Bayesian automated cortical segmentation for neonatal MRI

Zane Chou, Natacha Paquette, Bhavana Ganesh, Yalin Wang, Rafael Ceschin, Marvin D. Nelson, Luke Macyszyn, Bilwaj Gaonkar, Ashok Panigrahy, Natasha Lepore

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Several attempts have been made in the past few years to develop and implement an automated segmentation of neonatal brain structural MRI. However, accurate automated MRI segmentation remains challenging in this population because of the low signal-to-noise ratio, large partial volume effects and inter-individual anatomical variability of the neonatal brain. In this paper, we propose a learning method for segmenting the whole brain cortical grey matter on neonatal T2-weighted images. We trained our algorithm using a neonatal dataset composed of 3 fullterm and 4 preterm infants scanned at term equivalent age. Our segmentation pipeline combines the FAST algorithm from the FSL library software and a Bayesian segmentation approach to create a threshold matrix that minimizes the error of mislabeling brain tissue types. Our method shows promising results with our pilot training set. In both preterm and full-term neonates, automated Bayesian segmentation generates a smoother and more consistent parcellation compared to FAST, while successfully removing the subcortical structure and cleaning the edges of the cortical grey matter. This method show promising refinement of the FAST segmentation by considerably reducing manual input and editing required from the user, and further improving reliability and processing time of neonatal MR images. Further improvement will include a larger dataset of training images acquired from different manufacturers.

Original languageEnglish (US)
Title of host publication13th International Conference on Medical Information Processing and Analysis
PublisherSPIE
Volume10572
ISBN (Electronic)9781510616332
DOIs
StatePublished - Jan 1 2017
Externally publishedYes
Event13th International Conference on Medical Information Processing and Analysis, SIPAIM 2017 - San Andres Island, Colombia
Duration: Oct 5 2017Oct 7 2017

Other

Other13th International Conference on Medical Information Processing and Analysis, SIPAIM 2017
CountryColombia
CitySan Andres Island
Period10/5/1710/7/17

Fingerprint

Magnetic resonance imaging
brain
Brain
Segmentation
pilot training
editing
cleaning
learning
Cleaning
Signal to noise ratio
signal to noise ratios
education
Pipelines
Term
Tissue
computer programs
Large Data Sets
Fast Algorithm
thresholds
Refinement

Keywords

  • Brain tissue segmentation
  • Cortical grey matter (cGM)
  • Magnetic resonance imaging (MRI)
  • Neonatal brain
  • Prematurity
  • Unmyelinated white matter (uWM)

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Chou, Z., Paquette, N., Ganesh, B., Wang, Y., Ceschin, R., Nelson, M. D., ... Lepore, N. (2017). Bayesian automated cortical segmentation for neonatal MRI. In 13th International Conference on Medical Information Processing and Analysis (Vol. 10572). [105720R] SPIE. https://doi.org/10.1117/12.2285217

Bayesian automated cortical segmentation for neonatal MRI. / Chou, Zane; Paquette, Natacha; Ganesh, Bhavana; Wang, Yalin; Ceschin, Rafael; Nelson, Marvin D.; Macyszyn, Luke; Gaonkar, Bilwaj; Panigrahy, Ashok; Lepore, Natasha.

13th International Conference on Medical Information Processing and Analysis. Vol. 10572 SPIE, 2017. 105720R.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chou, Z, Paquette, N, Ganesh, B, Wang, Y, Ceschin, R, Nelson, MD, Macyszyn, L, Gaonkar, B, Panigrahy, A & Lepore, N 2017, Bayesian automated cortical segmentation for neonatal MRI. in 13th International Conference on Medical Information Processing and Analysis. vol. 10572, 105720R, SPIE, 13th International Conference on Medical Information Processing and Analysis, SIPAIM 2017, San Andres Island, Colombia, 10/5/17. https://doi.org/10.1117/12.2285217
Chou Z, Paquette N, Ganesh B, Wang Y, Ceschin R, Nelson MD et al. Bayesian automated cortical segmentation for neonatal MRI. In 13th International Conference on Medical Information Processing and Analysis. Vol. 10572. SPIE. 2017. 105720R https://doi.org/10.1117/12.2285217
Chou, Zane ; Paquette, Natacha ; Ganesh, Bhavana ; Wang, Yalin ; Ceschin, Rafael ; Nelson, Marvin D. ; Macyszyn, Luke ; Gaonkar, Bilwaj ; Panigrahy, Ashok ; Lepore, Natasha. / Bayesian automated cortical segmentation for neonatal MRI. 13th International Conference on Medical Information Processing and Analysis. Vol. 10572 SPIE, 2017.
@inproceedings{478f617eff6c4baeb696f8fe23cd5638,
title = "Bayesian automated cortical segmentation for neonatal MRI",
abstract = "Several attempts have been made in the past few years to develop and implement an automated segmentation of neonatal brain structural MRI. However, accurate automated MRI segmentation remains challenging in this population because of the low signal-to-noise ratio, large partial volume effects and inter-individual anatomical variability of the neonatal brain. In this paper, we propose a learning method for segmenting the whole brain cortical grey matter on neonatal T2-weighted images. We trained our algorithm using a neonatal dataset composed of 3 fullterm and 4 preterm infants scanned at term equivalent age. Our segmentation pipeline combines the FAST algorithm from the FSL library software and a Bayesian segmentation approach to create a threshold matrix that minimizes the error of mislabeling brain tissue types. Our method shows promising results with our pilot training set. In both preterm and full-term neonates, automated Bayesian segmentation generates a smoother and more consistent parcellation compared to FAST, while successfully removing the subcortical structure and cleaning the edges of the cortical grey matter. This method show promising refinement of the FAST segmentation by considerably reducing manual input and editing required from the user, and further improving reliability and processing time of neonatal MR images. Further improvement will include a larger dataset of training images acquired from different manufacturers.",
keywords = "Brain tissue segmentation, Cortical grey matter (cGM), Magnetic resonance imaging (MRI), Neonatal brain, Prematurity, Unmyelinated white matter (uWM)",
author = "Zane Chou and Natacha Paquette and Bhavana Ganesh and Yalin Wang and Rafael Ceschin and Nelson, {Marvin D.} and Luke Macyszyn and Bilwaj Gaonkar and Ashok Panigrahy and Natasha Lepore",
year = "2017",
month = "1",
day = "1",
doi = "10.1117/12.2285217",
language = "English (US)",
volume = "10572",
booktitle = "13th International Conference on Medical Information Processing and Analysis",
publisher = "SPIE",
address = "United States",

}

TY - GEN

T1 - Bayesian automated cortical segmentation for neonatal MRI

AU - Chou, Zane

AU - Paquette, Natacha

AU - Ganesh, Bhavana

AU - Wang, Yalin

AU - Ceschin, Rafael

AU - Nelson, Marvin D.

AU - Macyszyn, Luke

AU - Gaonkar, Bilwaj

AU - Panigrahy, Ashok

AU - Lepore, Natasha

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Several attempts have been made in the past few years to develop and implement an automated segmentation of neonatal brain structural MRI. However, accurate automated MRI segmentation remains challenging in this population because of the low signal-to-noise ratio, large partial volume effects and inter-individual anatomical variability of the neonatal brain. In this paper, we propose a learning method for segmenting the whole brain cortical grey matter on neonatal T2-weighted images. We trained our algorithm using a neonatal dataset composed of 3 fullterm and 4 preterm infants scanned at term equivalent age. Our segmentation pipeline combines the FAST algorithm from the FSL library software and a Bayesian segmentation approach to create a threshold matrix that minimizes the error of mislabeling brain tissue types. Our method shows promising results with our pilot training set. In both preterm and full-term neonates, automated Bayesian segmentation generates a smoother and more consistent parcellation compared to FAST, while successfully removing the subcortical structure and cleaning the edges of the cortical grey matter. This method show promising refinement of the FAST segmentation by considerably reducing manual input and editing required from the user, and further improving reliability and processing time of neonatal MR images. Further improvement will include a larger dataset of training images acquired from different manufacturers.

AB - Several attempts have been made in the past few years to develop and implement an automated segmentation of neonatal brain structural MRI. However, accurate automated MRI segmentation remains challenging in this population because of the low signal-to-noise ratio, large partial volume effects and inter-individual anatomical variability of the neonatal brain. In this paper, we propose a learning method for segmenting the whole brain cortical grey matter on neonatal T2-weighted images. We trained our algorithm using a neonatal dataset composed of 3 fullterm and 4 preterm infants scanned at term equivalent age. Our segmentation pipeline combines the FAST algorithm from the FSL library software and a Bayesian segmentation approach to create a threshold matrix that minimizes the error of mislabeling brain tissue types. Our method shows promising results with our pilot training set. In both preterm and full-term neonates, automated Bayesian segmentation generates a smoother and more consistent parcellation compared to FAST, while successfully removing the subcortical structure and cleaning the edges of the cortical grey matter. This method show promising refinement of the FAST segmentation by considerably reducing manual input and editing required from the user, and further improving reliability and processing time of neonatal MR images. Further improvement will include a larger dataset of training images acquired from different manufacturers.

KW - Brain tissue segmentation

KW - Cortical grey matter (cGM)

KW - Magnetic resonance imaging (MRI)

KW - Neonatal brain

KW - Prematurity

KW - Unmyelinated white matter (uWM)

UR - http://www.scopus.com/inward/record.url?scp=85038423138&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85038423138&partnerID=8YFLogxK

U2 - 10.1117/12.2285217

DO - 10.1117/12.2285217

M3 - Conference contribution

VL - 10572

BT - 13th International Conference on Medical Information Processing and Analysis

PB - SPIE

ER -