evidoria

← Back to browse

Good practice Imported

VectorCam — Uganda's AI smartphone tool for malaria vector surveillance

Uganda · Kampala · See the Uganda profile · See the Kampala profile

Evidence: Descriptive / self-reported Top 66% 53/100 · Ask Evidence Copilot about this practice

Developed by Makerere University and Johns Hopkins with Uganda's Ministry of Health, VectorCam uses AI computer vision to identify mosquito species at 90% accuracy via smartphone, enabling real-time malaria vector surveillance across 22 districts in resource-limited settings.

90 %
Species identification accuracy (2024 field validation)
12 districts
Districts deployed (2024/2025)
22 districts
Target districts (2025 target)
20 seconds
Classification time per sample
VectorCam — Uganda's AI smartphone tool for malaria vector surveillance

Details

Maturity
Scaling
Promoter
Makerere University School of Public Health / Johns Hopkins Bloomberg School of Public Health / Uganda Ministry of Health
Period
2022–present
Keywords
health, malaria, vector surveillance, public health

Context

VectorCam is a smartphone-based AI application developed by Makerere University School of Public Health, the Johns Hopkins Bloomberg School of Public Health Center for Global Digital Health Innovation, and Uganda's Ministry of Health. It uses computer vision and deep learning, via a smartphone-attached microscope adapter, to identify mosquito species, sex, and blood-meal status directly in the field, without needing centralized laboratory infrastructure. The tool is intended to support malaria vector surveillance and targeted vector control in resource-limited districts where trained entomologists are scarce.

Objectives

The project aims to enable real-time malaria vector surveillance without centralized laboratories, improve targeted vector control through accurate species identification, and extend surveillance capacity to resource-limited settings lacking trained entomologists.

Activities

After a pilot phase from November 2022 to April 2024, the tool was deployed across Ugandan districts with field validation of species-identification accuracy and operational research monitoring embedded in a subset of districts. As of 2025, Makerere University, Johns Hopkins and the Ministry of Health are scaling the project from 12 to a target of 22 districts, with research embedded in 12 of them, and results posted to Uganda's DHIS system for district and national access.

Results

Field validation reported 90% accuracy for mosquito species identification, with classification completed in about 20 seconds per sample. The tool is deployed in 12 districts as of 2024/2025, with rollout targeted to reach 22 districts.

Conclusions

The evidence to date is limited to technical-performance and deployment-level indicators; no randomized controlled trial or formal impact evaluation has been published. Evidoria's own source assessment cautions that causal claims about population-level malaria outcomes should be treated carefully pending formal evaluation, which is planned as part of the Ministry of Health-led scale-up.

Implementation
Indicative cost
Medium (€50k–€500k) — No public budget figures were found; the project is funded by the Bill & Melinda Gates Foundation. Hardware (smartphone + low-cost microscope adapter) is inexpensive per unit, but multi-district deployment, DHIS integration and embedded operational research across up to 22 districts imply a moderate ongoing programme cost — estimated as medium in the absence of published figures.
Time to results
Long (> 3 years) — Pilot ran November 2022-April 2024; scale-up to 22 districts (with research embedded in 12) was announced in August/September 2025 and is ongoing, indicating a multi-year (long) implementation horizon.
Staffing & skills
Makerere University School of Public Health (research/field validation), Johns Hopkins Bloomberg School of Public Health, Center for Global Digital Health Innovation (technical development), Uganda Ministry of Health (rollout and DHIS integration)

Conditions for success

  • Smartphone + microscope-adapter hardware availability in district health offices
  • Integration with Uganda's DHIS reporting system for data uptake by district/national health authorities
  • Embedded operational research capacity to validate performance as it scales
  • Sustained donor funding (Bill & Melinda Gates Foundation) through the pilot-to-scale transition

Common failure modes

  • Species-identification accuracy claims (90%) are not yet paired with a published, peer-reviewed evaluation methodology
  • Scale-up to 22 districts depends on continued donor and Ministry of Health commitment
  • No causal evidence yet linking surveillance data to actual reductions in malaria transmission or vector control effectiveness

Where it fits

Governance type
national ministry of health with university/academic partners
Scale
sub-national, scaling from 12 to 22 districts
Income level
low-income (Uganda)

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