An adaptive method for fully automatic liver segmentation in medical MRI-images

Roaa G. Mohamed, Noha A. Seada, Salma Hamdy, Mostafa G. M. Mostafa

Research output: Contribution to journalArticle

19 Downloads (Pure)

Abstract

Despite the importance of the liver segmentation in the medical images for efficient noninvasive diagnosis, few studies found in the literatures for fully automated methods for liver segmentation in Magnetic Resonance Imaging (MRI) compared to that in Computed Tomography (CT) scans. Motivated by this, we propose an adaptive fully automatic liver segmentation method for MRI images based on thresholding and Bayesian classification. Bayesian classifications have proved to be highly robust to various image degradations. It only requires a small amount of training data to estimate the parameters necessary for classification, which is a huge advantage in medical applications. Furthermore, the Bayesian model is robust when large uncertainties are involved in medical image analysis problems. The proposed method is successfully tested on many MRI cases acquired from different patients, in various sizes. Experiments proved the robustness of the proposed liver automatic segmentation process even on data from different scanner types. The segmentation accuracy of the model has a mean Dice Similarity Coefficient (DSC) of 95.5% for MRI datasets.
Original languageEnglish
Pages (from-to)12-18
Number of pages7
JournalInternational Journal of Computer Applications
Volume179
Issue number4
DOIs
Publication statusPublished - 31 Dec 2017
Externally publishedYes

Fingerprint

Liver
Medical applications
Image analysis
Tomography
Degradation
Magnetic Resonance Imaging
Experiments

Cite this

G. Mohamed, Roaa ; A. Seada, Noha ; Hamdy, Salma ; Mostafa, Mostafa G. M. . / An adaptive method for fully automatic liver segmentation in medical MRI-images. In: International Journal of Computer Applications. 2017 ; Vol. 179, No. 4. pp. 12-18.
@article{e1b5c17807cb40a5bb4a7c9a3eb7dcb0,
title = "An adaptive method for fully automatic liver segmentation in medical MRI-images",
abstract = "Despite the importance of the liver segmentation in the medical images for efficient noninvasive diagnosis, few studies found in the literatures for fully automated methods for liver segmentation in Magnetic Resonance Imaging (MRI) compared to that in Computed Tomography (CT) scans. Motivated by this, we propose an adaptive fully automatic liver segmentation method for MRI images based on thresholding and Bayesian classification. Bayesian classifications have proved to be highly robust to various image degradations. It only requires a small amount of training data to estimate the parameters necessary for classification, which is a huge advantage in medical applications. Furthermore, the Bayesian model is robust when large uncertainties are involved in medical image analysis problems. The proposed method is successfully tested on many MRI cases acquired from different patients, in various sizes. Experiments proved the robustness of the proposed liver automatic segmentation process even on data from different scanner types. The segmentation accuracy of the model has a mean Dice Similarity Coefficient (DSC) of 95.5{\%} for MRI datasets.",
author = "{G. Mohamed}, Roaa and {A. Seada}, Noha and Salma Hamdy and Mostafa, {Mostafa G. M.}",
year = "2017",
month = "12",
day = "31",
doi = "10.5120/ijca2017915917",
language = "English",
volume = "179",
pages = "12--18",
journal = "International Journal of Computer Applications",
issn = "0975-8887",
publisher = "Foundation of Computer Science (FCS)",
number = "4",

}

An adaptive method for fully automatic liver segmentation in medical MRI-images. / G. Mohamed, Roaa; A. Seada, Noha; Hamdy, Salma; Mostafa, Mostafa G. M. .

In: International Journal of Computer Applications, Vol. 179, No. 4, 31.12.2017, p. 12-18.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An adaptive method for fully automatic liver segmentation in medical MRI-images

AU - G. Mohamed, Roaa

AU - A. Seada, Noha

AU - Hamdy, Salma

AU - Mostafa, Mostafa G. M.

PY - 2017/12/31

Y1 - 2017/12/31

N2 - Despite the importance of the liver segmentation in the medical images for efficient noninvasive diagnosis, few studies found in the literatures for fully automated methods for liver segmentation in Magnetic Resonance Imaging (MRI) compared to that in Computed Tomography (CT) scans. Motivated by this, we propose an adaptive fully automatic liver segmentation method for MRI images based on thresholding and Bayesian classification. Bayesian classifications have proved to be highly robust to various image degradations. It only requires a small amount of training data to estimate the parameters necessary for classification, which is a huge advantage in medical applications. Furthermore, the Bayesian model is robust when large uncertainties are involved in medical image analysis problems. The proposed method is successfully tested on many MRI cases acquired from different patients, in various sizes. Experiments proved the robustness of the proposed liver automatic segmentation process even on data from different scanner types. The segmentation accuracy of the model has a mean Dice Similarity Coefficient (DSC) of 95.5% for MRI datasets.

AB - Despite the importance of the liver segmentation in the medical images for efficient noninvasive diagnosis, few studies found in the literatures for fully automated methods for liver segmentation in Magnetic Resonance Imaging (MRI) compared to that in Computed Tomography (CT) scans. Motivated by this, we propose an adaptive fully automatic liver segmentation method for MRI images based on thresholding and Bayesian classification. Bayesian classifications have proved to be highly robust to various image degradations. It only requires a small amount of training data to estimate the parameters necessary for classification, which is a huge advantage in medical applications. Furthermore, the Bayesian model is robust when large uncertainties are involved in medical image analysis problems. The proposed method is successfully tested on many MRI cases acquired from different patients, in various sizes. Experiments proved the robustness of the proposed liver automatic segmentation process even on data from different scanner types. The segmentation accuracy of the model has a mean Dice Similarity Coefficient (DSC) of 95.5% for MRI datasets.

U2 - 10.5120/ijca2017915917

DO - 10.5120/ijca2017915917

M3 - Article

VL - 179

SP - 12

EP - 18

JO - International Journal of Computer Applications

JF - International Journal of Computer Applications

SN - 0975-8887

IS - 4

ER -