Locating Fetal Facial Surface, Oral Cavity and Airways by a 3D Ultrasound Calibration Using a Novel Cones' Phantom

Rong XU; Jun OHYA; Yoshinobu SATO; Bo ZHANG; Masakatsu G. FUJIE

doi:10.1587/transinf.E97.D.1324

Locating Fetal Facial Surface, Oral Cavity and Airways by a 3D Ultrasound Calibration Using a Novel Cones' Phantom

Rong XU, Jun OHYA, Yoshinobu SATO, Bo ZHANG, Masakatsu G. FUJIE

Full Text Views

0

Cite this

Summary :

Toward the actualization of an automatic navigation system for fetoscopic tracheal occlusion (FETO) surgery, this paper proposes a 3D ultrasound (US) calibration-based approach that can locate the fetal facial surface, oral cavity, and airways by a registration between a 3D fetal model and 3D US images. The proposed approach consists of an offline process and online process. The offline process first reconstructs the 3D fetal model with the anatomies of the oral cavity and airways. Then, a point-based 3D US calibration system based on real-time 3D US images, an electromagnetic (EM) tracking device, and a novel cones' phantom, computes the matrix that transforms the 3D US image space into the world coordinate system. In the online process, by scanning the mother's body with a 3D US probe, 3D US images containing the fetus are obtained. The fetal facial surface extracted from the 3D US images is registered to the 3D fetal model using an ICP-based (iterative closest point) algorithm and the calibration matrices, so that the fetal facial surface as well as the oral cavity and airways are located. The results indicate that the 3D US calibration system achieves an FRE (fiducial registration error) of 1.49±0.44mm and a TRE (target registration error) of 1.81±0.56mm by using 24 fiducial points from two US volumes. A mean TRE of 1.55±0.46 mm is also achieved for measuring location accuracy of the 3D fetal facial surface extracted from 3D US images by 14 target markers, and mean location errors of 2.51±0.47 mm and 3.04±0.59 mm are achieved for indirectly measuring location accuracy of the pharynx and the entrance of the trachea, respectively, which satisfy the requirement of the FETO surgery.

Publication: IEICE TRANSACTIONS on Information Vol.E97-D No.5 pp.1324-1335

Publication Date: 2014/05/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E97.D.1324

Type of Manuscript: PAPER

Category: Biological Engineering

Authors

Rong XU
  Waseda University
Jun OHYA
  Waseda University
Yoshinobu SATO
  Osaka University
Bo ZHANG
  Waseda University
Masakatsu G. FUJIE
  Waseda University

Keyword

3D location, fetal oral cavity and airways, 3D ultrasound calibration, iterative closest point algorithm, 3D fetal model, 3D electromagnetic tracking device

Cite this

Copy

Rong XU, Jun OHYA, Yoshinobu SATO, Bo ZHANG, Masakatsu G. FUJIE, "Locating Fetal Facial Surface, Oral Cavity and Airways by a 3D Ultrasound Calibration Using a Novel Cones' Phantom" in IEICE TRANSACTIONS on Information, vol. E97-D, no. 5, pp. 1324-1335, May 2014, doi: 10.1587/transinf.E97.D.1324.
Abstract: Toward the actualization of an automatic navigation system for fetoscopic tracheal occlusion (FETO) surgery, this paper proposes a 3D ultrasound (US) calibration-based approach that can locate the fetal facial surface, oral cavity, and airways by a registration between a 3D fetal model and 3D US images. The proposed approach consists of an offline process and online process. The offline process first reconstructs the 3D fetal model with the anatomies of the oral cavity and airways. Then, a point-based 3D US calibration system based on real-time 3D US images, an electromagnetic (EM) tracking device, and a novel cones' phantom, computes the matrix that transforms the 3D US image space into the world coordinate system. In the online process, by scanning the mother's body with a 3D US probe, 3D US images containing the fetus are obtained. The fetal facial surface extracted from the 3D US images is registered to the 3D fetal model using an ICP-based (iterative closest point) algorithm and the calibration matrices, so that the fetal facial surface as well as the oral cavity and airways are located. The results indicate that the 3D US calibration system achieves an FRE (fiducial registration error) of 1.49±0.44mm and a TRE (target registration error) of 1.81±0.56mm by using 24 fiducial points from two US volumes. A mean TRE of 1.55±0.46 mm is also achieved for measuring location accuracy of the 3D fetal facial surface extracted from 3D US images by 14 target markers, and mean location errors of 2.51±0.47 mm and 3.04±0.59 mm are achieved for indirectly measuring location accuracy of the pharynx and the entrance of the trachea, respectively, which satisfy the requirement of the FETO surgery.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E97.D.1324/_p

Copy

@ARTICLE{e97-d_5_1324,
author={Rong XU, Jun OHYA, Yoshinobu SATO, Bo ZHANG, Masakatsu G. FUJIE, },
journal={IEICE TRANSACTIONS on Information},
title={Locating Fetal Facial Surface, Oral Cavity and Airways by a 3D Ultrasound Calibration Using a Novel Cones' Phantom},
year={2014},
volume={E97-D},
number={5},
pages={1324-1335},
abstract={Toward the actualization of an automatic navigation system for fetoscopic tracheal occlusion (FETO) surgery, this paper proposes a 3D ultrasound (US) calibration-based approach that can locate the fetal facial surface, oral cavity, and airways by a registration between a 3D fetal model and 3D US images. The proposed approach consists of an offline process and online process. The offline process first reconstructs the 3D fetal model with the anatomies of the oral cavity and airways. Then, a point-based 3D US calibration system based on real-time 3D US images, an electromagnetic (EM) tracking device, and a novel cones' phantom, computes the matrix that transforms the 3D US image space into the world coordinate system. In the online process, by scanning the mother's body with a 3D US probe, 3D US images containing the fetus are obtained. The fetal facial surface extracted from the 3D US images is registered to the 3D fetal model using an ICP-based (iterative closest point) algorithm and the calibration matrices, so that the fetal facial surface as well as the oral cavity and airways are located. The results indicate that the 3D US calibration system achieves an FRE (fiducial registration error) of 1.49±0.44mm and a TRE (target registration error) of 1.81±0.56mm by using 24 fiducial points from two US volumes. A mean TRE of 1.55±0.46 mm is also achieved for measuring location accuracy of the 3D fetal facial surface extracted from 3D US images by 14 target markers, and mean location errors of 2.51±0.47 mm and 3.04±0.59 mm are achieved for indirectly measuring location accuracy of the pharynx and the entrance of the trachea, respectively, which satisfy the requirement of the FETO surgery.},
keywords={},
doi={10.1587/transinf.E97.D.1324},
ISSN={1745-1361},
month={May},}

Copy

TY - JOUR
TI - Locating Fetal Facial Surface, Oral Cavity and Airways by a 3D Ultrasound Calibration Using a Novel Cones' Phantom
T2 - IEICE TRANSACTIONS on Information
SP - 1324
EP - 1335
AU - Rong XU
AU - Jun OHYA
AU - Yoshinobu SATO
AU - Bo ZHANG
AU - Masakatsu G. FUJIE
PY - 2014
DO - 10.1587/transinf.E97.D.1324
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E97-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2014
AB - Toward the actualization of an automatic navigation system for fetoscopic tracheal occlusion (FETO) surgery, this paper proposes a 3D ultrasound (US) calibration-based approach that can locate the fetal facial surface, oral cavity, and airways by a registration between a 3D fetal model and 3D US images. The proposed approach consists of an offline process and online process. The offline process first reconstructs the 3D fetal model with the anatomies of the oral cavity and airways. Then, a point-based 3D US calibration system based on real-time 3D US images, an electromagnetic (EM) tracking device, and a novel cones' phantom, computes the matrix that transforms the 3D US image space into the world coordinate system. In the online process, by scanning the mother's body with a 3D US probe, 3D US images containing the fetus are obtained. The fetal facial surface extracted from the 3D US images is registered to the 3D fetal model using an ICP-based (iterative closest point) algorithm and the calibration matrices, so that the fetal facial surface as well as the oral cavity and airways are located. The results indicate that the 3D US calibration system achieves an FRE (fiducial registration error) of 1.49±0.44mm and a TRE (target registration error) of 1.81±0.56mm by using 24 fiducial points from two US volumes. A mean TRE of 1.55±0.46 mm is also achieved for measuring location accuracy of the 3D fetal facial surface extracted from 3D US images by 14 target markers, and mean location errors of 2.51±0.47 mm and 3.04±0.59 mm are achieved for indirectly measuring location accuracy of the pharynx and the entrance of the trachea, respectively, which satisfy the requirement of the FETO surgery.
ER -