A Review of Publicly Available Automatic Brain Segmentation Methodologies, Machine Learning Models, Recent Advancements, and Their Comparison

Abstract

Background: The noninvasive study of the structure and functions of the brain using neuroimaging techniques is increasingly being used for its clinical and research perspective. The morphological and volumetric changes in several regions and structures of brains are associated with the prognosis of neurological disorders such as Alzheimer’s disease, epilepsy, schizophrenia, etc. and the early identification of such changes can have huge clinical significance. The accurate segmentation of three-dimensional brain magnetic resonance images into tissue types (i.e., grey matter, white matter, cerebrospinal fluid) and brain structures, thus, has huge importance as they can act as early biomarkers. The manual segmentation though considered the “gold standard” is time-consuming, subjective, and not suitable for bigger neuroimaging studies. Several automatic segmentation tools and algorithms have been developed over the years; the machine learning models particularly those using deep convolutional neural network (CNN) architecture are increasingly being applied to improve the accuracy of automatic methods. Purpose: The purpose of the study is to understand the current and emerging state of automatic segmentation tools, their comparison, machine learning models, their reliability, and shortcomings with an intent to focus on the development of improved methods and algorithms. Methods: The study focuses on the review of publicly available neuroimaging tools, their comparison, and emerging machine learning models particularly those based on CNN architecture developed and published during the last five years. Conclusion: Several software tools developed by various research groups and made publicly available for automatic segmentation of the brain show variability in their results in several comparison studies and have not attained the level of reliability required for clinical studies. The machine learning models particularly three dimensional fully convolutional network models can provide a robust and efficient alternative with relation to publicly available tools but perform poorly on unseen datasets. The challenges related to training, computation cost, reproducibility, and validation across distinct scanning modalities for machine learning models need to be addressed.

A Quantitative Imaging Biomarker Supporting Radiological Assessment of Hippocampal Sclerosis Derived From Deep Learning-Based Segmentation of T1w-MRI

“…Deep learning-based methods have been shown to outperform atlas-based methods for neuroanatomy segmentation (17)(18)(19)(20). Convolutional neural network (CNN) architectures have also been successfully used specifically to segment the hippocampus (21)(22)(23)(24).…”

3D modeling and printing for complex biventricular repair of double outlet right ventricle

“…However, due to the considerable heterogeneity, both in the patient-specific anatomy but also in the image data, this approach can be considered as beneficial in the current stage of research, as it allows to accurately assess the relevant aspects of the anatomy and mitigate imaging artifacts via the experience of the heart team. The advance of machine learning-based algorithms for image processing was already successful in providing automated tools for reconstruction of different anatomical structures ( 26 – 28 ), including the heart ( 27 ), albeit mostly for those with normal physiology. For these approaches to be applicable for congenital heart defects, a joint effort by multiple centers is most likely required to provide the necessary case numbers and sufficiently heterogenous image data for the approach to be widely applicable.…”

Beam orientation optimization for coherent X-ray scattering from distributed deep targets

“…A variety of methodologies and models have been developed for automatic segmentation of specific brain regions (see, for instance, Ref. [14]). Modeling the scattering signal of temporal and frontal target regions in a 2D scout image resulted in beam effectiveness that varied significantly, up to 2 orders of magnitude.…”