Publication
MAL

Diagnostic performance of the miLab MAL AI-driven automated digital microscopy system versus standard microscopy for Plasmodium spp. detection

This prospective evaluation compared miLab MAL with conventional diagnostic methods using 162 blood samples collected over six months. The system showed promising Plasmodium detection performance, with limitations in species identification and parasite quantification.

Journal of Clinical Microbiology (JCM)

|

October 30, 2026

|

Lorra Monpierre, et al.

Abstract

This study evaluated miLab™ MAL, an AI-based automated digital microscopy platform, for detecting Plasmodium species in a clinical laboratory setting in France. In 162 clinical samples, the system achieved a sensitivity of 90.4% and a specificity of 98.7% compared with LAMP and/or thick blood smears, and a sensitivity of 97.4% and a specificity of 97.7% compared with expert thin blood smear microscopy (Cohen’s κ = 0.95). While the platform effectively detected Plasmodium spp., it faced challenges in species-level differentiation and tended to underestimate parasitemia.

SensitivitySpecificityCohen’s κ
LAMP and/or Thick Blood Smear90.4%98.7%0.89
Thin Blood Smear Microscopy97.4%97.7%0.95

Table 1: Diagnostic Performance of miLab™ MAL vs. Conventional Methods

Key Highlights

  • 97.4% sensitivity · 97.7% specificity · Cohen’s κ 0.95 vs. expert thin smear microscopy

Keywords

malaria
Plasmodium
France
AI diagnostics
digital microscopy

Related Publications