Feasibility study of catheter segmentation in 3D Frustum Ultrasounds by DCNN

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

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

Abstract

3D ultrasound has been developed rapidly in medical intervention therapies, such as cardiac catheterization. Image-based catheter detection is studied to help sonographer to timely localize the instrument in the 3D US images. Conventionally, the 3D imaging methods are based on the Cartesian domain, which is limited by bandwidth and information lose when it is converted from the original acquisition space---Frustum domain. The catheter segmentation in the Frustum space helps to reduce the computational cost and improve efficiency. In this paper, we present a catheter segmentation method in 3D Frustum image via a deep convolutional network (DCNN). To accelerate the prediction efficiency on whole US Frustum volume, a filter-based pre-selection is applied to reduce the computational cost of the DCNN. Based on experiments on the ex-vivo dataset, our proposed method can segment the catheter in Frustum images with 0.67 Dice score within 3 seconds.
LanguageEnglish
Title of host publicationImage-Guided Procedures, Robotic Interventions, and Modeling
Subtitle of host publicationImage-Guided Procedures, Robotic Interventions, and Modeling
StateAccepted/In press - 2020
Event2020 SPIE Medical Imaging: Image-Guided Procedures, Robotic Interventions, and Modeling - Houston, United States
Duration: 16 Feb 202019 Feb 2020

Conference

Conference2020 SPIE Medical Imaging
CountryUnited States
CityHouston
Period16/02/2019/02/20

Fingerprint

Catheters
Ultrasonics
Costs
Bandwidth
Imaging techniques
Experiments

Cite this

Min, L., Yang, H., Shan, C., Kolen, A. F., & de With, P. (Accepted/In press). Feasibility study of catheter segmentation in 3D Frustum Ultrasounds by DCNN. In Image-Guided Procedures, Robotic Interventions, and Modeling: Image-Guided Procedures, Robotic Interventions, and Modeling
Min, Lan ; Yang, Hongxu ; Shan, Caifeng ; Kolen, Alexander F. ; de With, Peter. / Feasibility study of catheter segmentation in 3D Frustum Ultrasounds by DCNN. Image-Guided Procedures, Robotic Interventions, and Modeling: Image-Guided Procedures, Robotic Interventions, and Modeling. 2020.
@inproceedings{7ef52ecfc5eb4274a5e1ac4801139cfa,
title = "Feasibility study of catheter segmentation in 3D Frustum Ultrasounds by DCNN",
abstract = "3D ultrasound has been developed rapidly in medical intervention therapies, such as cardiac catheterization. Image-based catheter detection is studied to help sonographer to timely localize the instrument in the 3D US images. Conventionally, the 3D imaging methods are based on the Cartesian domain, which is limited by bandwidth and information lose when it is converted from the original acquisition space---Frustum domain. The catheter segmentation in the Frustum space helps to reduce the computational cost and improve efficiency. In this paper, we present a catheter segmentation method in 3D Frustum image via a deep convolutional network (DCNN). To accelerate the prediction efficiency on whole US Frustum volume, a filter-based pre-selection is applied to reduce the computational cost of the DCNN. Based on experiments on the ex-vivo dataset, our proposed method can segment the catheter in Frustum images with 0.67 Dice score within 3 seconds.",
author = "Lan Min and Hongxu Yang and Caifeng Shan and Kolen, {Alexander F.} and {de With}, Peter",
year = "2020",
language = "English",
booktitle = "Image-Guided Procedures, Robotic Interventions, and Modeling",

}

Min, L, Yang, H, Shan, C, Kolen, AF & de With, P 2020, Feasibility study of catheter segmentation in 3D Frustum Ultrasounds by DCNN. in Image-Guided Procedures, Robotic Interventions, and Modeling: Image-Guided Procedures, Robotic Interventions, and Modeling. 2020 SPIE Medical Imaging, Houston, United States, 16/02/20.

Feasibility study of catheter segmentation in 3D Frustum Ultrasounds by DCNN. / Min, Lan; Yang, Hongxu; Shan, Caifeng; Kolen, Alexander F.; de With, Peter.

Image-Guided Procedures, Robotic Interventions, and Modeling: Image-Guided Procedures, Robotic Interventions, and Modeling. 2020.

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

TY - GEN

T1 - Feasibility study of catheter segmentation in 3D Frustum Ultrasounds by DCNN

AU - Min,Lan

AU - Yang,Hongxu

AU - Shan,Caifeng

AU - Kolen,Alexander F.

AU - de With,Peter

PY - 2020

Y1 - 2020

N2 - 3D ultrasound has been developed rapidly in medical intervention therapies, such as cardiac catheterization. Image-based catheter detection is studied to help sonographer to timely localize the instrument in the 3D US images. Conventionally, the 3D imaging methods are based on the Cartesian domain, which is limited by bandwidth and information lose when it is converted from the original acquisition space---Frustum domain. The catheter segmentation in the Frustum space helps to reduce the computational cost and improve efficiency. In this paper, we present a catheter segmentation method in 3D Frustum image via a deep convolutional network (DCNN). To accelerate the prediction efficiency on whole US Frustum volume, a filter-based pre-selection is applied to reduce the computational cost of the DCNN. Based on experiments on the ex-vivo dataset, our proposed method can segment the catheter in Frustum images with 0.67 Dice score within 3 seconds.

AB - 3D ultrasound has been developed rapidly in medical intervention therapies, such as cardiac catheterization. Image-based catheter detection is studied to help sonographer to timely localize the instrument in the 3D US images. Conventionally, the 3D imaging methods are based on the Cartesian domain, which is limited by bandwidth and information lose when it is converted from the original acquisition space---Frustum domain. The catheter segmentation in the Frustum space helps to reduce the computational cost and improve efficiency. In this paper, we present a catheter segmentation method in 3D Frustum image via a deep convolutional network (DCNN). To accelerate the prediction efficiency on whole US Frustum volume, a filter-based pre-selection is applied to reduce the computational cost of the DCNN. Based on experiments on the ex-vivo dataset, our proposed method can segment the catheter in Frustum images with 0.67 Dice score within 3 seconds.

M3 - Conference contribution

BT - Image-Guided Procedures, Robotic Interventions, and Modeling

ER -

Min L, Yang H, Shan C, Kolen AF, de With P. Feasibility study of catheter segmentation in 3D Frustum Ultrasounds by DCNN. In Image-Guided Procedures, Robotic Interventions, and Modeling: Image-Guided Procedures, Robotic Interventions, and Modeling. 2020.