An Automatic Segmentation of T2-FLAIR Multiple Sclerosis Lesions

Multiple sclerosis diagnosis and patient follow-up can be helped by an evaluation of the lesion load in MRI sequences. A lot of automatic methods to segment these lesions are available in the literature. The MICCAI workshop Multiple Sclerosis (MS) lesion segmentation Challenge 08 allows to test and compare these algorithms. This paper presents a method designed to detect hyperintense signal area on T2-FLAIR sequence and its results on the Challenge test data. The proposed algorithm uses only three conventional MRI sequences: T1, T2 and T2-FLAIR. First, images are cropped, spatially unbiased and skull-stripped. A segmentation of the brain into its different compartments is performed on the T1 and the T2 sequences. From these segmentations, a threshold for the T2-FLAIR sequence is automatically computed. Then postprocessing operations select the most plausible lesions in the obtained hyperintense signals. Average global result on the test data (80/100) is close to the inter-expert variability (90/100).

[1]  À. Rovira,et al.  Isolated demyelinating syndromes: comparison of different MR imaging criteria to predict conversion to clinically definite multiple sclerosis. , 2000, AJNR. American journal of neuroradiology.

[2]  Sébastien Ourselin,et al.  Block Matching: A General Framework to Improve Robustness of Rigid Registration of Medical Images , 2000, MICCAI.

[3]  Nicholas Ayache,et al.  Maximum Likelihood Estimation of the Bias Field in MR Brain Images: Investigating Different Modelings of the Imaging Process , 2001, MICCAI.

[4]  S. Reingold,et al.  Diagnostic criteria for multiple sclerosis: 2005 revisions to the “McDonald Criteria” , 2005, Annals of neurology.

[5]  Stephen M Smith,et al.  Fast robust automated brain extraction , 2002, Human brain mapping.

[6]  N. Ayache,et al.  Atlas-based automatic segmentation of MR images: validation study on the brainstem in radiotherapy context. , 2005, International journal of radiation oncology, biology, physics.

[7]  N. Ayache,et al.  An efficient locally affine framework for the smooth registration of anatomical structures , 2008, Medical Image Anal..

[8]  Arthur W. Toga,et al.  Skull-stripping magnetic resonance brain images using a model-based level set , 2006, NeuroImage.

[9]  Grégoire Malandain,et al.  Improved EM-Based Tissue Segmentation and Partial Volume Effect Quantification in Multi-sequence Brain MRI , 2004, MICCAI.

[10]  R Kikinis,et al.  Automatic identification of gray matter structures from MRI to improve the segmentation of white matter lesions. , 1995, Journal of image guided surgery.

[11]  P. Narayana,et al.  Compensation for surface coil sensitivity variation in magnetic resonance imaging. , 1988, Magnetic resonance imaging.

[12]  Richard M. Leahy,et al.  Surface-based labeling of cortical anatomy using a deformable atlas , 1997, IEEE Transactions on Medical Imaging.

[13]  Terry M. Peters,et al.  3D statistical neuroanatomical models from 305 MRI volumes , 1993, 1993 IEEE Conference Record Nuclear Science Symposium and Medical Imaging Conference.

[14]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[15]  A. Compston,et al.  Recommended diagnostic criteria for multiple sclerosis: Guidelines from the international panel on the diagnosis of multiple sclerosis , 2001, Annals of neurology.

[16]  Nicholas Ayache,et al.  Revue des approches de segmentation des lésions de sclérose en plaques dans les séquences conventionnelles IRM , 2009 .

[17]  D. Louis Collins,et al.  Model-based 3-D segmentation of multiple sclerosis lesions in magnetic resonance brain images , 1995, IEEE Trans. Medical Imaging.

[18]  R. Leahy,et al.  Magnetic Resonance Image Tissue Classification Using a Partial Volume Model , 2001, NeuroImage.