Three-dimensional multi-scale line filter for segmentation and visualization of curvilinear structures in medical images

Yoshinobu Sato; Shin Nakajima; Nobuyuki Shiraga; Hideki Atsumi; Shigeyuki Yoshida; Thomas Koller; Guido Gerig; Ron Kikinis

doi:10.1016/S1361-8415(98)80009-1

Three-dimensional multi-scale line filter for segmentation and visualization of curvilinear structures in medical images

Yoshinobu Sato, Shin Nakajima, Nobuyuki Shiraga, Hideki Atsumi, Shigeyuki Yoshida, Thomas Koller, Guido Gerig, Ron Kikinis

Research output: Contribution to journal › Article › peer-review

Abstract

This paper describes a method for the enhancement of curvilinear structures such as vessels and bronchi in three-dimensional (3-D) medical images. A 3-D line enhancement filter is developed with the aim of discriminating line structures from other structures and recovering line structures of various widths. The 3-D line filter is based on a combination of the eigenvalues of the 3-D Hessian matrix. Multi-scale integration is formulated by taking the maximum among single-scale filter responses, and its characteristics are examined to derive criteria for the selection of parameters in the formulation. The resultant multi-scale line-filtered images provide significantly improved segmentation and visualization of curvilinear structures. The usefulness of the method is demonstrated by the segmentation and visualization of brain vessels from magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA), bronchi from a chest CT, and liver vessels (portal veins) from an abdominal CT.

Original language	English (US)
Pages (from-to)	143-168
Number of pages	26
Journal	Medical Image Analysis
Volume	2
Issue number	2
DOIs	https://doi.org/10.1016/S1361-8415(98)80009-1
State	Published - 1998

Keywords

3-D image analysis
Line detection
Line measure
Multi-scale integration
Vessel enhancement

ASJC Scopus subject areas

Radiological and Ultrasound Technology
Radiology Nuclear Medicine and imaging
Computer Vision and Pattern Recognition
Health Informatics
Computer Graphics and Computer-Aided Design

Access to Document

10.1016/S1361-8415(98)80009-1

Cite this

@article{7f8bfd3f3a394df6bf18dea4aff0aee5,

title = "Three-dimensional multi-scale line filter for segmentation and visualization of curvilinear structures in medical images",

abstract = "This paper describes a method for the enhancement of curvilinear structures such as vessels and bronchi in three-dimensional (3-D) medical images. A 3-D line enhancement filter is developed with the aim of discriminating line structures from other structures and recovering line structures of various widths. The 3-D line filter is based on a combination of the eigenvalues of the 3-D Hessian matrix. Multi-scale integration is formulated by taking the maximum among single-scale filter responses, and its characteristics are examined to derive criteria for the selection of parameters in the formulation. The resultant multi-scale line-filtered images provide significantly improved segmentation and visualization of curvilinear structures. The usefulness of the method is demonstrated by the segmentation and visualization of brain vessels from magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA), bronchi from a chest CT, and liver vessels (portal veins) from an abdominal CT.",

keywords = "3-D image analysis, Line detection, Line measure, Multi-scale integration, Vessel enhancement",

author = "Yoshinobu Sato and Shin Nakajima and Nobuyuki Shiraga and Hideki Atsumi and Shigeyuki Yoshida and Thomas Koller and Guido Gerig and Ron Kikinis",

note = "Funding Information: This work was partly supported by NIH grant P01 CA67165-02. The authors would like to thank Dr Don Johann of New York University Medical School for providing MR and DSA images of the brain, Dr Shinjiro Kawato of Mitsubishi Electric Corporation for providing the 3-D thinning program, and Dr Ahbir Bhalerao for making the volume rendering package {\textquoteleft}VolPack{\textquoteright} available at the Surgical Planning Laboratory, Harvard University and Brigham and Women{\textquoteright}s Hospital. The first author would also like to thank Dr Carl-Fredric Westin for encouraging him to use 3-D local intensity structures, Dr Nobuhiko Hata for helping him adapt the environment of the Surgical Planning Laboratory, and Professor Frank Jolesz of Harvard Medical School and Professor Shinichi Tamura of Osaka University Medical School for their continuous encouragement.",

year = "1998",

doi = "10.1016/S1361-8415(98)80009-1",

language = "English (US)",

volume = "2",

pages = "143--168",

journal = "Medical Image Analysis",

issn = "1361-8415",

publisher = "Elsevier B.V.",

number = "2",

}

TY - JOUR

T1 - Three-dimensional multi-scale line filter for segmentation and visualization of curvilinear structures in medical images

AU - Sato, Yoshinobu

AU - Nakajima, Shin

AU - Shiraga, Nobuyuki

AU - Atsumi, Hideki

AU - Yoshida, Shigeyuki

AU - Koller, Thomas

AU - Gerig, Guido

AU - Kikinis, Ron

N1 - Funding Information: This work was partly supported by NIH grant P01 CA67165-02. The authors would like to thank Dr Don Johann of New York University Medical School for providing MR and DSA images of the brain, Dr Shinjiro Kawato of Mitsubishi Electric Corporation for providing the 3-D thinning program, and Dr Ahbir Bhalerao for making the volume rendering package ‘VolPack’ available at the Surgical Planning Laboratory, Harvard University and Brigham and Women’s Hospital. The first author would also like to thank Dr Carl-Fredric Westin for encouraging him to use 3-D local intensity structures, Dr Nobuhiko Hata for helping him adapt the environment of the Surgical Planning Laboratory, and Professor Frank Jolesz of Harvard Medical School and Professor Shinichi Tamura of Osaka University Medical School for their continuous encouragement.

PY - 1998

Y1 - 1998

N2 - This paper describes a method for the enhancement of curvilinear structures such as vessels and bronchi in three-dimensional (3-D) medical images. A 3-D line enhancement filter is developed with the aim of discriminating line structures from other structures and recovering line structures of various widths. The 3-D line filter is based on a combination of the eigenvalues of the 3-D Hessian matrix. Multi-scale integration is formulated by taking the maximum among single-scale filter responses, and its characteristics are examined to derive criteria for the selection of parameters in the formulation. The resultant multi-scale line-filtered images provide significantly improved segmentation and visualization of curvilinear structures. The usefulness of the method is demonstrated by the segmentation and visualization of brain vessels from magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA), bronchi from a chest CT, and liver vessels (portal veins) from an abdominal CT.

AB - This paper describes a method for the enhancement of curvilinear structures such as vessels and bronchi in three-dimensional (3-D) medical images. A 3-D line enhancement filter is developed with the aim of discriminating line structures from other structures and recovering line structures of various widths. The 3-D line filter is based on a combination of the eigenvalues of the 3-D Hessian matrix. Multi-scale integration is formulated by taking the maximum among single-scale filter responses, and its characteristics are examined to derive criteria for the selection of parameters in the formulation. The resultant multi-scale line-filtered images provide significantly improved segmentation and visualization of curvilinear structures. The usefulness of the method is demonstrated by the segmentation and visualization of brain vessels from magnetic resonance imaging (MRI) and magnetic resonance angiography (MRA), bronchi from a chest CT, and liver vessels (portal veins) from an abdominal CT.

KW - 3-D image analysis

KW - Line detection

KW - Line measure

KW - Multi-scale integration

KW - Vessel enhancement

UR - http://www.scopus.com/inward/record.url?scp=0032081365&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032081365&partnerID=8YFLogxK

U2 - 10.1016/S1361-8415(98)80009-1

DO - 10.1016/S1361-8415(98)80009-1

M3 - Article

C2 - 10646760

AN - SCOPUS:0032081365

SN - 1361-8415

VL - 2

SP - 143

EP - 168

JO - Medical Image Analysis

JF - Medical Image Analysis

IS - 2

ER -

Three-dimensional multi-scale line filter for segmentation and visualization of curvilinear structures in medical images

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this