ARTIFICIAL INTELLIGENCE IN HEMATOLOGY, FLAG, COUNT, CONFIRM: AUTOMATED CBC METRICS FOR HIGH PERFORMANCE MALARIA SCREENING

Abstract

Automated hematology analyzers have shown promise as adjuncts for malaria detection through generation of parasite‑related flags, service parameters and abnormal scattergrams. This study assessed the flagging capacity of the Sysmex XN‑1000i, evaluated prbc–WDF# and prbc–WNR# service parameters, and described characteristic scattergram abnormalities in malaria‑suspected samples.

Objective: To evaluate the diagnostic performance of the Sysmex XN-1000i in detecting malaria using the prbc flag, PRBC–WDF#, PRBC–WNR#, and scattergram abnormalities, with PCR as the reference standard.

Methods: A descriptive cross-sectional study was conducted from May to October 2020 at a tertiary hospital in Pakistan. Febrile patients suspected of malaria (n=87) underwent complete blood counts on the Sysmex XN-1000i. The presence of a prbc flag, PRBC WDF#, PRBC–WNR#, and abnormal WDF scattergrams were recorded. Real-time PCR served as the gold standard. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated.

Results: PRBC–WDF# and PRBC–WNR# both achieved 100% sensitivity, specificity, PPV, NPV, and accuracy, correctly identifying all PCR-confirmed cases and negatives. The prbc flag and scattergram abnormalities each demonstrated 97.6% sensitivity, 100% specificity, 100% PPV, and 33.3% NPV, missing two PCR-positive cases each.

Conclusion: The Sysmex XN-1000i’s PRBC–WDF# and PRBC–WNR# parameters provide highly reliable malaria detection, even at low parasite levels. Combining these parameters with prbc flagging and scattergram review can enhance early case identification in routine CBC workflows, offering a practical screening tool in endemic and resource-limited settings.