Nephroblastoma analysis in MRI images

Djibril Kaba, Nigel McFarlane, Feng Dong, Norbert Graf, Xujiong Ye

Research output: Contribution to journal › Article

Abstract

The annotation of the tumour from medical scans is a crucial step in nephroblastoma treatment. Therefore, an accurate and reliable segmentation method is needed to facilitate the evaluation and the treatments of the tumour. The proposed method serves this purpose by performing the segmentation of nephroblastoma in MRI scans. The segmentation is performed by adapting and a 2D free hand drawing tool to select a region of interest in the scan slices. Results from 24 patients show a mean root-mean-square error of 0.0481±0.0309, an average Dice coefficient of 0.9060±0.0549 and an average accuracy of 99.59% ±0.0039. Thus the proposed method demonstrated an effective agreement with manual annotations.
LanguageEnglish
Pages173-183
Number of pages11
JournalImage Analysis and Stereology
Volume38
Issue number2
DOIs
Publication statusPublished - 31 Jul 2019

Fingerprint

annotations
Magnetic resonance imaging
Tumors
tumors
root-mean-square errors
Mean square error
evaluation
coefficients
Magnetic Resonance Imaging

Keywords

  • continuous max-flow
  • graph segmentation
  • kernel induced space
  • MRI image
  • nephroblastoma
  • Wilms tumour

Cite this

Kaba, D., McFarlane, N., Dong, F., Graf, N., & Ye, X. (2019). Nephroblastoma analysis in MRI images. 38(2), 173-183. https://doi.org/10.5566/ias.2000
Kaba, Djibril ; McFarlane, Nigel ; Dong, Feng ; Graf, Norbert ; Ye, Xujiong. / Nephroblastoma analysis in MRI images. 2019 ; Vol. 38, No. 2. pp. 173-183.
@article{9d35c4c2490e4ca8a38cda0ce7e19401,
title = "Nephroblastoma analysis in MRI images",
abstract = "The annotation of the tumour from medical scans is a crucial step in nephroblastoma treatment. Therefore, an accurate and reliable segmentation method is needed to facilitate the evaluation and the treatments of the tumour. The proposed method serves this purpose by performing the segmentation of nephroblastoma in MRI scans. The segmentation is performed by adapting and a 2D free hand drawing tool to select a region of interest in the scan slices. Results from 24 patients show a mean root-mean-square error of 0.0481±0.0309, an average Dice coefficient of 0.9060±0.0549 and an average accuracy of 99.59{\%} ±0.0039. Thus the proposed method demonstrated an effective agreement with manual annotations.",
keywords = "continuous max-flow, graph segmentation, kernel induced space, MRI image, nephroblastoma, Wilms tumour",
author = "Djibril Kaba and Nigel McFarlane and Feng Dong and Norbert Graf and Xujiong Ye",
year = "2019",
month = "7",
day = "31",
doi = "10.5566/ias.2000",
language = "English",
volume = "38",
pages = "173--183",
number = "2",

}

Kaba, D, McFarlane, N, Dong, F, Graf, N & Ye, X 2019, 'Nephroblastoma analysis in MRI images' vol. 38, no. 2, pp. 173-183. https://doi.org/10.5566/ias.2000

Nephroblastoma analysis in MRI images. / Kaba, Djibril; McFarlane, Nigel; Dong, Feng; Graf, Norbert; Ye, Xujiong.

Vol. 38, No. 2, 31.07.2019, p. 173-183.

Research output: Contribution to journal › Article

TY - JOUR

T1 - Nephroblastoma analysis in MRI images

AU - Kaba, Djibril

AU - McFarlane, Nigel

AU - Dong, Feng

AU - Graf, Norbert

AU - Ye, Xujiong

PY - 2019/7/31

Y1 - 2019/7/31

N2 - The annotation of the tumour from medical scans is a crucial step in nephroblastoma treatment. Therefore, an accurate and reliable segmentation method is needed to facilitate the evaluation and the treatments of the tumour. The proposed method serves this purpose by performing the segmentation of nephroblastoma in MRI scans. The segmentation is performed by adapting and a 2D free hand drawing tool to select a region of interest in the scan slices. Results from 24 patients show a mean root-mean-square error of 0.0481±0.0309, an average Dice coefficient of 0.9060±0.0549 and an average accuracy of 99.59% ±0.0039. Thus the proposed method demonstrated an effective agreement with manual annotations.

AB - The annotation of the tumour from medical scans is a crucial step in nephroblastoma treatment. Therefore, an accurate and reliable segmentation method is needed to facilitate the evaluation and the treatments of the tumour. The proposed method serves this purpose by performing the segmentation of nephroblastoma in MRI scans. The segmentation is performed by adapting and a 2D free hand drawing tool to select a region of interest in the scan slices. Results from 24 patients show a mean root-mean-square error of 0.0481±0.0309, an average Dice coefficient of 0.9060±0.0549 and an average accuracy of 99.59% ±0.0039. Thus the proposed method demonstrated an effective agreement with manual annotations.

KW - continuous max-flow

KW - graph segmentation

KW - kernel induced space

KW - MRI image

KW - nephroblastoma

KW - Wilms tumour

U2 - 10.5566/ias.2000

DO - 10.5566/ias.2000

M3 - Article

VL - 38

SP - 173

EP - 183

IS - 2

ER -

Kaba D, McFarlane N, Dong F, Graf N, Ye X. Nephroblastoma analysis in MRI images. 2019 Jul 31;38(2):173-183. https://doi.org/10.5566/ias.2000

Access to Document