The challenge of cerebral magnetic resonance imaging in neonates: A new method using mathematical morphology for the segmentation of structures including diffuse excessive high signal intensities.

Yongchao Xu,Catherine Adamsbaum,Alessio Virzì,Héléne Urien,Isabelle Bloch,Thierry Géraud,Baptiste Morel,Élodie Puybareau,Sonia Dahdouh

doi:10.1016/j.media.2018.05.003

Abstract

Preterm birth is a multifactorial condition associated with increased morbidity and mortality. Diffuse excessive high signal intensity (DEHSI) has been recently described on T2-weighted MR sequences in this population and thought to be associated with neuropathologies. To date, no robust and reproducible method to assess the presence of white matter hyperintensities has been developed, perhaps explaining the current controversy over their prognostic value. The aim of this paper is to propose a new semi-automated framework to detect DEHSI on neonatal brain MR images having a particular pattern due to the physiological lack of complete myelination of the white matter. A novel method for semi- automatic segmentation of neonatal brain structures and DEHSI, based on mathematical morphology and on max-tree representations of the images is thus described. It is a mandatory first step to identify and clinically assess homogeneous cohorts of neonates for DEHSI and/or volume of any other segmented structures. Implemented in a user-friendly interface, the method makes it straightforward to select relevant markers of structures to be segmented, and if needed, apply eventually manual corrections. This method responds to the increasing need for providing medical experts with semi-automatic tools for image analysis, and overcomes the limitations of visual analysis alone, prone to subjectivity and variability. Experimental results demonstrate that the method is accurate, with excellent reproducibility and with very few manual corrections needed. Although the method was intended initially for images acquired at 1.5T, which corresponds to the usual clinical practice, preliminary results on images acquired at 3T suggest that the proposed approach can be generalized.

Highlights

Input Image Ground truth Our segmentationThanks to progress in neonatology, more than 85% of premature newborns survive [50]
A reproducible automatic segmentation of Diffuse excessive high signal intensity (DEHSI) may benefit from the segmentation of newborn brain MR images into different brain tissues, i.e., cortical gray matter (CoGM), basal ganglia and thalami (BGT), white matter (WM), ventricles (Vent), and cerebrospinal fluid (CSF)
We propose a complete pipeline dedicated to 1.5T images for segmenting different neonatal brain tissues and white matter hyperintensities usually observed in supratentorial2 slices, extending our preliminary work for some tissues in [34]

Summary

Introduction

Input Image Ground truth Our segmentationThanks to progress in neonatology, more than 85% of premature newborns survive [50]. Automated quantification of the cortical folding in a population of preterms, newborns and infants has been investigated in [16], showing promising results. This highlights the need for robust knowledge of normal versus pathological patterns in terms of volume, morphology, and signal intensities. A reproducible automatic (or semiautomatic) segmentation of DEHSI may benefit from the segmentation of newborn brain MR images into different brain tissues, i.e., cortical gray matter (CoGM), basal ganglia and thalami (BGT), white matter (WM), ventricles (Vent), and cerebrospinal fluid (CSF). The segmentation of newborn brain MR images provides new, quantitative information [11, 46] about the maturation of the brain: in particular the gyration process [17] and the myelination process [13, 55]

Objectives

Methods

Results

Discussion

Conclusion