Objective: To establish an easy method for performing an automated differential cell count on bronchoalveolar lavage (BAL) cytocentrifuge samples using morphometric parameters acquired by a digital image analyzer.
Study design: The study population comprised 20 satisfactory cytocentrifuge preparations routinely processed in our laboratory. Images of at least 40 fields of interest were digitized and stored. The images were analyzed to assess the planimetric parameters. These were used for cell identification. Basic morphometric parameters were acquired by an automated image analysis procedure. The results obtained by this method are compared to those of the manual counting procedure.
Results: Following sequential analysis, computerized identification corresponded, in > 97% of the cells, to the results of manual cell typing. After data analysis, < 1% of cells remained unidentifiable.
Conclusion: Although the number of specimens evaluated is not yet large enough to allow a definitive statement, the use of image analysis systems seems a promising approach in automated differential cell counting in BAL preparations.