Louis, bordering Mauritania), Kédougou (mining border region) and Pikine (urban Dakar) health districts.
A peer-reviewed evaluation published in Frontiers in Tropical Diseases (2026), covering the pilot's April-October 2025 operation, found 330 alerts issued, of which 144 (43.6%) were validated; validated human alerts made up about 75% of these, with the remainder animal, plant and environmental signals. Digital reporting cut average detection time by 39%, from 4.6 days under conventional surveillance to 2.8 days, and in some cases flagged signals up to ten days before official confirmation.
The same evaluation reported real operational limits: an average alert-validation delay of 34.1 hours, and rural investigation delays of 48 to over 72 hours, with laboratory confirmation still taking four to five days — meaning the AI layer speeds up detection, but the response chain around it remains a bottleneck.

Read the full analysis: https://www.frontiersin.org/journals/tropical-diseases/articles/10.3389/fitd.2026.1778438/full