White Blood Cell Detection and Classification in Blood Smear Images Using a One-Stage Object Detector and Similarity Learning

Fierro-Radilla, Atoany Nazareth; Bolaños Cacho, Monica Larre; Pérez-Daniel, Karina Ruby; Arredondo Valle, Armando; López Figueroa, Carlos Alberto; Benitez-Garcia, Gibran

doi:10.1007/978-3-031-19493-1_27

White Blood Cell Detection and Classification in Blood Smear Images Using a One-Stage Object Detector and Similarity Learning

Journal

Advances in Computational Intelligence

Lecture Notes in Computer Science

ISSN

0302-9743

1611-3349

Date Issued

2022

Author(s)

Fierro-Radilla, Atoany Nazareth

Bolaños Cacho, Monica Larre

Pérez-Daniel, Karina Ruby

Facultad de Ingeniería - CampCM

Arredondo Valle, Armando

López Figueroa, Carlos Alberto

Facultad de Ingeniería - CampCM

Benitez-Garcia, Gibran

Type

text::book::book part

DOI

10.1007/978-3-031-19493-1_27

URL

https://scripta.up.edu.mx/handle/20.500.12552/4189

Abstract

White blood cells are a fundamental part of the immune system which protect human body against infections and diseases. The complete blood count is a routine analysis that provides doctors information about the patients. Monitoring the immune system allows doctor to select preventive treatments against several diseases. The complete blood count relies in a rigorous observation of a blood sample through a microscope; the accuracy of the result depends on the expertise and time of the analyst. In this paper, a novel vision-based method using convolutional neural networks for white blood cell detection and classification is presented. The results show the proposed method is robust against the huge number of easy negatives in object detection, as well, the high inter-class similarity among images can be reduced for a better white blood cell classification. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.