Segmentation of 3-D High-Frequency Ultrasound Images of Human Lymph Nodes Using Graph Cut with Energy Functional Adapted to Local Intensity Distribution

Jen Wei Kuo, Jonathan Mamou, Yao Wang, Emi Saegusa-Beecroft, Junji MacHi, Ernest J. Feleppa

Research output: Contribution to journalArticle

Abstract

Previous studies by our group have shown that 3-D high-frequency quantitative ultrasound (QUS) methods have the potential to differentiate metastatic lymph nodes (LNs) from cancer-free LNs dissected from human cancer patients. To successfully perform these methods inside the LN parenchyma (LNP), an automatic segmentation method is highly desired to exclude the surrounding thin layer of fat from QUS processing and accurately correct for ultrasound attenuation. In high-frequency ultrasound images of LNs, the intensity distribution of LNP and fat varies spatially because of acoustic attenuation and focusing effects. Thus, the intensity contrast between two object regions (e.g., LNP and fat) is also spatially varying. In our previous work, nested graph cut (GC) demonstrated its ability to simultaneously segment LNP, fat, and the outer phosphate-buffered saline bath even when some boundaries are lost because of acoustic attenuation and focusing effects. This paper describes a novel approach called GC with locally adaptive energy to further deal with spatially varying distributions of LNP and fat caused by inhomogeneous acoustic attenuation. The proposed method achieved Dice similarity coefficients of 0.937±0.035 when compared with expert manual segmentation on a representative data set consisting of 115 3-D LN images obtained from colorectal cancer patients.

Original languageEnglish (US)
Article number8006247
Pages (from-to)1514-1525
Number of pages12
JournalIEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
Volume64
Issue number10
DOIs
StatePublished - Oct 1 2017

Fingerprint

lymphatic system
fats
Oils and fats
acoustic attenuation
Ultrasonics
Acoustics
cancer
energy
baths
phosphates
Phosphates
attenuation
coefficients
Processing

Keywords

  • Graph cuts (GCs)
  • lymph node (LN)
  • segmentation
  • ultrasound

ASJC Scopus subject areas

  • Instrumentation
  • Acoustics and Ultrasonics
  • Electrical and Electronic Engineering

Cite this

Segmentation of 3-D High-Frequency Ultrasound Images of Human Lymph Nodes Using Graph Cut with Energy Functional Adapted to Local Intensity Distribution. / Kuo, Jen Wei; Mamou, Jonathan; Wang, Yao; Saegusa-Beecroft, Emi; MacHi, Junji; Feleppa, Ernest J.

In: IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, Vol. 64, No. 10, 8006247, 01.10.2017, p. 1514-1525.

Research output: Contribution to journalArticle

@article{73d3c95d0a484395b3153710e394d90b,
title = "Segmentation of 3-D High-Frequency Ultrasound Images of Human Lymph Nodes Using Graph Cut with Energy Functional Adapted to Local Intensity Distribution",
abstract = "Previous studies by our group have shown that 3-D high-frequency quantitative ultrasound (QUS) methods have the potential to differentiate metastatic lymph nodes (LNs) from cancer-free LNs dissected from human cancer patients. To successfully perform these methods inside the LN parenchyma (LNP), an automatic segmentation method is highly desired to exclude the surrounding thin layer of fat from QUS processing and accurately correct for ultrasound attenuation. In high-frequency ultrasound images of LNs, the intensity distribution of LNP and fat varies spatially because of acoustic attenuation and focusing effects. Thus, the intensity contrast between two object regions (e.g., LNP and fat) is also spatially varying. In our previous work, nested graph cut (GC) demonstrated its ability to simultaneously segment LNP, fat, and the outer phosphate-buffered saline bath even when some boundaries are lost because of acoustic attenuation and focusing effects. This paper describes a novel approach called GC with locally adaptive energy to further deal with spatially varying distributions of LNP and fat caused by inhomogeneous acoustic attenuation. The proposed method achieved Dice similarity coefficients of 0.937±0.035 when compared with expert manual segmentation on a representative data set consisting of 115 3-D LN images obtained from colorectal cancer patients.",
keywords = "Graph cuts (GCs), lymph node (LN), segmentation, ultrasound",
author = "Kuo, {Jen Wei} and Jonathan Mamou and Yao Wang and Emi Saegusa-Beecroft and Junji MacHi and Feleppa, {Ernest J.}",
year = "2017",
month = "10",
day = "1",
doi = "10.1109/TUFFC.2017.2737948",
language = "English (US)",
volume = "64",
pages = "1514--1525",
journal = "IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control",
issn = "0885-3010",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "10",

}

TY - JOUR

T1 - Segmentation of 3-D High-Frequency Ultrasound Images of Human Lymph Nodes Using Graph Cut with Energy Functional Adapted to Local Intensity Distribution

AU - Kuo, Jen Wei

AU - Mamou, Jonathan

AU - Wang, Yao

AU - Saegusa-Beecroft, Emi

AU - MacHi, Junji

AU - Feleppa, Ernest J.

PY - 2017/10/1

Y1 - 2017/10/1

N2 - Previous studies by our group have shown that 3-D high-frequency quantitative ultrasound (QUS) methods have the potential to differentiate metastatic lymph nodes (LNs) from cancer-free LNs dissected from human cancer patients. To successfully perform these methods inside the LN parenchyma (LNP), an automatic segmentation method is highly desired to exclude the surrounding thin layer of fat from QUS processing and accurately correct for ultrasound attenuation. In high-frequency ultrasound images of LNs, the intensity distribution of LNP and fat varies spatially because of acoustic attenuation and focusing effects. Thus, the intensity contrast between two object regions (e.g., LNP and fat) is also spatially varying. In our previous work, nested graph cut (GC) demonstrated its ability to simultaneously segment LNP, fat, and the outer phosphate-buffered saline bath even when some boundaries are lost because of acoustic attenuation and focusing effects. This paper describes a novel approach called GC with locally adaptive energy to further deal with spatially varying distributions of LNP and fat caused by inhomogeneous acoustic attenuation. The proposed method achieved Dice similarity coefficients of 0.937±0.035 when compared with expert manual segmentation on a representative data set consisting of 115 3-D LN images obtained from colorectal cancer patients.

AB - Previous studies by our group have shown that 3-D high-frequency quantitative ultrasound (QUS) methods have the potential to differentiate metastatic lymph nodes (LNs) from cancer-free LNs dissected from human cancer patients. To successfully perform these methods inside the LN parenchyma (LNP), an automatic segmentation method is highly desired to exclude the surrounding thin layer of fat from QUS processing and accurately correct for ultrasound attenuation. In high-frequency ultrasound images of LNs, the intensity distribution of LNP and fat varies spatially because of acoustic attenuation and focusing effects. Thus, the intensity contrast between two object regions (e.g., LNP and fat) is also spatially varying. In our previous work, nested graph cut (GC) demonstrated its ability to simultaneously segment LNP, fat, and the outer phosphate-buffered saline bath even when some boundaries are lost because of acoustic attenuation and focusing effects. This paper describes a novel approach called GC with locally adaptive energy to further deal with spatially varying distributions of LNP and fat caused by inhomogeneous acoustic attenuation. The proposed method achieved Dice similarity coefficients of 0.937±0.035 when compared with expert manual segmentation on a representative data set consisting of 115 3-D LN images obtained from colorectal cancer patients.

KW - Graph cuts (GCs)

KW - lymph node (LN)

KW - segmentation

KW - ultrasound

UR - http://www.scopus.com/inward/record.url?scp=85029009722&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029009722&partnerID=8YFLogxK

U2 - 10.1109/TUFFC.2017.2737948

DO - 10.1109/TUFFC.2017.2737948

M3 - Article

C2 - 28796617

AN - SCOPUS:85029009722

VL - 64

SP - 1514

EP - 1525

JO - IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control

JF - IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control

SN - 0885-3010

IS - 10

M1 - 8006247

ER -