A Novel Deep Learning Approach in Haematology for Classification of Leucocytes.

摘要: This paper presents a comparison between two different Computer Aided Diagnosis systems for classification of five types leucocytes located in the tail Peripheral Blood Smears: Lymphocytes, Monocytes, Neutrophils, Basophils and Eosinophils. In particular, we have evaluated compared performance previous feature-based Back Propagation Neural Network classifier with novel classifiers both based on Deep Learning using Convolutional Networks introduced this study. All are built considering same dataset images acquired The experimental results, reported terms accuracy, sensitivity, specificity precision, show that strategies could be discussed from clinical technical point view.