EasyScan GO reaches 91.1% sensitivity for malaria at 11 sites

AI system

System: EasyScan GO digital malaria microscopy device with convolutional-neural-network image analysis of thick blood films
Model: A convolutional neural network that analyzes thick blood films.
Release: Version 1.15.17b.
Application developer: Intellectual Ventures' Global Good Fund (software and original hardware); Motic (production prototype device, software licensed to Motic)
Role: The network analyzed microscope images for malaria parasites.
Configuration: The deployment used software version 1.15.17b.
Inputs and outputs: Not assessed
Autonomy: Not assessed
Workflow: EasyScan GO devices donated and set up with manufacturer training at each site. Giemsa-stained thick blood films read in parallel by two expert microscopists as reference. STARD-reported multi-center observational design across 11 countries (ClinicalTrials.gov NCT03512678).
Human oversight: Not assessed
Relation to AI: The machine-learning device produced the readings compared against expert microscopy, so the measured accuracy is the AI's own performance. The evidence is a single peer-reviewed study funded by the maker (Global Good Fund) with maker-employed co-authors who provided inputs to study design and training. The authors state the study did not evaluate performance under routine field operation and that all parameters were significantly affected by slide quality. No independent or outcome-level evidence was found.
AI contribution: The machine-learning device produced the readings compared against expert microscopy, so the measured accuracy is the AI's own performance. The evidence is a single peer-reviewed study funded by the maker (Global Good Fund) with maker-employed co-authors who provided inputs to study design and training. The authors state the study did not evaluate performance under routine field operation and that all parameters were significantly affected by slide quality. No independent or outcome-level evidence was found.

Benefits

Outcome identifiers: Not assessed
Benefit: Diagnostic sensitivity was 91.1% (95% CI 88.9–92.7) and specificity 75.6% (73.1–78.0)
Category: Improved Task Accuracy
Status: Not assessed
Beneficiary groups: People. Organizations. Society.
Beneficiaries: Not assessed
Reach: Not assessed
Distribution and access: Not assessed
Duration: Not assessed
Expected?: Yes
Costs and harms: Not assessed

Measurement

Studies: Not assessed
Study design: Not assessed
Measurement: Not assessed
Setting and sample: Not assessed
Measures: Diagnostic sensitivity was 91.1% (95% CI 88.9–92.7) and specificity 75.6% (73.1–78.0)
Comparison: expert microscopy
Results: Result: diagnostic sensitivity of the EasyScan Go device was 91.1% (95% CI 88.9–92.7) and specificity was 75.6% (95% CI 73.1–78.0).
Additional value from AI: Not assessed
Limitations: Not assessed
Denominators: Not assessed
Materials for replication: Not assessed

Money

Kind of value: Not assessed
Amount: Not assessed
Currency and prices: Not assessed
Period and unit: Not assessed
Realization: Not assessed
Recipients and payers: Not assessed
Gross or net: Not assessed
Included costs: Not assessed
Comparison: Not assessed
Method: Not assessed
Return and payback: Not assessed

People and places

Participants: Developer: Intellectual Ventures' Global Good Fund (software and original hardware); Motic (production prototype device, software licensed to Motic). Operator: Research sites in 11 countries under a multi-center study with University of Oxford / Mahidol-Oxford Tropical Medicine Research Unit-affiliated investigators (research use, not routine service).
Jurisdictions: Benefit: 11 sites in Burkina Faso, Kenya, Republic of the Congo, Senegal, South Africa, Uganda, Bangladesh, Cambodia, Nepal, Thailand and Brazil. Developer: United States.
Ownership and control: Not assessed
Dependencies: Not assessed
Cross-border connections: Not assessed
Geopolitical implications: Not assessed

Follow-up

Conditions for replication: Good slide quality, which the authors found significantly affected all performance parameters. Parasitemia above about 200 parasites/uL, below which sensitivity fell to 57%. Thick-film-only software (the tested version did not separate non-falciparum species).
Replication attempts: 2 follow-up entries. Benefit recurred: 2.
Later findings: No later evaluation, approval or deployment is reported in the article.
Availability: Not assessed
Institutional decision: Not assessed
Grounds: Not assessed
Action and result: Not assessed
Success criterion: Not assessed

Evidence

Source numbers: AIBD-0352-source-0001–0004
Citations: Numbered references below.
Language: Not assessed
Source relationships: Developer
Original evidence: Not assessed
Access and versions: Not assessed
Supported findings: Not assessed
Evidence links: Not assessed
Evidence strength: Not assessed
Missing evidence: Not assessed

Grades

Assessed benefits: People. Organizations. Society.
Depth: Not assessed
Reasons: People: No patient-management or clinical outcome was measured. Organizations: Laboratory-level accuracy was measured under study conditions, not routine field operation. Society: No deployment at program scale is documented.
Observed or intended: Not assessed
Rubric: Not assessed
Assessor and date: Not assessed
Alternative assessments: Not assessed
Fill: The makers' claim

Entry details

Kind: Not assessed
Related entries: Not assessed
Field: Malaria diagnostics / digital microscopy. Global-health diagnostic research.
Purpose: Automated detection, species identification and quantification of malaria parasites in blood films to replace or support expert microscopy. The evaluated benefit (diagnostic accuracy) is that purpose.
Dates: Date basis unassessed: 2022 (precision unassessed). Updated: 21 October 2026.
Stage: Not assessed
Version: SHA-256: 165d995b2ec7…

References

Other identifiers: Not assessed
Category mappings: Not assessed
Brief: Not assessed
Revisions: 5 revisions. Latest: 21 October 2026.

Sources

Peer-reviewed study Developer

Das, D., Vongpromek, R., Assawariyathipat, T., et al. (2022). Field evaluation of the diagnostic performance of EasyScan GO: a digital malaria microscopy device based on machine-learning. Malaria Journal 21, 122. doi:10.1186/s12936-022-04146-1.
Peer-reviewed study Developer

Das, D., et al. (2022). Same article, full text (methods and limitations). Europe PMC PMC9004086.
Peer-reviewed study Developer

Das, D., Vongpromek, R., Assawariyathipat, T., et al. (2022). Field evaluation of the diagnostic performance of EasyScan GO: a digital malaria microscopy device based on machine-learning. Malaria Journal 21, 122. doi:10.1186/s12936-022-04146-1. Full text as served by the Europe PMC REST API (PMC9004086).