Hybrid Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening

Research output: Contribution to journalArticle

Abstract

Transfer learning using deep pre-trained convolutional neural networks is increasingly used to solve a large number of problems in the medical field. In spite of being trained using images with entirely different domain, these networks are flexible to adapt to solve a problem in a different domain too. Transfer learning involves fine-tuning a pre-trained network with optimal values of hyperparameters such as learning rate, batch size, and number of training epochs. The process of training the network identifies the relevant features for solving a specific problem. Adapting the pre-trained network to solve a different problem requires fine-tuning until relevant features are obtained. This is facilitated through the use of large number of filters present in the convolutional layers of pre-trained network. A very few features out of these features are useful for solving the problem in a different domain, while others are irrelevant, use of which may only reduce the efficacy of the network. However, by minimizing the number of filters required to solve the problem, the efficiency of the training the network can be improved. In this study, we consider identification of relevant filters using the pre-trained networks namely AlexNet and VGG-16 net to detect cervical cancer from cervix images. This paper presents a novel hybrid transfer learning technique, in which a CNN is built and trained from scratch, with initial weights of only those filters which were identified as relevant using AlexNet and VGG-16 net. This study used 2198 cervix images with 1090 belonging to negative class and 1108 to positive class. Our experiment using hybrid transfer learning achieved an accuracy of 91.46%.

Original languageEnglish
JournalJournal of Digital Imaging
DOIs
Publication statusAccepted/In press - 01-01-2019

Fingerprint

Early Detection of Cancer
Cervix Uteri
Uterine Cervical Neoplasms
Screening
Tuning
Neural networks
Experiments
Learning
Efficiency
Weights and Measures
Transfer (Psychology)
Deep learning

All Science Journal Classification (ASJC) codes

  • Radiological and Ultrasound Technology
  • Radiology Nuclear Medicine and imaging
  • Computer Science Applications

Cite this

@article{ff73c1d2b15a4334bf9c9f625de1de66,
title = "Hybrid Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening",
abstract = "Transfer learning using deep pre-trained convolutional neural networks is increasingly used to solve a large number of problems in the medical field. In spite of being trained using images with entirely different domain, these networks are flexible to adapt to solve a problem in a different domain too. Transfer learning involves fine-tuning a pre-trained network with optimal values of hyperparameters such as learning rate, batch size, and number of training epochs. The process of training the network identifies the relevant features for solving a specific problem. Adapting the pre-trained network to solve a different problem requires fine-tuning until relevant features are obtained. This is facilitated through the use of large number of filters present in the convolutional layers of pre-trained network. A very few features out of these features are useful for solving the problem in a different domain, while others are irrelevant, use of which may only reduce the efficacy of the network. However, by minimizing the number of filters required to solve the problem, the efficiency of the training the network can be improved. In this study, we consider identification of relevant filters using the pre-trained networks namely AlexNet and VGG-16 net to detect cervical cancer from cervix images. This paper presents a novel hybrid transfer learning technique, in which a CNN is built and trained from scratch, with initial weights of only those filters which were identified as relevant using AlexNet and VGG-16 net. This study used 2198 cervix images with 1090 belonging to negative class and 1108 to positive class. Our experiment using hybrid transfer learning achieved an accuracy of 91.46{\%}.",
author = "Vidya Kudva and Keerthana Prasad and Shyamala Guruvare",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s10278-019-00269-1",
language = "English",
journal = "Journal of Digital Imaging",
issn = "0897-1889",
publisher = "Springer New York",

}

TY - JOUR

T1 - Hybrid Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening

AU - Kudva, Vidya

AU - Prasad, Keerthana

AU - Guruvare, Shyamala

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Transfer learning using deep pre-trained convolutional neural networks is increasingly used to solve a large number of problems in the medical field. In spite of being trained using images with entirely different domain, these networks are flexible to adapt to solve a problem in a different domain too. Transfer learning involves fine-tuning a pre-trained network with optimal values of hyperparameters such as learning rate, batch size, and number of training epochs. The process of training the network identifies the relevant features for solving a specific problem. Adapting the pre-trained network to solve a different problem requires fine-tuning until relevant features are obtained. This is facilitated through the use of large number of filters present in the convolutional layers of pre-trained network. A very few features out of these features are useful for solving the problem in a different domain, while others are irrelevant, use of which may only reduce the efficacy of the network. However, by minimizing the number of filters required to solve the problem, the efficiency of the training the network can be improved. In this study, we consider identification of relevant filters using the pre-trained networks namely AlexNet and VGG-16 net to detect cervical cancer from cervix images. This paper presents a novel hybrid transfer learning technique, in which a CNN is built and trained from scratch, with initial weights of only those filters which were identified as relevant using AlexNet and VGG-16 net. This study used 2198 cervix images with 1090 belonging to negative class and 1108 to positive class. Our experiment using hybrid transfer learning achieved an accuracy of 91.46%.

AB - Transfer learning using deep pre-trained convolutional neural networks is increasingly used to solve a large number of problems in the medical field. In spite of being trained using images with entirely different domain, these networks are flexible to adapt to solve a problem in a different domain too. Transfer learning involves fine-tuning a pre-trained network with optimal values of hyperparameters such as learning rate, batch size, and number of training epochs. The process of training the network identifies the relevant features for solving a specific problem. Adapting the pre-trained network to solve a different problem requires fine-tuning until relevant features are obtained. This is facilitated through the use of large number of filters present in the convolutional layers of pre-trained network. A very few features out of these features are useful for solving the problem in a different domain, while others are irrelevant, use of which may only reduce the efficacy of the network. However, by minimizing the number of filters required to solve the problem, the efficiency of the training the network can be improved. In this study, we consider identification of relevant filters using the pre-trained networks namely AlexNet and VGG-16 net to detect cervical cancer from cervix images. This paper presents a novel hybrid transfer learning technique, in which a CNN is built and trained from scratch, with initial weights of only those filters which were identified as relevant using AlexNet and VGG-16 net. This study used 2198 cervix images with 1090 belonging to negative class and 1108 to positive class. Our experiment using hybrid transfer learning achieved an accuracy of 91.46%.

UR - http://www.scopus.com/inward/record.url?scp=85076884393&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85076884393&partnerID=8YFLogxK

U2 - 10.1007/s10278-019-00269-1

DO - 10.1007/s10278-019-00269-1

M3 - Article

AN - SCOPUS:85076884393

JO - Journal of Digital Imaging

JF - Journal of Digital Imaging

SN - 0897-1889

ER -