Utilization of a Digital Automated Cell Morphology Analyzer Results for Determining Differential White Blood Cell Counts in a Turkish Pediatric Population.

Abstract

Background: Manual morphological analysis of peripheral blood smears (PBS) with light microscopy is an essential diagnostic and monitoring tool. Recently, automated morphology analyzers have been developed that can preclassify cells using artificial intelligence algorithms. This study aims to evaluate the preliminary leukocyte classification capabilities of the MC-80 digital morphology analyzer, a novel system, in pediatric patients and compare its performance with that of manual microscopy, the current gold standard.

Methods: This retrospective study was conducted at SBU Bursa Yüksek İhtisas Training and Research Hospital between September 5 and 29, 2022. Blood samples from 153 consecutive pediatric patients (age range: 0-18 years; median age: 3 years) undergoing simultaneous hemograms and PBS analyses were assessed using both the MC-80 digital morphology analyzer and manual microscopy.

Results: Spearman rank correlation coefficients indicated a high correlation for neutrophils (rho = 0.742; 95% CI: 0.661-0.807) and lymphocytes (rho = 0.745; 95% CI: 0.666-0.810) while the correlation for blast cells was significantly lower (rho = 0.079; 95% CI: -0.099-0.238). Concordance between the MC-80 and manual microscopy was minimal for monocytes (κ = 0.21; 95% CI: 0.11-0.29) and negligible for blast cells (κ = 0.08; 95% CI: 0.00-0.17).

Conclusions: The MC-80 digital morphology analyzer shows acceptable preliminary classification for neutrophils and lymphocytes; further development is required before it can be routinely implemented in laboratory workflows.