Automatic Finger Tendon Segmentation from Ultrasound Images Using Deep Learning

Chan Pang Kuok, Bo Siang Tsai, Tai-Hua Yang, Fong-chin Su, I. Ming Jou, Yung-Nien Sun

研究成果: Conference contribution

摘要

Ultrasound imaging is the most commonly applied method for the diagnosis and surgery of a trigger finger. However, the ultrasound images are noisy and the boundaries of tissues are usually very unclear and fuzzy. Therefore, an automatic computer assisted tool for the tissues segmentation is desired and developed. The segmentation results of the conventional methods were satisfactory but they usually depended on the prior knowledge. Recently, the deep-learning convolutional neural network (CNN) shows amazing performance on image processing and it can process the image end-to-end. In this study, we propose a finger tendon segmentation CNN which overcomes the requirement of prior knowledge and gives promising results on ultrasound images. The evaluation result is remarkable high with DSC 0.884 on 380 testing images and the prediction time is fast by 0.027 s per image. This work, to our best of knowledge, is the first deep learning finger tendon segmentation method from transverse ultrasound images.

原文English
主出版物標題New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers
編輯Chuan-Yu Chang, Chien-Chou Lin, Horng-Horng Lin
發行者Springer Verlag
頁面778-784
頁數7
ISBN(列印)9789811391897
DOIs
出版狀態Published - 2019 一月 1
事件23rd International Computer Symposium, ICS 2018 - Yunlin, Taiwan
持續時間: 2018 十二月 202018 十二月 22

出版系列

名字Communications in Computer and Information Science
1013
ISSN(列印)1865-0929

Conference

Conference23rd International Computer Symposium, ICS 2018
國家Taiwan
城市Yunlin
期間18-12-2018-12-22

指紋

Ultrasound Image
Tendons
Segmentation
Ultrasonics
Prior Knowledge
Neural Networks
Tissue
Neural networks
Ultrasound
Trigger
Surgery
Image Processing
Image processing
Transverse
Imaging
Imaging techniques
Testing
Learning
Deep learning
Prediction

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Mathematics(all)

引用此文

Kuok, C. P., Tsai, B. S., Yang, T-H., Su, F., Jou, I. M., & Sun, Y-N. (2019). Automatic Finger Tendon Segmentation from Ultrasound Images Using Deep Learning. 於 C-Y. Chang, C-C. Lin, & H-H. Lin (編輯), New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers (頁 778-784). (Communications in Computer and Information Science; 卷 1013). Springer Verlag. https://doi.org/10.1007/978-981-13-9190-3_84
Kuok, Chan Pang ; Tsai, Bo Siang ; Yang, Tai-Hua ; Su, Fong-chin ; Jou, I. Ming ; Sun, Yung-Nien. / Automatic Finger Tendon Segmentation from Ultrasound Images Using Deep Learning. New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers. 編輯 / Chuan-Yu Chang ; Chien-Chou Lin ; Horng-Horng Lin. Springer Verlag, 2019. 頁 778-784 (Communications in Computer and Information Science).
@inproceedings{76c54879106c4b8fba1e28a408404b66,
title = "Automatic Finger Tendon Segmentation from Ultrasound Images Using Deep Learning",
abstract = "Ultrasound imaging is the most commonly applied method for the diagnosis and surgery of a trigger finger. However, the ultrasound images are noisy and the boundaries of tissues are usually very unclear and fuzzy. Therefore, an automatic computer assisted tool for the tissues segmentation is desired and developed. The segmentation results of the conventional methods were satisfactory but they usually depended on the prior knowledge. Recently, the deep-learning convolutional neural network (CNN) shows amazing performance on image processing and it can process the image end-to-end. In this study, we propose a finger tendon segmentation CNN which overcomes the requirement of prior knowledge and gives promising results on ultrasound images. The evaluation result is remarkable high with DSC 0.884 on 380 testing images and the prediction time is fast by 0.027 s per image. This work, to our best of knowledge, is the first deep learning finger tendon segmentation method from transverse ultrasound images.",
author = "Kuok, {Chan Pang} and Tsai, {Bo Siang} and Tai-Hua Yang and Fong-chin Su and Jou, {I. Ming} and Yung-Nien Sun",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-981-13-9190-3_84",
language = "English",
isbn = "9789811391897",
series = "Communications in Computer and Information Science",
publisher = "Springer Verlag",
pages = "778--784",
editor = "Chuan-Yu Chang and Chien-Chou Lin and Horng-Horng Lin",
booktitle = "New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers",
address = "Germany",

}

Kuok, CP, Tsai, BS, Yang, T-H, Su, F, Jou, IM & Sun, Y-N 2019, Automatic Finger Tendon Segmentation from Ultrasound Images Using Deep Learning. 於 C-Y Chang, C-C Lin & H-H Lin (編輯), New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers. Communications in Computer and Information Science, 卷 1013, Springer Verlag, 頁 778-784, 23rd International Computer Symposium, ICS 2018, Yunlin, Taiwan, 18-12-20. https://doi.org/10.1007/978-981-13-9190-3_84

Automatic Finger Tendon Segmentation from Ultrasound Images Using Deep Learning. / Kuok, Chan Pang; Tsai, Bo Siang; Yang, Tai-Hua; Su, Fong-chin; Jou, I. Ming; Sun, Yung-Nien.

New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers. 編輯 / Chuan-Yu Chang; Chien-Chou Lin; Horng-Horng Lin. Springer Verlag, 2019. p. 778-784 (Communications in Computer and Information Science; 卷 1013).

研究成果: Conference contribution

TY - GEN

T1 - Automatic Finger Tendon Segmentation from Ultrasound Images Using Deep Learning

AU - Kuok, Chan Pang

AU - Tsai, Bo Siang

AU - Yang, Tai-Hua

AU - Su, Fong-chin

AU - Jou, I. Ming

AU - Sun, Yung-Nien

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Ultrasound imaging is the most commonly applied method for the diagnosis and surgery of a trigger finger. However, the ultrasound images are noisy and the boundaries of tissues are usually very unclear and fuzzy. Therefore, an automatic computer assisted tool for the tissues segmentation is desired and developed. The segmentation results of the conventional methods were satisfactory but they usually depended on the prior knowledge. Recently, the deep-learning convolutional neural network (CNN) shows amazing performance on image processing and it can process the image end-to-end. In this study, we propose a finger tendon segmentation CNN which overcomes the requirement of prior knowledge and gives promising results on ultrasound images. The evaluation result is remarkable high with DSC 0.884 on 380 testing images and the prediction time is fast by 0.027 s per image. This work, to our best of knowledge, is the first deep learning finger tendon segmentation method from transverse ultrasound images.

AB - Ultrasound imaging is the most commonly applied method for the diagnosis and surgery of a trigger finger. However, the ultrasound images are noisy and the boundaries of tissues are usually very unclear and fuzzy. Therefore, an automatic computer assisted tool for the tissues segmentation is desired and developed. The segmentation results of the conventional methods were satisfactory but they usually depended on the prior knowledge. Recently, the deep-learning convolutional neural network (CNN) shows amazing performance on image processing and it can process the image end-to-end. In this study, we propose a finger tendon segmentation CNN which overcomes the requirement of prior knowledge and gives promising results on ultrasound images. The evaluation result is remarkable high with DSC 0.884 on 380 testing images and the prediction time is fast by 0.027 s per image. This work, to our best of knowledge, is the first deep learning finger tendon segmentation method from transverse ultrasound images.

UR - http://www.scopus.com/inward/record.url?scp=85069735935&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069735935&partnerID=8YFLogxK

U2 - 10.1007/978-981-13-9190-3_84

DO - 10.1007/978-981-13-9190-3_84

M3 - Conference contribution

AN - SCOPUS:85069735935

SN - 9789811391897

T3 - Communications in Computer and Information Science

SP - 778

EP - 784

BT - New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers

A2 - Chang, Chuan-Yu

A2 - Lin, Chien-Chou

A2 - Lin, Horng-Horng

PB - Springer Verlag

ER -

Kuok CP, Tsai BS, Yang T-H, Su F, Jou IM, Sun Y-N. Automatic Finger Tendon Segmentation from Ultrasound Images Using Deep Learning. 於 Chang C-Y, Lin C-C, Lin H-H, 編輯, New Trends in Computer Technologies and Applications - 23rd International Computer Symposium, ICS 2018, Revised Selected Papers. Springer Verlag. 2019. p. 778-784. (Communications in Computer and Information Science). https://doi.org/10.1007/978-981-13-9190-3_84