Journal
JOURNAL OF CLINICAL LABORATORY ANALYSIS
Volume 31, Issue 2, Pages -Publisher
WILEY
DOI: 10.1002/jcla.22024
Keywords
cytology; hematology; image processing; leukemia; pathology
Categories
Funding
- Ministry of Economy and Competitiveness of Spain [DPI 2015-64493-R]
- Universitat Politecnica de Catalunya
Ask authors/readers for more resources
BackgroundAutomated peripheral blood (PB) image analyzers usually underestimate the total number of blast cells, mixing them up with reactive or normal lymphocytes. Therefore, they are not able to discriminate between myeloid or lymphoid blast cell lineages. The objective of the proposed work is to achieve automatic discrimination of reactive lymphoid cells (RLC), lymphoid and myeloid blast cells and to obtain their morphologic patterns through feature analysis. MethodsIn the training stage, a set of 696 blood cell images was selected in 32 patients (myeloid acute leukemia, lymphoid precursor neoplasms and viral or other infections). For classification, we used support vector machines, testing different combinations of feature categories and feature selection techniques. Further, a validation was implemented using the selected features over 220 images from 15 new patients (five corresponding to each category). ResultsBest discrimination accuracy in the training was obtained with feature selection from the whole feature set (90.1%). We selected 60 features, showing significant differences (P<0.001) in the mean values of the different cell groups. Nucleus-cytoplasm ratio was the most important feature for the cell classification, and color-texture features from the cytoplasm were also important. In the validation stage, the overall classification accuracy and the true-positive rates for RLC, myeloid and lymphoid blast cells were 80%, 85%, 82% and 74%, respectively. ConclusionThe methodology appears to be able to recognize reactive lymphocytes well, especially between reactive lymphocytes and lymphoblasts.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available