Cervical Cell Classification using Learning Vector Quantization (LVQ) Based on Shape and Statistical Features

Erlinda Metta Dewi,Retna Apsari,Endah Purwanti

doi:10.3991/ijoe.v15i02.9796

Erlinda Metta Dewi, Retna Apsari + Show 1 more

Open Access

https://doi.org/10.3991/ijoe.v15i02.9796

Copy DOI

Abstract

This research was conducted to design a system that is able to classify cervical cells into two classes, namely normal cells or abnormal cells. We use digital images of single cervical as research materials and Learning Vector Quantization (LVQ) as classification method. Prior to classification, the nucleus areas of single cervical cell images were segmented and features were extracted. The features used in this study are 7 kinds of which consist of 2 types of feature, namely shape features and statistical features. The shape features used are area, perimeter, shape factor, and roundness of the nucleus, while the statistical features of the grayscale image histogram used are mean, standard deviation, and entropy. LVQ optimal parameter values based on the highest accuracy of training data, are learning rate 0.1 and learning rate reduction 0.5. The highest accuracy of system obtained from 45 testing data is 93.33%.

Highlights

Cervical cancer ranks as the second most common type of cancer affecting women around the world and 80% of death cases come from developing countries [1]
Before it can be classified into normal cells or abnormal cells, a digital image of single cervical cell must pass through a series of processes in advance to obtain the results of shape and statistical features values that will be used for classification’s input using Learning Vector Quantization (LVQ) method
These processes begin with a series of grayscaling, digital image processing, and nucleus segmentation to obtain the values of shape features

Summary

Introduction

Cervical cancer ranks as the second most common type of cancer affecting women around the world and 80% of death cases come from developing countries [1]. One method for the early detection of cervical cancer is by using pap smear test. This test is a method of visual inspection of the cervical cells under a microscope to detect any abnormal cells that could potentially develop into cancer cells. Diagnosis of pap smear test results have high false rate ranged between 5% -50% [3]. Because the inspection is done manually, the examination of cervical cells susceptible to misinterpretation. Attempts to classify a single cervical cell have been conducted by several researchers. Norup extract 20 kinds of shape features from a single cervical cell that iJOE ‒ Vol 15, No 2, 2019

Methods

Findings

Discussion

Conclusion