Deep Convolution Neural Network for Malignancy Detection and Classification in Microscopic Uterine Cervix Cell Images

Shanthi P B, Faraz Faruqi, Hareesha K S, Ranjini Kudva

Research output: Contribution to journalArticle

Abstract

OBJECTIVE: Automated Pap smear cervical screening is one of the most effective imaging based cancer detection
tools used for categorizing cervical cell images as normal and abnormal. Traditional classification methods depend on
hand-engineered features and show limitations in large, diverse datasets. Effective feature extraction requires an efficient
image preprocessing and segmentation, which remains prominent challenge in the field of Pathology. In this paper, a
deep learning concept is used for cell image classification in large datasets. METHODS: This relatively proposed novel
method, combines abstract and complicated representations of data acquired in a hierarchical architecture. Convolution
Neural Network (CNN) learns meaningful kernels that simulate the extraction of visual features such as edges, size,
shape and colors in image classification. A deep prediction model is built using such a CNN network to classify the
various grades of cancer: normal, mild, moderate, severe and carcinoma. It is an effective computational model which
uses multiple processing layers to learn complex features. A large dataset is prepared for this study by systematically
augmenting the images in Herlev dataset. RESULT: Among the three sets considered for the study, the first set of single
cell enhanced original images achieved an accuracy of 94.1% for 5 class, 96.2% for 4 class, 94.8% for 3 class and
95.7% for 2 class problems. The second set includes contour extracted images showed an accuracy of 92.14%, 92.9%,
94.7% and 89.9% for 5, 4, 3 and 2 class problems. The third set of binary images showed 85.07% for 5 class, 84%
for 4 class, 92.07% for 3 class and highest accuracy of 99.97% for 2 class problems. CONCLUSION: The experimental
results of the proposed model showed an effective classification of different grades of cancer in cervical cell images,
exhibiting the extensive potential of deep learning in Pap smear cell image classification.

Original languageEnglish
Pages (from-to)3447-3456
Number of pages10
JournalAsian Pacific journal of cancer prevention : APJCP
Volume20
Issue number11
DOIs
Publication statusPublished - 01-11-2019

Fingerprint

Cervix Uteri
Papanicolaou Test
Neoplasms
Learning
Uterine Cervical Neoplasms
Hand
Color
Pathology
Carcinoma
Datasets

All Science Journal Classification (ASJC) codes

  • Epidemiology
  • Oncology
  • Public Health, Environmental and Occupational Health
  • Cancer Research

Cite this

@article{bec66e9b4fb64c3286b8937a3d9f3a5f,
title = "Deep Convolution Neural Network for Malignancy Detection and Classification in Microscopic Uterine Cervix Cell Images",
abstract = "OBJECTIVE: Automated Pap smear cervical screening is one of the most effective imaging based cancer detection tools used for categorizing cervical cell images as normal and abnormal. Traditional classification methods depend on hand-engineered features and show limitations in large, diverse datasets. Effective feature extraction requires an efficient image preprocessing and segmentation, which remains prominent challenge in the field of Pathology. In this paper, a deep learning concept is used for cell image classification in large datasets. METHODS: This relatively proposed novel method, combines abstract and complicated representations of data acquired in a hierarchical architecture. Convolution Neural Network (CNN) learns meaningful kernels that simulate the extraction of visual features such as edges, size, shape and colors in image classification. A deep prediction model is built using such a CNN network to classify the various grades of cancer: normal, mild, moderate, severe and carcinoma. It is an effective computational model which uses multiple processing layers to learn complex features. A large dataset is prepared for this study by systematically augmenting the images in Herlev dataset. RESULT: Among the three sets considered for the study, the first set of single cell enhanced original images achieved an accuracy of 94.1{\%} for 5 class, 96.2{\%} for 4 class, 94.8{\%} for 3 class and 95.7{\%} for 2 class problems. The second set includes contour extracted images showed an accuracy of 92.14{\%}, 92.9{\%}, 94.7{\%} and 89.9{\%} for 5, 4, 3 and 2 class problems. The third set of binary images showed 85.07{\%} for 5 class, 84{\%} for 4 class, 92.07{\%} for 3 class and highest accuracy of 99.97{\%} for 2 class problems. CONCLUSION: The experimental results of the proposed model showed an effective classification of different grades of cancer in cervical cell images, exhibiting the extensive potential of deep learning in Pap smear cell image classification.",
author = "{P B}, Shanthi and Faraz Faruqi and {K S}, Hareesha and Ranjini Kudva",
year = "2019",
month = "11",
day = "1",
doi = "10.31557/APJCP.2019.20.11.3447",
language = "English",
volume = "20",
pages = "3447--3456",
journal = "Asian Pacific Journal of Cancer Prevention",
issn = "1513-7368",
publisher = "Asian Pacific Organization for Cancer Prevention",
number = "11",

}

TY - JOUR

T1 - Deep Convolution Neural Network for Malignancy Detection and Classification in Microscopic Uterine Cervix Cell Images

AU - P B, Shanthi

AU - Faruqi, Faraz

AU - K S, Hareesha

AU - Kudva, Ranjini

PY - 2019/11/1

Y1 - 2019/11/1

N2 - OBJECTIVE: Automated Pap smear cervical screening is one of the most effective imaging based cancer detection tools used for categorizing cervical cell images as normal and abnormal. Traditional classification methods depend on hand-engineered features and show limitations in large, diverse datasets. Effective feature extraction requires an efficient image preprocessing and segmentation, which remains prominent challenge in the field of Pathology. In this paper, a deep learning concept is used for cell image classification in large datasets. METHODS: This relatively proposed novel method, combines abstract and complicated representations of data acquired in a hierarchical architecture. Convolution Neural Network (CNN) learns meaningful kernels that simulate the extraction of visual features such as edges, size, shape and colors in image classification. A deep prediction model is built using such a CNN network to classify the various grades of cancer: normal, mild, moderate, severe and carcinoma. It is an effective computational model which uses multiple processing layers to learn complex features. A large dataset is prepared for this study by systematically augmenting the images in Herlev dataset. RESULT: Among the three sets considered for the study, the first set of single cell enhanced original images achieved an accuracy of 94.1% for 5 class, 96.2% for 4 class, 94.8% for 3 class and 95.7% for 2 class problems. The second set includes contour extracted images showed an accuracy of 92.14%, 92.9%, 94.7% and 89.9% for 5, 4, 3 and 2 class problems. The third set of binary images showed 85.07% for 5 class, 84% for 4 class, 92.07% for 3 class and highest accuracy of 99.97% for 2 class problems. CONCLUSION: The experimental results of the proposed model showed an effective classification of different grades of cancer in cervical cell images, exhibiting the extensive potential of deep learning in Pap smear cell image classification.

AB - OBJECTIVE: Automated Pap smear cervical screening is one of the most effective imaging based cancer detection tools used for categorizing cervical cell images as normal and abnormal. Traditional classification methods depend on hand-engineered features and show limitations in large, diverse datasets. Effective feature extraction requires an efficient image preprocessing and segmentation, which remains prominent challenge in the field of Pathology. In this paper, a deep learning concept is used for cell image classification in large datasets. METHODS: This relatively proposed novel method, combines abstract and complicated representations of data acquired in a hierarchical architecture. Convolution Neural Network (CNN) learns meaningful kernels that simulate the extraction of visual features such as edges, size, shape and colors in image classification. A deep prediction model is built using such a CNN network to classify the various grades of cancer: normal, mild, moderate, severe and carcinoma. It is an effective computational model which uses multiple processing layers to learn complex features. A large dataset is prepared for this study by systematically augmenting the images in Herlev dataset. RESULT: Among the three sets considered for the study, the first set of single cell enhanced original images achieved an accuracy of 94.1% for 5 class, 96.2% for 4 class, 94.8% for 3 class and 95.7% for 2 class problems. The second set includes contour extracted images showed an accuracy of 92.14%, 92.9%, 94.7% and 89.9% for 5, 4, 3 and 2 class problems. The third set of binary images showed 85.07% for 5 class, 84% for 4 class, 92.07% for 3 class and highest accuracy of 99.97% for 2 class problems. CONCLUSION: The experimental results of the proposed model showed an effective classification of different grades of cancer in cervical cell images, exhibiting the extensive potential of deep learning in Pap smear cell image classification.

UR - http://www.scopus.com/inward/record.url?scp=85075547012&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075547012&partnerID=8YFLogxK

U2 - 10.31557/APJCP.2019.20.11.3447

DO - 10.31557/APJCP.2019.20.11.3447

M3 - Article

C2 - 31759371

AN - SCOPUS:85075547012

VL - 20

SP - 3447

EP - 3456

JO - Asian Pacific Journal of Cancer Prevention

JF - Asian Pacific Journal of Cancer Prevention

SN - 1513-7368

IS - 11

ER -