Related Experiment Video
Updated: Jul 17, 2025
Multiplexed Fluorescent Immunohistochemical Staining, Imaging, and Analysis in Histological Samples of Lymphoma
Published on: January 9, 2019
Artificial intelligence and the blood film: Performance of the MC-80 digital morphology analyzer in samples with
Gina Zini1,2, Francesca Mancini3, Elena Rossi1,2
1Catholic University of Sacred Heart, Hematology Service, Rome, Italy.
Introduction:
Implementing artificial intelligence-based instruments in hematology laboratories requires evidence of efficiency in classifying pathological cells. In two-Universities, we assessed the performance of the Mindray® MC-80 for hematology patients with frequent leukemic and dysplastic cells.
Methods:
The Mindray MC-80® locates and pre-classifies cells in blood films. In a two-university study, four films were prepared from 591 samples, two each for the analyser MC-80 and the microscope reference method, using reagents from two different manufacturers. We used Microsoft Excel® statistics for imprecision and distributional inaccuracy and a matrix table model (H20-A2 CLSI standard) for sensitivity, specificity and predictive value for atypical cells.
Results:
The results indicate minimal within-run imprecision (ICSH method) and good intra-method consistency even on duplicate analysis of 413 samples with a high incidence of hematological abnormalities (r = 0.942 or more, except for basophils, r = 0.841, and reactive lymphocytes, r = 0.847). Distributional inaccuracy was also very low compared to the microscope reference, with a pass rate higher than 80% for pathological cells (except 75.1% for reactive lymphocytes). The primary causes of discrepancy were bizarre shapes of dysplastic neutrophils and inconsistent nomenclature for lymphoma cells. Sensitivity for critical samples containing cells typically absent in circulating blood (immature or malignant) was 98.8% for immature granulocytes, 83.8% for all types of neoplastic cells, 93.6% for reactive lymphocytes and 97.5% for nucleated red blood cells. The negative predictive values of MC-80 were 98.8% for immature granulocytes, 88.4% for the different types of neoplastic cells, 97.8% for reactive lymphocytes, and 96.9% for nucleated red blood cells.
Conclusion:
Our study highlights the outstanding diagnostic performance of this artificial intelligence-based blood film analyzer for hematology patients with circulating abnormal cells. We appreciated the morphological harmonization of cells observed on the screen and those seen in the microscope.
