[ESCMID 2026] AI-Based Digital Microscopy for Malaria Diagnosis: Preliminary Validation and Cost-Effectiveness

AI-based digital microscopy offers high diagnostic accuracy and significant economic and operational advantages over conventional microscopy and RDTs in non-endemic settings.

ESCMID 2026

April 17, 2026

P. Pitzinger, et al.

[ESCMID 2026] AI-Based Digital Microscopy for Malaria Diagnosis: Preliminary Validation and Cost-Effectiveness

Unauthorized downloading or distribution of this poster is strictly prohibited.

Abstract

This study evaluated the real-life performance of an AI-based microscopy system (miLab MAL) against routine diagnostics (conventional microscopy and RDTs) for malaria in 134 blood samples from returning travelers in Berlin. The miLab MAL system demonstrated high diagnostic accuracy with 100% sensitivity and 97.2% specificity, showing no false negatives. In comparison, RDT sensitivity was lower at 85.2%. Operationally, the AI system provided results in a mean time of 16 minutes. Crucially, the mean operational diagnostic cost per sample for AI screening was €9.39, resulting in a 72% cost reduction compared to the routine €33.43. The conclusions support integrating AI-based digital microscopy into routine diagnostics due to its accuracy and clear economic and operational advantages.

	Result	Comparison / Notes
Sensitivity	100%	RDT: 85.2%
Specificity	97.2%	-
Accuracy	97.8%	-
Mean time to result	16 min	Faster than conventional microscopy + RDT
Mean cost per test	€9.39	72% reduction compared to conventional testing (€33.43)