Automated catheter localization in volumetric ultrasound using 3D patch-wise U-net with focal loss

Hongxu Yang, Caifeng Shan, Alexander F. Kolen, Peter de With

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

3D ultrasound (US) imaging has become an attractive option for image-guided interventions. Fast and accurate catheter localization in 3D cardiac US can improve the outcome and efficiency of the cardiac interventions. In this paper, we propose a catheter localization method for 3D cardiac US using the patch-wise semantic segmentation with model fitting. Our 3D U-Net is trained with the focal loss of cross-entropy, which makes the network to focus more on samples that are difficult to classify. Moreover, we adopt a dense sampling strategy to overcome the extremely imbalanced catheter occupation in the 3D US data. Extensive experiments on our challenging ex-vivo dataset show that the proposed method achieves an F-1 score of 65.1% for catheter segmentation, outperforming the state-of-the-art methods. With this, our method can localize RF-ablation catheters with an average error of 1.28 mm.
LanguageEnglish
Title of host publicationIEEE International Conference on Image Processing 2019
PublisherInstitute of Electrical and Electronics Engineers
DOIs
StatePublished - 2019
Event26th IEEE International Conference on Image Processing (ICIP 2019) - Taipei, Taiwan
Duration: 22 Sep 201925 Sep 2019

Conference

Conference26th IEEE International Conference on Image Processing (ICIP 2019)
Abbreviated titleICIP 2019
CountryTaiwan
CityTaipei
Period22/09/1925/09/19

Fingerprint

Catheters
Ultrasonics
Ablation
Entropy
Semantics
Sampling
Imaging techniques
Experiments

Cite this

Yang, H., Shan, C., Kolen, A. F., & de With, P. (2019). Automated catheter localization in volumetric ultrasound using 3D patch-wise U-net with focal loss. In IEEE International Conference on Image Processing 2019 Institute of Electrical and Electronics Engineers. DOI: 10.1109/ICIP.2019.8803045
Yang, Hongxu ; Shan, Caifeng ; Kolen, Alexander F. ; de With, Peter. / Automated catheter localization in volumetric ultrasound using 3D patch-wise U-net with focal loss. IEEE International Conference on Image Processing 2019. Institute of Electrical and Electronics Engineers, 2019.
@inproceedings{d96a6aca4c674ac687a79afe415d6f72,
title = "Automated catheter localization in volumetric ultrasound using 3D patch-wise U-net with focal loss",
abstract = "3D ultrasound (US) imaging has become an attractive option for image-guided interventions. Fast and accurate catheter localization in 3D cardiac US can improve the outcome and efficiency of the cardiac interventions. In this paper, we propose a catheter localization method for 3D cardiac US using the patch-wise semantic segmentation with model fitting. Our 3D U-Net is trained with the focal loss of cross-entropy, which makes the network to focus more on samples that are difficult to classify. Moreover, we adopt a dense sampling strategy to overcome the extremely imbalanced catheter occupation in the 3D US data. Extensive experiments on our challenging ex-vivo dataset show that the proposed method achieves an F-1 score of 65.1{\%} for catheter segmentation, outperforming the state-of-the-art methods. With this, our method can localize RF-ablation catheters with an average error of 1.28 mm.",
author = "Hongxu Yang and Caifeng Shan and Kolen, {Alexander F.} and {de With}, Peter",
year = "2019",
doi = "10.1109/ICIP.2019.8803045",
language = "English",
booktitle = "IEEE International Conference on Image Processing 2019",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Yang, H, Shan, C, Kolen, AF & de With, P 2019, Automated catheter localization in volumetric ultrasound using 3D patch-wise U-net with focal loss. in IEEE International Conference on Image Processing 2019. Institute of Electrical and Electronics Engineers, 26th IEEE International Conference on Image Processing (ICIP 2019), Taipei, Taiwan, 22/09/19. DOI: 10.1109/ICIP.2019.8803045

Automated catheter localization in volumetric ultrasound using 3D patch-wise U-net with focal loss. / Yang, Hongxu; Shan, Caifeng; Kolen, Alexander F.; de With, Peter.

IEEE International Conference on Image Processing 2019. Institute of Electrical and Electronics Engineers, 2019.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Automated catheter localization in volumetric ultrasound using 3D patch-wise U-net with focal loss

AU - Yang,Hongxu

AU - Shan,Caifeng

AU - Kolen,Alexander F.

AU - de With,Peter

PY - 2019

Y1 - 2019

N2 - 3D ultrasound (US) imaging has become an attractive option for image-guided interventions. Fast and accurate catheter localization in 3D cardiac US can improve the outcome and efficiency of the cardiac interventions. In this paper, we propose a catheter localization method for 3D cardiac US using the patch-wise semantic segmentation with model fitting. Our 3D U-Net is trained with the focal loss of cross-entropy, which makes the network to focus more on samples that are difficult to classify. Moreover, we adopt a dense sampling strategy to overcome the extremely imbalanced catheter occupation in the 3D US data. Extensive experiments on our challenging ex-vivo dataset show that the proposed method achieves an F-1 score of 65.1% for catheter segmentation, outperforming the state-of-the-art methods. With this, our method can localize RF-ablation catheters with an average error of 1.28 mm.

AB - 3D ultrasound (US) imaging has become an attractive option for image-guided interventions. Fast and accurate catheter localization in 3D cardiac US can improve the outcome and efficiency of the cardiac interventions. In this paper, we propose a catheter localization method for 3D cardiac US using the patch-wise semantic segmentation with model fitting. Our 3D U-Net is trained with the focal loss of cross-entropy, which makes the network to focus more on samples that are difficult to classify. Moreover, we adopt a dense sampling strategy to overcome the extremely imbalanced catheter occupation in the 3D US data. Extensive experiments on our challenging ex-vivo dataset show that the proposed method achieves an F-1 score of 65.1% for catheter segmentation, outperforming the state-of-the-art methods. With this, our method can localize RF-ablation catheters with an average error of 1.28 mm.

U2 - 10.1109/ICIP.2019.8803045

DO - 10.1109/ICIP.2019.8803045

M3 - Conference contribution

BT - IEEE International Conference on Image Processing 2019

PB - Institute of Electrical and Electronics Engineers

ER -

Yang H, Shan C, Kolen AF, de With P. Automated catheter localization in volumetric ultrasound using 3D patch-wise U-net with focal loss. In IEEE International Conference on Image Processing 2019. Institute of Electrical and Electronics Engineers. 2019. Available from, DOI: 10.1109/ICIP.2019.8803045