mHealth One, the new version of mHealth developed by TICANALYSE, enables the use of artificial intelligence to improve the provision of community health services. Côte d’Ivoire has rolled it out in the Gontougo region, with support from UNICEF and under the leadership of the Ministry of Health, Public Hygiene and Universal Health Cover. Since 5 October 2026, community health workers in the region have been collecting their data on the platform.
A three-tier AI system
mHealth One’s artificial intelligence is deployed across all three levels of the healthcare system, and each level receives the support it needs to make decisions.
At an operational level, the community health worker receives support during consultations. Based on the symptoms recorded, mHealth Care provides treatment recommendations and guidance on how to proceed, in line with national protocols. The community health worker knows when to treat, advise or refer a patient to the health centre, even without an internet connection.
At a tactical level, nurses and district teams monitor activity in mHealth Monitor. The AI identifies inconsistent data, unusual variations and agents who are no longer submitting data. It generates automated reports and issues alerts, enabling supervisors to focus on situations that require intervention.
At a strategic level, decision-makers query the data in everyday French using the mHealth AI Assistant. A question such as ‘How many children under the age of five have been treated for malaria this month in Tanda?’ is answered in the form of a graph or table, without the need for a technical report.
A phased roll-out
The roll-out covers the health districts of Bondoukou, Koun-Fao, Sandégué, Tanda and Transua. The mHealth experts attended a refresher course on 10 and 11 September, and then trained end-users in the field from 20 September to 3 October. The DSCPS, the DISD and the DIS were involved at every stage.
Building on a tried and tested foundation
mHealth has been in use in Côte d’Ivoire for several years. An evaluation carried out by the INSP and UNICEF amongst 280 users found an average score of 4.62 out of 5 and an acceptability rate of 100 per cent.
FR 
EN