A Generative Probabilistic Model and Discriminative Extensions for Brain Lesion Segmentation— With Application to Tumor and Stroke

Abstract

We introduce a generative probabilistic model for segmentation of brain lesions in multi-dimensional images that generalizes the EM segmenter, a common approach for modelling brain images using Gaussian mixtures and a probabilistic tissue atlas that employs expectation-maximization (EM) to estimate the label map for a new image. Our model augments the probabilistic atlas of the healthy tissues with a latent atlas of the lesion. We derive an estimation algorithm with closed-form EM update equations. The method extracts a latent atlas prior distribution and the lesion posterior distributions jointly from the image data. It delineates lesion areas individually in each channel, allowing for differences in lesion appearance across modalities, an important feature of many brain tumor imaging sequences. We also propose discriminative model extensions to map the output of the generative model to arbitrary labels with semantic and biological meaning, such as “tumor core” or “fluid-filled structure”, but without a one-to-one correspondence to the hypo-or hyper-intense lesion areas identified by the generative model. We test the approach in two image sets: the publicly available BRATS set of glioma patient scans, and multimodal brain images of patients with acute and subacute ischemic stroke. We find the generative model that has been designed for tumor lesions to generalize well to stroke images, and the generative-discriminative model to be one of the top ranking methods in the BRATS evaluation.

Automated Segmentation of Hyperintense Regions in FLAIR MRI Using Deep Learning

“…Recently, Menze et al (8) proposed an algorithm based on a generative probabilistic model aimed at tumor and stroke region segmentations. Their work reported a Dice index mean value of 0.73 ± 0.13 when aiming to segment the complete FLAIR lesion, with a 0.86 ± 0.06 interobserver variability.…”

“…The brain tumor segmentation algorithms commonly described in the literature usually exploit classical image analysis techniques or pattern recognition techniques (6–8) with the more recent approaches using deep convolutional neural networks (9–16). …”

A Crossbred Approach for Effective Brain Stroke Lesion Segmentation

“…The clear image of the affected region of the brain image is not provided by the above techniques, but, for the early identification of the brain infection, the MRI is more efficient than the CT [10]. The brain stroke detection techniques consolidate many methods such as preprocessing, segmentation, feature extraction and classification [11][12][13]. The researchers proposed many techniques for the segmentation and classification of images.…”

Automatic selection of localized region-based active contour models using image content analysis applied to brain tumor segmentation

“…Many efforts were made in recent years to develop human-free intervention methods that can achieve similar results to those obtained by physicians. Numerous techniques have been used for this purpose, which can be grouped in 2 classes, i.e., supervised [1][2][3][4][5][6] and unsupervised methods [7][8][9][10][11]. In Refs.…”

“…[1,2], convolutional neural networks were used for learning and then they were applied to identify abnormal areas. Menze et al [3] and Cordier et al [6] employed atlas-based methods to detect brain lesions. By treating the tumor segmentation as a labeling problem [4], the local independent projection-based classification algorithm was used to classify each voxel into different classes.…”