# A computational geometry approach to automated pulmonary fissure segmentation in CT examinations

Jiantao Pu, Joseph K. Leader, Bin Zheng, Friedrich D Knollmann, Carl Fuhrman, Frank C. Sciurba, David Gur

Research output: Contribution to journalArticle

80 Citations (Scopus)

### Abstract

Identification of pulmonary fissures, which form the boundaries between the lobes in the lungs, may be useful during clinical interpretation of computed tomography (CT) examinations to assess the early presence and characterization of manifestation of several lung diseases. Motivated by the unique nature of the surface shape of pulmonary fissures in 3-D space, we developed a new automated scheme using computational geometry methods to detect and segment fissures depicted on CT images. After a geometric modeling of the lung volume using the marching cubes algorithm, Laplacian smoothing is applied iteratively to enhance pulmonary fissures by depressing nonfissure structures while smoothing the surfaces of lung fissures. Next, an extended Gaussian image based procedure is used to locate the fissures in a statistical manner that approximates the fissures using a set of plane patches. This approach has several advantages such as independence of anatomic knowledge of the lung structure except the surface shape of fissures, limited sensitivity to other lung structures, and ease of implementation. The scheme performance was evaluated by two experienced thoracic radiologists using a set of 100 images (slices) randomly selected from 10 screening CT examinations. In this preliminary evaluation 98.7% and 94.9% of scheme segmented fissure voxels are within 2 mm of the fissures marked independently by two radiologists in the testing image dataset. Using the scheme detected fissures as reference, 89.4% and 90.1% of manually marked fissure points have distance ${\leq}2$ mm to the reference suggesting a possible under-segmentation of the scheme. The case-based root mean square (rms) distances (errors) between our scheme and the radiologist ranged from $1.48 \pm 0.92$ to $2.04 \pm 3.88$ mm. The discrepancy of fissure detection results between the automated scheme and either radiologist is smaller in this dataset than the interreader variability.

Original language English (US) 4703240 710-719 10 IEEE Transactions on Medical Imaging 28 5 https://doi.org/10.1109/TMI.2008.2010441 Published - May 2009 Yes

### Fingerprint

Computational geometry
Tomography
Lung
Pulmonary diseases
Screening
Testing
Lung Diseases
Thorax

### Keywords

• Extended Gaussian image (EGI)
• Pulmonary fissure
• Segmentation
• Shape analysis

### ASJC Scopus subject areas

• Electrical and Electronic Engineering
• Computer Science Applications
• Software

### Cite this

A computational geometry approach to automated pulmonary fissure segmentation in CT examinations. / Pu, Jiantao; Leader, Joseph K.; Zheng, Bin; Knollmann, Friedrich D; Fuhrman, Carl; Sciurba, Frank C.; Gur, David.

In: IEEE Transactions on Medical Imaging, Vol. 28, No. 5, 4703240, 05.2009, p. 710-719.

Research output: Contribution to journalArticle

Pu, Jiantao ; Leader, Joseph K. ; Zheng, Bin ; Knollmann, Friedrich D ; Fuhrman, Carl ; Sciurba, Frank C. ; Gur, David. / A computational geometry approach to automated pulmonary fissure segmentation in CT examinations. In: IEEE Transactions on Medical Imaging. 2009 ; Vol. 28, No. 5. pp. 710-719.
@article{e3fea026b9944c1fa18ca593bc86aaf5,
title = "A computational geometry approach to automated pulmonary fissure segmentation in CT examinations",
abstract = "Identification of pulmonary fissures, which form the boundaries between the lobes in the lungs, may be useful during clinical interpretation of computed tomography (CT) examinations to assess the early presence and characterization of manifestation of several lung diseases. Motivated by the unique nature of the surface shape of pulmonary fissures in 3-D space, we developed a new automated scheme using computational geometry methods to detect and segment fissures depicted on CT images. After a geometric modeling of the lung volume using the marching cubes algorithm, Laplacian smoothing is applied iteratively to enhance pulmonary fissures by depressing nonfissure structures while smoothing the surfaces of lung fissures. Next, an extended Gaussian image based procedure is used to locate the fissures in a statistical manner that approximates the fissures using a set of plane patches. This approach has several advantages such as independence of anatomic knowledge of the lung structure except the surface shape of fissures, limited sensitivity to other lung structures, and ease of implementation. The scheme performance was evaluated by two experienced thoracic radiologists using a set of 100 images (slices) randomly selected from 10 screening CT examinations. In this preliminary evaluation 98.7{\%} and 94.9{\%} of scheme segmented fissure voxels are within 2 mm of the fissures marked independently by two radiologists in the testing image dataset. Using the scheme detected fissures as reference, 89.4{\%} and 90.1{\%} of manually marked fissure points have distance ${\leq}2$ mm to the reference suggesting a possible under-segmentation of the scheme. The case-based root mean square (rms) distances (errors) between our scheme and the radiologist ranged from $1.48 \pm 0.92$ to $2.04 \pm 3.88$ mm. The discrepancy of fissure detection results between the automated scheme and either radiologist is smaller in this dataset than the interreader variability.",
keywords = "Computer-aided detection (CAD), Extended Gaussian image (EGI), Pulmonary fissure, Segmentation, Shape analysis",
author = "Jiantao Pu and Leader, {Joseph K.} and Bin Zheng and Knollmann, {Friedrich D} and Carl Fuhrman and Sciurba, {Frank C.} and David Gur",
year = "2009",
month = "5",
doi = "10.1109/TMI.2008.2010441",
language = "English (US)",
volume = "28",
pages = "710--719",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - A computational geometry approach to automated pulmonary fissure segmentation in CT examinations

AU - Pu, Jiantao

AU - Zheng, Bin

AU - Knollmann, Friedrich D

AU - Fuhrman, Carl

AU - Sciurba, Frank C.

AU - Gur, David

PY - 2009/5

Y1 - 2009/5

N2 - Identification of pulmonary fissures, which form the boundaries between the lobes in the lungs, may be useful during clinical interpretation of computed tomography (CT) examinations to assess the early presence and characterization of manifestation of several lung diseases. Motivated by the unique nature of the surface shape of pulmonary fissures in 3-D space, we developed a new automated scheme using computational geometry methods to detect and segment fissures depicted on CT images. After a geometric modeling of the lung volume using the marching cubes algorithm, Laplacian smoothing is applied iteratively to enhance pulmonary fissures by depressing nonfissure structures while smoothing the surfaces of lung fissures. Next, an extended Gaussian image based procedure is used to locate the fissures in a statistical manner that approximates the fissures using a set of plane patches. This approach has several advantages such as independence of anatomic knowledge of the lung structure except the surface shape of fissures, limited sensitivity to other lung structures, and ease of implementation. The scheme performance was evaluated by two experienced thoracic radiologists using a set of 100 images (slices) randomly selected from 10 screening CT examinations. In this preliminary evaluation 98.7% and 94.9% of scheme segmented fissure voxels are within 2 mm of the fissures marked independently by two radiologists in the testing image dataset. Using the scheme detected fissures as reference, 89.4% and 90.1% of manually marked fissure points have distance ${\leq}2$ mm to the reference suggesting a possible under-segmentation of the scheme. The case-based root mean square (rms) distances (errors) between our scheme and the radiologist ranged from $1.48 \pm 0.92$ to $2.04 \pm 3.88$ mm. The discrepancy of fissure detection results between the automated scheme and either radiologist is smaller in this dataset than the interreader variability.

AB - Identification of pulmonary fissures, which form the boundaries between the lobes in the lungs, may be useful during clinical interpretation of computed tomography (CT) examinations to assess the early presence and characterization of manifestation of several lung diseases. Motivated by the unique nature of the surface shape of pulmonary fissures in 3-D space, we developed a new automated scheme using computational geometry methods to detect and segment fissures depicted on CT images. After a geometric modeling of the lung volume using the marching cubes algorithm, Laplacian smoothing is applied iteratively to enhance pulmonary fissures by depressing nonfissure structures while smoothing the surfaces of lung fissures. Next, an extended Gaussian image based procedure is used to locate the fissures in a statistical manner that approximates the fissures using a set of plane patches. This approach has several advantages such as independence of anatomic knowledge of the lung structure except the surface shape of fissures, limited sensitivity to other lung structures, and ease of implementation. The scheme performance was evaluated by two experienced thoracic radiologists using a set of 100 images (slices) randomly selected from 10 screening CT examinations. In this preliminary evaluation 98.7% and 94.9% of scheme segmented fissure voxels are within 2 mm of the fissures marked independently by two radiologists in the testing image dataset. Using the scheme detected fissures as reference, 89.4% and 90.1% of manually marked fissure points have distance ${\leq}2$ mm to the reference suggesting a possible under-segmentation of the scheme. The case-based root mean square (rms) distances (errors) between our scheme and the radiologist ranged from $1.48 \pm 0.92$ to $2.04 \pm 3.88$ mm. The discrepancy of fissure detection results between the automated scheme and either radiologist is smaller in this dataset than the interreader variability.

KW - Extended Gaussian image (EGI)

KW - Pulmonary fissure

KW - Segmentation

KW - Shape analysis

UR - http://www.scopus.com/inward/record.url?scp=65549169642&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=65549169642&partnerID=8YFLogxK

U2 - 10.1109/TMI.2008.2010441

DO - 10.1109/TMI.2008.2010441

M3 - Article

C2 - 19272987

AN - SCOPUS:65549169642

VL - 28

SP - 710

EP - 719

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 5

M1 - 4703240

ER -