Spectral-based automatic labeling and refining of human cortical sulcal curves using expert-provided examples

Ilwoo Lyu, Jun Kyung Seong, Sung Yong Shin, Kiho Im, Jee Hoon Roh, Min Jeong Kim, Geon Ha Kim, Jong Hun Kim, Alan C. Evans, Duk L. Na, Jong Min Lee

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

We present a spectral-based method for automatically labeling and refining major sulcal curves of a human cerebral cortex. Given a set of input (unlabeled) sulcal curves automatically extracted from a cortical surface and a collection of expert-provided examples (labeled sulcal curves), our objective is to identify the input major sulcal curves and assign their neuroanatomical labels, and then refines these curves based on the expert-provided example data, without employing any atlas-based registration scheme as preprocessing. In order to construct the example data, neuroanatomists manually labeled a set of 24 major sulcal curves (12 each for the left and right hemispheres) for each individual subject according to a precise protocol. We collected 30 sets of such curves from 30 subjects. Given the raw input sulcal curve set of a subject, we choose the most similar example curve to each input curve in the set to label and refine the latter according to the former. We adapt a spectral matching algorithm to choose the example curve by exploiting the sulcal curve features and their relationship. The high dimensionality of sulcal curve data in spectral matching is addressed by using their multi-resolution representations, which greatly reduces time and space complexities. Our method provides consistent labeling and refining results even under high variability of cortical sulci across the subjects. Through experiments we show that the results are comparable in accuracy to those done manually. Most output curves exhibited accuracy values higher than 80%, and the mean accuracy values of the curves in the left and the right hemispheres were 84.69% and 84.58%, respectively.

Original languageEnglish
Pages (from-to)142-157
Number of pages16
JournalNeuroImage
Volume52
Issue number1
DOIs
Publication statusPublished - 2010 Aug 1
Externally publishedYes

Fingerprint

Atlases
Cerebral Cortex

Keywords

  • Labeling
  • Refining
  • Spectral matching
  • Sulcal curve
  • Sulcal variability

ASJC Scopus subject areas

  • Cognitive Neuroscience
  • Neurology

Cite this

Spectral-based automatic labeling and refining of human cortical sulcal curves using expert-provided examples. / Lyu, Ilwoo; Seong, Jun Kyung; Shin, Sung Yong; Im, Kiho; Roh, Jee Hoon; Kim, Min Jeong; Kim, Geon Ha; Kim, Jong Hun; Evans, Alan C.; Na, Duk L.; Lee, Jong Min.

In: NeuroImage, Vol. 52, No. 1, 01.08.2010, p. 142-157.

Research output: Contribution to journalArticle

Lyu, I, Seong, JK, Shin, SY, Im, K, Roh, JH, Kim, MJ, Kim, GH, Kim, JH, Evans, AC, Na, DL & Lee, JM 2010, 'Spectral-based automatic labeling and refining of human cortical sulcal curves using expert-provided examples', NeuroImage, vol. 52, no. 1, pp. 142-157. https://doi.org/10.1016/j.neuroimage.2010.03.076
Lyu, Ilwoo ; Seong, Jun Kyung ; Shin, Sung Yong ; Im, Kiho ; Roh, Jee Hoon ; Kim, Min Jeong ; Kim, Geon Ha ; Kim, Jong Hun ; Evans, Alan C. ; Na, Duk L. ; Lee, Jong Min. / Spectral-based automatic labeling and refining of human cortical sulcal curves using expert-provided examples. In: NeuroImage. 2010 ; Vol. 52, No. 1. pp. 142-157.
@article{c3fa8bb7d132434e907ad6c34925fe69,
title = "Spectral-based automatic labeling and refining of human cortical sulcal curves using expert-provided examples",
abstract = "We present a spectral-based method for automatically labeling and refining major sulcal curves of a human cerebral cortex. Given a set of input (unlabeled) sulcal curves automatically extracted from a cortical surface and a collection of expert-provided examples (labeled sulcal curves), our objective is to identify the input major sulcal curves and assign their neuroanatomical labels, and then refines these curves based on the expert-provided example data, without employing any atlas-based registration scheme as preprocessing. In order to construct the example data, neuroanatomists manually labeled a set of 24 major sulcal curves (12 each for the left and right hemispheres) for each individual subject according to a precise protocol. We collected 30 sets of such curves from 30 subjects. Given the raw input sulcal curve set of a subject, we choose the most similar example curve to each input curve in the set to label and refine the latter according to the former. We adapt a spectral matching algorithm to choose the example curve by exploiting the sulcal curve features and their relationship. The high dimensionality of sulcal curve data in spectral matching is addressed by using their multi-resolution representations, which greatly reduces time and space complexities. Our method provides consistent labeling and refining results even under high variability of cortical sulci across the subjects. Through experiments we show that the results are comparable in accuracy to those done manually. Most output curves exhibited accuracy values higher than 80{\%}, and the mean accuracy values of the curves in the left and the right hemispheres were 84.69{\%} and 84.58{\%}, respectively.",
keywords = "Labeling, Refining, Spectral matching, Sulcal curve, Sulcal variability",
author = "Ilwoo Lyu and Seong, {Jun Kyung} and Shin, {Sung Yong} and Kiho Im and Roh, {Jee Hoon} and Kim, {Min Jeong} and Kim, {Geon Ha} and Kim, {Jong Hun} and Evans, {Alan C.} and Na, {Duk L.} and Lee, {Jong Min}",
year = "2010",
month = "8",
day = "1",
doi = "10.1016/j.neuroimage.2010.03.076",
language = "English",
volume = "52",
pages = "142--157",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press Inc.",
number = "1",

}

TY - JOUR

T1 - Spectral-based automatic labeling and refining of human cortical sulcal curves using expert-provided examples

AU - Lyu, Ilwoo

AU - Seong, Jun Kyung

AU - Shin, Sung Yong

AU - Im, Kiho

AU - Roh, Jee Hoon

AU - Kim, Min Jeong

AU - Kim, Geon Ha

AU - Kim, Jong Hun

AU - Evans, Alan C.

AU - Na, Duk L.

AU - Lee, Jong Min

PY - 2010/8/1

Y1 - 2010/8/1

N2 - We present a spectral-based method for automatically labeling and refining major sulcal curves of a human cerebral cortex. Given a set of input (unlabeled) sulcal curves automatically extracted from a cortical surface and a collection of expert-provided examples (labeled sulcal curves), our objective is to identify the input major sulcal curves and assign their neuroanatomical labels, and then refines these curves based on the expert-provided example data, without employing any atlas-based registration scheme as preprocessing. In order to construct the example data, neuroanatomists manually labeled a set of 24 major sulcal curves (12 each for the left and right hemispheres) for each individual subject according to a precise protocol. We collected 30 sets of such curves from 30 subjects. Given the raw input sulcal curve set of a subject, we choose the most similar example curve to each input curve in the set to label and refine the latter according to the former. We adapt a spectral matching algorithm to choose the example curve by exploiting the sulcal curve features and their relationship. The high dimensionality of sulcal curve data in spectral matching is addressed by using their multi-resolution representations, which greatly reduces time and space complexities. Our method provides consistent labeling and refining results even under high variability of cortical sulci across the subjects. Through experiments we show that the results are comparable in accuracy to those done manually. Most output curves exhibited accuracy values higher than 80%, and the mean accuracy values of the curves in the left and the right hemispheres were 84.69% and 84.58%, respectively.

AB - We present a spectral-based method for automatically labeling and refining major sulcal curves of a human cerebral cortex. Given a set of input (unlabeled) sulcal curves automatically extracted from a cortical surface and a collection of expert-provided examples (labeled sulcal curves), our objective is to identify the input major sulcal curves and assign their neuroanatomical labels, and then refines these curves based on the expert-provided example data, without employing any atlas-based registration scheme as preprocessing. In order to construct the example data, neuroanatomists manually labeled a set of 24 major sulcal curves (12 each for the left and right hemispheres) for each individual subject according to a precise protocol. We collected 30 sets of such curves from 30 subjects. Given the raw input sulcal curve set of a subject, we choose the most similar example curve to each input curve in the set to label and refine the latter according to the former. We adapt a spectral matching algorithm to choose the example curve by exploiting the sulcal curve features and their relationship. The high dimensionality of sulcal curve data in spectral matching is addressed by using their multi-resolution representations, which greatly reduces time and space complexities. Our method provides consistent labeling and refining results even under high variability of cortical sulci across the subjects. Through experiments we show that the results are comparable in accuracy to those done manually. Most output curves exhibited accuracy values higher than 80%, and the mean accuracy values of the curves in the left and the right hemispheres were 84.69% and 84.58%, respectively.

KW - Labeling

KW - Refining

KW - Spectral matching

KW - Sulcal curve

KW - Sulcal variability

UR - http://www.scopus.com/inward/record.url?scp=77953287123&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953287123&partnerID=8YFLogxK

U2 - 10.1016/j.neuroimage.2010.03.076

DO - 10.1016/j.neuroimage.2010.03.076

M3 - Article

C2 - 20363334

AN - SCOPUS:77953287123

VL - 52

SP - 142

EP - 157

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 1

ER -