Detail publikace

Automatic Extraction of Pathological Area in 2D MR Brain Scan

Originální název

Automatic Extraction of Pathological Area in 2D MR Brain Scan

Anglický název

Automatic Extraction of Pathological Area in 2D MR Brain Scan

Jazyk

en

Originální abstrakt

The aim of this work is to propose the fully automated pathological area extraction. The proposed method is based on multi-resolution symmetry analysis and automatic thresholding. The multi-contrast MRI (T2 and FLAIR images) is used. Since the method is based on symmetry, it works for both axial and coronal planes. In these both planes of healthy brain, the approximate left-right symmetry exists and it is used as the prior knowledge for searching the approximate pathology location. The algorithm was tested on 357 axial and 443 coronal real images from publicly available BRATS databases containing 3D brain volumes afflicted by a brain tumor. The results were evaluated by Dice Coefficient (axial: 0.85, coronal: 0.82) and by Accuracy (axial: 0.96, coronal: 0.94).

Anglický abstrakt

The aim of this work is to propose the fully automated pathological area extraction. The proposed method is based on multi-resolution symmetry analysis and automatic thresholding. The multi-contrast MRI (T2 and FLAIR images) is used. Since the method is based on symmetry, it works for both axial and coronal planes. In these both planes of healthy brain, the approximate left-right symmetry exists and it is used as the prior knowledge for searching the approximate pathology location. The algorithm was tested on 357 axial and 443 coronal real images from publicly available BRATS databases containing 3D brain volumes afflicted by a brain tumor. The results were evaluated by Dice Coefficient (axial: 0.85, coronal: 0.82) and by Accuracy (axial: 0.96, coronal: 0.94).

BibTex


@inproceedings{BUT109170,
  author="Pavel {Dvořák} and Karel {Bartušek} and Eva {Gescheidtová}",
  title="Automatic Extraction of Pathological Area in 2D MR Brain Scan",
  annote="The aim of this work is to propose the fully automated pathological area extraction. The proposed method is based on multi-resolution symmetry analysis and automatic thresholding. The multi-contrast MRI (T2 and FLAIR images) is used. Since the method is based on symmetry, it works for both axial and coronal planes. In these both planes of healthy brain, the approximate left-right symmetry exists and it is used as the prior knowledge for searching the approximate pathology location. The algorithm was tested on 357 axial and 443 coronal real images from publicly available BRATS databases containing 3D brain volumes afflicted by a brain tumor. The results were evaluated by Dice Coefficient (axial: 0.85, coronal: 0.82) and by Accuracy (axial: 0.96, coronal: 0.94).",
  booktitle="PIERS 2014 Guangzhou Proceedings",
  chapter="109170",
  howpublished="online",
  year="2014",
  month="august",
  pages="1885--1889",
  type="conference paper"
}