A Robust Energy Minimization Algorithm for MS-Lesion Segmentation

Abstract

The detection of multiple sclerosis lesion is important for many neuroimaging studies. In this paper, a new automatic robust algorithm for lesion segmentation based on MR images is proposed. This method takes full advantage of the decomposition of MR images into the true image that characterizes a physical property of the tissues and the bias field that accounts for the intensity inhomogeneity. An energy function is defined in term of the property of true image and bias field. The energy minimization is proposed for seeking the optimal segmentation result of lesions and white matter. Then postprocessing operations is used to select the most plausible lesions in the obtained hyperintense signals. The experimental results show that our approach is effective and robust for the lesion segmentation.

A Bibliography of Multiple Sclerosis Lesions Detection Methods using Brain MRIs

“…To obtain the intensity normalization maps, Gaussian Mixture Model (GMM) estimation is used, and to estimate loss function, MLE is used. Gong et al [101] used robust energy minimization based approach to detect lesions by using FLAIR images.…”

“…Khayati et al [25] 2008 Bayesian FLAIR MSL Harmouche et al [98] 2015 MRF T1, T2, PD, FLAIR T1-HL, T2-HL Shiee et al [97] 2010 FCM T1, T2, PD WHL Harmouche et al [87] 2006 Bayesian + MRF T2-w, PD-w T2 HL Bricq et al [88] 2008 HMM + TLE T1-w, FLAIR MSL Prastawa et al [90] 2008 PDF + MCD T1-w, T2-w, FLAIR MSL At-Ali et al [91] 2005 TLE T1-w, T2-w PD-w T1-HL, T2-HL Gao et al [94] 2013 EnM T1-w, T2-w, FLAIR WML Gong et al [101] 2015 EnM FLAIR MSL Tomas-Fernandez et al [103] 2015 GMM T1, T2, FLAIR MSL Elliott et al [104] 2013 Bayesian + RF FLAIR MSL Gao et al [105] 2014 EnM + BFE T1-w, T2-w, FLAIR MSL Zangeneh et al [113] 2016 GMM + ANN T1-w, T2-w, FLAIR MSL Zhao et al [118] 2017 EnM T1-w, FLAIR MSL Supervised Roy et al [4] 2013 SVM T1-w, T2-w, FLAIR MSL Deshpande et al [128] 2015 ADL T1-w MPRAGE, T2-w, PD, FLAIR MSL, WM, GM Tissues Zhang et al [130] 2007 ANN + KNN T2-w MSL, WM Tissues Cabezas et al [131] 2013 Gentleboost Algorithm PD-w, T2-w HL Cabezas et al [132] 2014 Gentleboost Algorithm T1-w, T2-w, PD-w, FLAIR MSL Guizard et al [146] 2015 RMNMS T1-w, T2-w, FLAIR MSL Jesson et al [133] 2015 MRF + RF T1, T2, FLAIR GML Kuwazuru et al [134] 2012 ANN + levelSet method T1, T2, FLAIR MSL Abdullah et al [135] 2012 SVM T1, T2 MSL Khastavaneh et al [137] 2015 MTANN + FIS T1-w, T2-w, FLAIR HL Veronese et al [138] 2014 SVM DIR GML…”