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

Vidya Kudva; Keerthana Prasad; Shyamala Guruvare

doi:10.1007/978-981-15-0626-0_25

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

Vidya Kudva^*
, Keerthana Prasad
, Shyamala Guruvare

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

10 Citations (Scopus)

Abstract

Automated analysis of digital cervix images acquired during Visual Inspection with Acetic acid (VIA) is found to be of great help to aid the physicians to diagnose cervical cancer. Traditional classification methods require many features to distinguish between normal and abnormal cervix. Selection of distinct visual features which well represent the data and at the same time are capable of performing discriminative learning is complex. This problem can be overcome using deep learning approaches. Transfer learning is one of the deep learning approaches, which facilitates the use of a pre-trained network for a specific problem at hand. This paper presents a transfer learning using AlexNet, which is a pre-trained convolutional neural network, for classification of the cervix images into two classes namely negative and positive. This study used 2198 cervix images with 1090 belonging to negative class and 1108 to positive class. Our experiment using AlexNet for transfer learning achieved an accuracy of 0.934.

Original language	English
Title of host publication	Advances in Communication, Signal Processing, VLSI, and Embedded Systems - Select Proceedings of VSPICE 2019
Editors	Shubhakar Kalya, Muralidhar Kulkarni, K. S. Shivaprakasha
Publisher	Springer Paris
Pages	299-312
Number of pages	14
ISBN (Print)	9789811506253
DOIs	https://doi.org/10.1007/978-981-15-0626-0_25
Publication status	Published - 01-01-2020
Event	International Conference on VLSI, Signal Processing, Power Systems, Illumination and Lighting Control, Communication and Embedded Systems, VSPICE 2019 - Deralakatte, India Duration: 23-05-2019 → 24-05-2019

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	614
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	International Conference on VLSI, Signal Processing, Power Systems, Illumination and Lighting Control, Communication and Embedded Systems, VSPICE 2019
Country/Territory	India
City	Deralakatte
Period	23-05-19 → 24-05-19

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

All Science Journal Classification (ASJC) codes

Industrial and Manufacturing Engineering

Access to Document

10.1007/978-981-15-0626-0_25

Cite this

Kudva, V., Prasad, K., & Guruvare, S. (2020). Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening. In S. Kalya, M. Kulkarni, & K. S. Shivaprakasha (Eds.), Advances in Communication, Signal Processing, VLSI, and Embedded Systems - Select Proceedings of VSPICE 2019 (pp. 299-312). (Lecture Notes in Electrical Engineering; Vol. 614). Springer Paris. https://doi.org/10.1007/978-981-15-0626-0_25

Kudva, Vidya ; Prasad, Keerthana ; Guruvare, Shyamala. / Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening. Advances in Communication, Signal Processing, VLSI, and Embedded Systems - Select Proceedings of VSPICE 2019. editor / Shubhakar Kalya ; Muralidhar Kulkarni ; K. S. Shivaprakasha. Springer Paris, 2020. pp. 299-312 (Lecture Notes in Electrical Engineering).

@inproceedings{16fb06b8e587493cafca940492ce9571,

title = "Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening",

abstract = "Automated analysis of digital cervix images acquired during Visual Inspection with Acetic acid (VIA) is found to be of great help to aid the physicians to diagnose cervical cancer. Traditional classification methods require many features to distinguish between normal and abnormal cervix. Selection of distinct visual features which well represent the data and at the same time are capable of performing discriminative learning is complex. This problem can be overcome using deep learning approaches. Transfer learning is one of the deep learning approaches, which facilitates the use of a pre-trained network for a specific problem at hand. This paper presents a transfer learning using AlexNet, which is a pre-trained convolutional neural network, for classification of the cervix images into two classes namely negative and positive. This study used 2198 cervix images with 1090 belonging to negative class and 1108 to positive class. Our experiment using AlexNet for transfer learning achieved an accuracy of 0.934.",

author = "Vidya Kudva and Keerthana Prasad and Shyamala Guruvare",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Singapore Pte Ltd.; International Conference on VLSI, Signal Processing, Power Systems, Illumination and Lighting Control, Communication and Embedded Systems, VSPICE 2019 ; Conference date: 23-05-2019 Through 24-05-2019",

year = "2020",

month = jan,

day = "1",

doi = "10.1007/978-981-15-0626-0\_25",

language = "English",

isbn = "9789811506253",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Paris",

pages = "299--312",

editor = "Shubhakar Kalya and Muralidhar Kulkarni and Shivaprakasha, \{K. S.\}",

booktitle = "Advances in Communication, Signal Processing, VLSI, and Embedded Systems - Select Proceedings of VSPICE 2019",

address = "France",

}

Kudva, V, Prasad, K & Guruvare, S 2020, Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening. in S Kalya, M Kulkarni & KS Shivaprakasha (eds), Advances in Communication, Signal Processing, VLSI, and Embedded Systems - Select Proceedings of VSPICE 2019. Lecture Notes in Electrical Engineering, vol. 614, Springer Paris, pp. 299-312, International Conference on VLSI, Signal Processing, Power Systems, Illumination and Lighting Control, Communication and Embedded Systems, VSPICE 2019, Deralakatte, India, 23-05-19. https://doi.org/10.1007/978-981-15-0626-0_25

Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening. / Kudva, Vidya; Prasad, Keerthana ; Guruvare, Shyamala.
Advances in Communication, Signal Processing, VLSI, and Embedded Systems - Select Proceedings of VSPICE 2019. ed. / Shubhakar Kalya; Muralidhar Kulkarni; K. S. Shivaprakasha. Springer Paris, 2020. p. 299-312 (Lecture Notes in Electrical Engineering; Vol. 614).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

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

AU - Kudva, Vidya

AU - Prasad, Keerthana

AU - Guruvare, Shyamala

PY - 2020/1/1

Y1 - 2020/1/1

N2 - Automated analysis of digital cervix images acquired during Visual Inspection with Acetic acid (VIA) is found to be of great help to aid the physicians to diagnose cervical cancer. Traditional classification methods require many features to distinguish between normal and abnormal cervix. Selection of distinct visual features which well represent the data and at the same time are capable of performing discriminative learning is complex. This problem can be overcome using deep learning approaches. Transfer learning is one of the deep learning approaches, which facilitates the use of a pre-trained network for a specific problem at hand. This paper presents a transfer learning using AlexNet, which is a pre-trained convolutional neural network, for classification of the cervix images into two classes namely negative and positive. This study used 2198 cervix images with 1090 belonging to negative class and 1108 to positive class. Our experiment using AlexNet for transfer learning achieved an accuracy of 0.934.

AB - Automated analysis of digital cervix images acquired during Visual Inspection with Acetic acid (VIA) is found to be of great help to aid the physicians to diagnose cervical cancer. Traditional classification methods require many features to distinguish between normal and abnormal cervix. Selection of distinct visual features which well represent the data and at the same time are capable of performing discriminative learning is complex. This problem can be overcome using deep learning approaches. Transfer learning is one of the deep learning approaches, which facilitates the use of a pre-trained network for a specific problem at hand. This paper presents a transfer learning using AlexNet, which is a pre-trained convolutional neural network, for classification of the cervix images into two classes namely negative and positive. This study used 2198 cervix images with 1090 belonging to negative class and 1108 to positive class. Our experiment using AlexNet for transfer learning achieved an accuracy of 0.934.

UR - https://www.scopus.com/pages/publications/85076974336

UR - https://www.scopus.com/pages/publications/85076974336#tab=citedBy

U2 - 10.1007/978-981-15-0626-0_25

DO - 10.1007/978-981-15-0626-0_25

M3 - Conference contribution

AN - SCOPUS:85076974336

SN - 9789811506253

T3 - Lecture Notes in Electrical Engineering

SP - 299

EP - 312

BT - Advances in Communication, Signal Processing, VLSI, and Embedded Systems - Select Proceedings of VSPICE 2019

A2 - Kalya, Shubhakar

A2 - Kulkarni, Muralidhar

A2 - Shivaprakasha, K. S.

PB - Springer Paris

T2 - International Conference on VLSI, Signal Processing, Power Systems, Illumination and Lighting Control, Communication and Embedded Systems, VSPICE 2019

Y2 - 23 May 2019 through 24 May 2019

ER -

Kudva V, Prasad K , Guruvare S. Transfer Learning for Classification of Uterine Cervix Images for Cervical Cancer Screening. In Kalya S, Kulkarni M, Shivaprakasha KS, editors, Advances in Communication, Signal Processing, VLSI, and Embedded Systems - Select Proceedings of VSPICE 2019. Springer Paris. 2020. p. 299-312. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-15-0626-0_25

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

Abstract

Publication series

Conference

UN SDGs

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this