Automation of detection of cervical cancer using convolutional neural networks

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Classification of digital cervical images acquired during visual inspection with acetic acid (VIA) is an important step in automated image-based cervical cancer detection. Many algorithms have been developed for classification of cervical images based on extracting mathematical features and classifying these images. Deciding the suitability of a feature and learning the algorithm is a complex task. On the other hand, convolutional neural networks (CNNs) self-learn most suitable hierarchical features from the raw input image. In this paper, we demonstrate the feasibility of using a shallow layer CNN for classification of image patches of cervical images as cancerous or not cancerous. We used cervix images acquired after the application of 3%–5% acetic acid using an Android device in 102 women. Of these, 42 cervix images belonged in the VIA-positive category (pathologic) and 60 in the VIA-negative category (healthy controls). A total of 275 image patches of 15 × 15 pixels were manually extracted from VIA-positive areas, and we considered these patches as positive examples. Similarly, 409 image patches were extracted from VIA-negative areas and were labeled as VIA negative. These image patches were classified using a shallow layer CNN composed of a layer each of convolutional, rectified linear unit, pooling, and two fully connected layers. A classification accuracy of 100% is achieved using shallow CNN.

Original languageEnglish
Pages (from-to)135-145
Number of pages11
JournalCritical Reviews in Biomedical Engineering
Volume46
Issue number2
DOIs
Publication statusPublished - 01-01-2018

Fingerprint

Acetic acid
Automation
Neural networks
Inspection
Pixels

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering

Cite this

@article{f8965024b1184a04ac96004eb1424bb4,
title = "Automation of detection of cervical cancer using convolutional neural networks",
abstract = "Classification of digital cervical images acquired during visual inspection with acetic acid (VIA) is an important step in automated image-based cervical cancer detection. Many algorithms have been developed for classification of cervical images based on extracting mathematical features and classifying these images. Deciding the suitability of a feature and learning the algorithm is a complex task. On the other hand, convolutional neural networks (CNNs) self-learn most suitable hierarchical features from the raw input image. In this paper, we demonstrate the feasibility of using a shallow layer CNN for classification of image patches of cervical images as cancerous or not cancerous. We used cervix images acquired after the application of 3{\%}–5{\%} acetic acid using an Android device in 102 women. Of these, 42 cervix images belonged in the VIA-positive category (pathologic) and 60 in the VIA-negative category (healthy controls). A total of 275 image patches of 15 × 15 pixels were manually extracted from VIA-positive areas, and we considered these patches as positive examples. Similarly, 409 image patches were extracted from VIA-negative areas and were labeled as VIA negative. These image patches were classified using a shallow layer CNN composed of a layer each of convolutional, rectified linear unit, pooling, and two fully connected layers. A classification accuracy of 100{\%} is achieved using shallow CNN.",
author = "Vidya Kudva and Keerthana Prasad and Shyamala Guruvare",
year = "2018",
month = "1",
day = "1",
doi = "10.1615/CritRevBiomedEng.2018026019",
language = "English",
volume = "46",
pages = "135--145",
journal = "Critical Reviews in Biomedical Engineering",
issn = "0278-940X",
publisher = "Begell House Inc.",
number = "2",

}

Automation of detection of cervical cancer using convolutional neural networks. / Kudva, Vidya; Prasad, Keerthana; Guruvare, Shyamala.

In: Critical Reviews in Biomedical Engineering, Vol. 46, No. 2, 01.01.2018, p. 135-145.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Automation of detection of cervical cancer using convolutional neural networks

AU - Kudva, Vidya

AU - Prasad, Keerthana

AU - Guruvare, Shyamala

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Classification of digital cervical images acquired during visual inspection with acetic acid (VIA) is an important step in automated image-based cervical cancer detection. Many algorithms have been developed for classification of cervical images based on extracting mathematical features and classifying these images. Deciding the suitability of a feature and learning the algorithm is a complex task. On the other hand, convolutional neural networks (CNNs) self-learn most suitable hierarchical features from the raw input image. In this paper, we demonstrate the feasibility of using a shallow layer CNN for classification of image patches of cervical images as cancerous or not cancerous. We used cervix images acquired after the application of 3%–5% acetic acid using an Android device in 102 women. Of these, 42 cervix images belonged in the VIA-positive category (pathologic) and 60 in the VIA-negative category (healthy controls). A total of 275 image patches of 15 × 15 pixels were manually extracted from VIA-positive areas, and we considered these patches as positive examples. Similarly, 409 image patches were extracted from VIA-negative areas and were labeled as VIA negative. These image patches were classified using a shallow layer CNN composed of a layer each of convolutional, rectified linear unit, pooling, and two fully connected layers. A classification accuracy of 100% is achieved using shallow CNN.

AB - Classification of digital cervical images acquired during visual inspection with acetic acid (VIA) is an important step in automated image-based cervical cancer detection. Many algorithms have been developed for classification of cervical images based on extracting mathematical features and classifying these images. Deciding the suitability of a feature and learning the algorithm is a complex task. On the other hand, convolutional neural networks (CNNs) self-learn most suitable hierarchical features from the raw input image. In this paper, we demonstrate the feasibility of using a shallow layer CNN for classification of image patches of cervical images as cancerous or not cancerous. We used cervix images acquired after the application of 3%–5% acetic acid using an Android device in 102 women. Of these, 42 cervix images belonged in the VIA-positive category (pathologic) and 60 in the VIA-negative category (healthy controls). A total of 275 image patches of 15 × 15 pixels were manually extracted from VIA-positive areas, and we considered these patches as positive examples. Similarly, 409 image patches were extracted from VIA-negative areas and were labeled as VIA negative. These image patches were classified using a shallow layer CNN composed of a layer each of convolutional, rectified linear unit, pooling, and two fully connected layers. A classification accuracy of 100% is achieved using shallow CNN.

UR - http://www.scopus.com/inward/record.url?scp=85055355401&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055355401&partnerID=8YFLogxK

U2 - 10.1615/CritRevBiomedEng.2018026019

DO - 10.1615/CritRevBiomedEng.2018026019

M3 - Article

C2 - 30055530

AN - SCOPUS:85055355401

VL - 46

SP - 135

EP - 145

JO - Critical Reviews in Biomedical Engineering

JF - Critical Reviews in Biomedical Engineering

SN - 0278-940X

IS - 2

ER -