Title: Searching for cell signatures in multidimensional feature spaces

Authors: Romuere Silva; Flávio Araújo; Mariana Rezende; Paulo Oliveira; Fátima Medeiros; Rodrigo Veras; Daniela Ushizima

Addresses: Department of Teleinformatics Engineering, Federal University of Ceará, Fortaleza, Ceará, Brazil; Information Systems, Federal University of Piauí, Picos, Piauí, Brazil; Berkeley Institute for Data Science, University of California – Berkeley, Berkeley, California, USA; Computational Research Division, Lawrence Berkeley National Laboratory, Berkeley, California, USA ' Department of Teleinformatics Engineering, Federal University of Ceará, Fortaleza, Ceará, Brazil; Information Systems, Federal University of Piauí, Picos, Piauí, Brazil; Berkeley Institute for Data Science, University of California – Berkeley, Berkeley, California, USA; Computational Research Division, Lawrence Berkeley National Laboratory, Berkeley, California, USA ' Department of Clinical Analysis, Federal University of Ouro Preto, Ouro Preto, Minas Gerais, Brazil ' Computing Department, Federal University of Ouro Preto, Ouro Preto, Minas Gerais, Brazil ' Department of Teleinformatics Engineering, Federal University of Ceará, Fortaleza, Ceará, Brazil ' Computing Department, Federal University of Piauí, Teresina, Piauí, Brazil ' Berkeley Institute for Data Science, University of California – Berkeley, Berkeley, California, USA; Computational Research Division, Lawrence Berkeley National Laboratory, Berkeley, California, USA

Abstract: Despite research on cervical cells since 1925, systems to automatically screen images from conventional Pap smear tests continue to be unavailable. One of the main challenges in deploying precise software tools is to validate cell signatures. In this paper, we introduce an analysis framework, CRIC-feat, that expedites the investigation of different image databases and respective descriptors, particularly applicable to Pap images. This paper provides a three-fold contribution: 1) we first review and discuss the main feature extraction protocols for cell description and implementations suitable for cervical cells; 2) we present a new application of Gray level run length (GLRLM) features to Pap images; 3) we evaluate 93 cell classification approaches, and provide a guideline for obtaining the most accurate description, based on two current public databases with digital images of real cells. Finally, we show that the nucleus information is preponderant in cell classification, particularly when considering the GLRLM feature set.

Keywords: medical image; feature extraction; cervical cells; quantitative microscopy; cell descriptors; classification.

DOI: 10.1504/IJBET.2021.116988

International Journal of Biomedical Engineering and Technology, 2021 Vol.36 No.3, pp.236 - 256

Received: 13 Jan 2018
Accepted: 01 May 2018
Published online: 11 Aug 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article