Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation

Zihao Guo; Jianqiao Zhou; Di Zhao

doi:10.1109/ITNEC48623.2020.9085093

Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation

Zihao Guo
, Jianqiao Zhou
, Di Zhao

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

10 Citations (Scopus)

Abstract

The segmentation of thyroid nodule ultrasonic image is a critical step for thyroid disease diagnosis. With the advent of medical big data, deep convolutional neural networks (DCNNs) have contributed to the analysis of medical image. However, there is still room for improving the accuracy of the result. In this paper, we employ several data pre-processing algorithms to amplify the feature of the original data as well as augment the whole dataset. Moreover, we use a deep learning model, improved DeepLab v3+ segmentation DCNN to achieve better training and prediction performance on thyroid nodule dataset. The results show that the dice similarity coefficient is measured to be 94.08% and accuracy is 97.91%, which reveals the advance nature of our system.

Original language	English
Title of host publication	Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020
Editors	Bing Xu, Kefen Mou
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	549-554
Number of pages	6
ISBN (Electronic)	9781728143903
DOIs	https://doi.org/10.1109/ITNEC48623.2020.9085093
Publication status	Published - Jun 2020
Externally published	Yes
Event	4th IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020 - Chongqing, China Duration: 12 Jun 2020 → 14 Jun 2020

Publication series

Name	Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020

Conference

Conference	4th IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020
Country/Territory	China
City	Chongqing
Period	12/06/20 → 14/06/20

UN SDGs

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

SDG 7 Affordable and Clean Energy

Keywords

Convolutional Neural Network
Data Augmentation Thyroid Nodule
Deep Learning
Medical Image
Sematic Segmentation
Ultrasonic Image

Access to Document

10.1109/ITNEC48623.2020.9085093

Cite this

Guo, Z., Zhou, J., & Zhao, D. (2020). Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation. In B. Xu, & K. Mou (Eds.), Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020 (pp. 549-554). Article 9085093 (Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ITNEC48623.2020.9085093

Guo, Zihao ; Zhou, Jianqiao ; Zhao, Di. / Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation. Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020. editor / Bing Xu ; Kefen Mou. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 549-554 (Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020).

@inproceedings{0ebf899d0ef844389fcddfd13cd93645,

title = "Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation",

abstract = "The segmentation of thyroid nodule ultrasonic image is a critical step for thyroid disease diagnosis. With the advent of medical big data, deep convolutional neural networks (DCNNs) have contributed to the analysis of medical image. However, there is still room for improving the accuracy of the result. In this paper, we employ several data pre-processing algorithms to amplify the feature of the original data as well as augment the whole dataset. Moreover, we use a deep learning model, improved DeepLab v3+ segmentation DCNN to achieve better training and prediction performance on thyroid nodule dataset. The results show that the dice similarity coefficient is measured to be 94.08\% and accuracy is 97.91\%, which reveals the advance nature of our system.",

keywords = "Convolutional Neural Network, Data Augmentation Thyroid Nodule, Deep Learning, Medical Image, Sematic Segmentation, Ultrasonic Image",

author = "Zihao Guo and Jianqiao Zhou and Di Zhao",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 4th IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020 ; Conference date: 12-06-2020 Through 14-06-2020",

year = "2020",

month = jun,

doi = "10.1109/ITNEC48623.2020.9085093",

language = "English",

series = "Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "549--554",

editor = "Bing Xu and Kefen Mou",

booktitle = "Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020",

address = "United States",

}

Guo, Z, Zhou, J & Zhao, D 2020, Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation. in B Xu & K Mou (eds), Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020., 9085093, Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020, Institute of Electrical and Electronics Engineers Inc., pp. 549-554, 4th IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020, Chongqing, China, 12/06/20. https://doi.org/10.1109/ITNEC48623.2020.9085093

Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation. / Guo, Zihao; Zhou, Jianqiao; Zhao, Di.
Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020. ed. / Bing Xu; Kefen Mou. Institute of Electrical and Electronics Engineers Inc., 2020. p. 549-554 9085093 (Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020).

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

TY - GEN

T1 - Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation

AU - Guo, Zihao

AU - Zhou, Jianqiao

AU - Zhao, Di

PY - 2020/6

Y1 - 2020/6

N2 - The segmentation of thyroid nodule ultrasonic image is a critical step for thyroid disease diagnosis. With the advent of medical big data, deep convolutional neural networks (DCNNs) have contributed to the analysis of medical image. However, there is still room for improving the accuracy of the result. In this paper, we employ several data pre-processing algorithms to amplify the feature of the original data as well as augment the whole dataset. Moreover, we use a deep learning model, improved DeepLab v3+ segmentation DCNN to achieve better training and prediction performance on thyroid nodule dataset. The results show that the dice similarity coefficient is measured to be 94.08% and accuracy is 97.91%, which reveals the advance nature of our system.

AB - The segmentation of thyroid nodule ultrasonic image is a critical step for thyroid disease diagnosis. With the advent of medical big data, deep convolutional neural networks (DCNNs) have contributed to the analysis of medical image. However, there is still room for improving the accuracy of the result. In this paper, we employ several data pre-processing algorithms to amplify the feature of the original data as well as augment the whole dataset. Moreover, we use a deep learning model, improved DeepLab v3+ segmentation DCNN to achieve better training and prediction performance on thyroid nodule dataset. The results show that the dice similarity coefficient is measured to be 94.08% and accuracy is 97.91%, which reveals the advance nature of our system.

KW - Convolutional Neural Network

KW - Data Augmentation Thyroid Nodule

KW - Deep Learning

KW - Medical Image

KW - Sematic Segmentation

KW - Ultrasonic Image

UR - https://www.scopus.com/pages/publications/85086253940

U2 - 10.1109/ITNEC48623.2020.9085093

DO - 10.1109/ITNEC48623.2020.9085093

M3 - Conference contribution

AN - SCOPUS:85086253940

T3 - Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020

SP - 549

EP - 554

BT - Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020

A2 - Xu, Bing

A2 - Mou, Kefen

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 4th IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020

Y2 - 12 June 2020 through 14 June 2020

ER -

Guo Z, Zhou J, Zhao D. Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation. In Xu B, Mou K, editors, Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020. Institute of Electrical and Electronics Engineers Inc. 2020. p. 549-554. 9085093. (Proceedings of 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2020). doi: 10.1109/ITNEC48623.2020.9085093

Thyroid Nodule Ultrasonic Imaging Segmentation Based on a Deep Learning Model and Data Augmentation

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this