Return
Diagnostic performance evaluation of the Siemens Healthineers Atellica HEMA 580 hematology analyzer for abnormal cell detection and optimization of immature myeloid cell cut-offs
D
C
DOI:10.1515/cclm-2026-0083.png)
Abstract
En 中文
Objectives We evaluated the diagnostic performance of the Atellica HEMA 580 hematology analyzer for complete blood count (CBC) parameters compared with the ADVIA 2120i and assessed its capability to flag abnormal cells, and determined optimal cut-offs for immature myeloid cell detection. Methods A total of 400 K(2)EDTA blood samples were analyzed. Peripheral blood smears were independently reviewed by two clinical pathologists, each performing a 200-cell differential count. Analytical correlation, regression, Bland-Altman, and receiver operating characteristic (ROC) curve analyses were used to assess performance. Optimal cut-offs for immature myeloid cell (IMC) detection were determined using ROC analysis and the Youden index. Results Manual microscopy identified 23 samples with blasts (>= 0.5 %), 177 with immature myeloid cells, and 61 with nucleated red blood cells (NRBCs). ROC analysis showed good accuracy for IMC detection using the IMG (immature granulocytes) parameter (AUC=0.81). An IMG cut-off >0.8 provided the best balance between sensitivity (40.7 %) and specificity (96.9 %), while large immature cells (LICs) >0.9 yielded higher sensitivity (60.5 %) at the expense of specificity (82.5 %). The Atellica HEMA 580 also demonstrated superior NRBC detection compared to ADVIA 2120i. Conclusions The Atellica HEMA 580 provided reliable performance in detecting immature myeloid cells and NRBCs. The IMG cut-off >0.8, determined through ROC optimization, supports its implementation for enhancing laboratory workflow efficiency while maintaining diagnostic accuracy. These findings support the clinical implementation of the Atellica HEMA 580 for routine hematology laboratories, particularly for enhanced detection of clinically significant immature myeloid populations.
Keywords:
immature myeloid cells
nucleated red blood cells (NRBCs)
hematology analyzer performance
receiver operating characteristic (ROC) analysis
immature granulocytes cut-off
automated cell flagging
Journal
C
IF:
3.7
Papers:
7.6K
Citations:
1.1W
Organization
No organization information available
