Comparison of Acute and Chronic Traumatic Brain Injury Using Semi-Automatic Multimodal Segmentation of MR Volumes

Laboratory of Neuro Imaging, University of California, Los Angeles, California 90095, USA.
Journal of neurotrauma (Impact Factor: 3.71). 07/2011; 28(11):2287-306. DOI: 10.1089/neu.2011.1920
Source: PubMed


Although neuroimaging is essential for prompt and proper management of traumatic brain injury (TBI), there is a regrettable and acute lack of robust methods for the visualization and assessment of TBI pathophysiology, especially for of the purpose of improving clinical outcome metrics. Until now, the application of automatic segmentation algorithms to TBI in a clinical setting has remained an elusive goal because existing methods have, for the most part, been insufficiently robust to faithfully capture TBI-related changes in brain anatomy. This article introduces and illustrates the combined use of multimodal TBI segmentation and time point comparison using 3D Slicer, a widely-used software environment whose TBI data processing solutions are openly available. For three representative TBI cases, semi-automatic tissue classification and 3D model generation are performed to perform intra-patient time point comparison of TBI using multimodal volumetrics and clinical atrophy measures. Identification and quantitative assessment of extra- and intra-cortical bleeding, lesions, edema, and diffuse axonal injury are demonstrated. The proposed tools allow cross-correlation of multimodal metrics from structural imaging (e.g., structural volume, atrophy measurements) with clinical outcome variables and other potential factors predictive of recovery. In addition, the workflows described are suitable for TBI clinical practice and patient monitoring, particularly for assessing damage extent and for the measurement of neuroanatomical change over time. With knowledge of general location, extent, and degree of change, such metrics can be associated with clinical measures and subsequently used to suggest viable treatment options.

Download full-text


Available from: Andrei Irimia
  • Source
    • "Tumor segmentation methods often borrow ideas from other brain tissue and other brain lesion segmentation methods that have achieved a considerable accuracy [8]. Brain lesions resulting from traumatic brain injuries [9], [10] and stroke [11], [12] are similar to glioma lesions in terms of size and multimodal intensity patterns, but have attracted little attention so far. Most automated algorithms for brain lesion segmentation rely on either generative or discriminative probabilistic models at the core of their processing pipeline. "
    [Show abstract] [Hide abstract]
    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 hypoor 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 generativediscriminative model to be one of the top ranking methods in the BRATS evaluation.
    Preview · Article · Nov 2015 · IEEE Transactions on Medical Imaging
  • Source
    • "Both studies (Strangman et al., 2010; Warner et al., 2010b) identified several structures, including the thalamus and hippocampus that are particularly affected by TBI and are of significant value when predicting clinical outcome. The automatic structural segmentation of MR brain scans of TBI patients remains, however, a difficult endeavour as most existing methods lack robustness towards TBI-related changes in anatomy (Irimia et al., 2011, 2012). In the acute phase contusions, the presence of blood, hydrocephalus and/or oedema can greatly affect the ability to accurately segment a brain. "
    [Show abstract] [Hide abstract]
    ABSTRACT: We propose a framework for the robust and fully-automatic segmentation of magnetic resonance (MR) brain images called "Multi-Atlas Label Propagation with Expectation-Maximisation based refinement" (MALP-EM). The presented approach is based on a robust registration approach (MAPER), highly performant label fusion (joint label fusion) and intensity-based label refinement using EM. We further adapt this framework to be applicable for the segmentation of brain images with gross changes in anatomy. We propose to account for consistent registration errors by relaxing anatomical priors obtained by multi-atlas propagation and a weighting scheme to locally combine anatomical atlas priors and intensity-refined posterior probabilities. The method is evaluated on a benchmark dataset used in a recent MICCAI segmentation challenge. In this context we show that MALP-EM is competitive for the segmentation of MR brain scans of healthy adults when compared to state-of-the-art automatic labelling techniques. To demonstrate the versatility of the proposed approach, we employed MALP-EM to segment 125 MR brain images into 134 regions from subjects who had sustained traumatic brain injury (TBI). We employ a protocol to assess segmentation quality if no manual reference labels are available. Based on this protocol, three independent, blinded raters confirmed on 13 MR brain scans with pathology that MALP-EM is superior to established label fusion techniques. We visually confirm the robustness of our segmentation approach on the full cohort and investigate the potential of derived symmetry-based imaging biomarkers that correlate with and predict clinically relevant variables in TBI such as the Marshall Classification (MC) or Glasgow Outcome Score (GOS). Specifically, we show that we are able to stratify TBI patients with favourable outcomes from non-favourable outcomes with 64.7% accuracy using acute-phase MR images and 66.8% accuracy using follow-up MR images. Furthermore, we are able to differentiate subjects with the presence of a mass lesion or midline shift from those with diffuse brain injury with 76.0% accuracy. The thalamus, putamen, pallidum and hippocampus are particularly affected. Their involvement predicts TBI disease progression.
    Full-text · Article · Feb 2015 · Medical Image Analysis
  • Source
    • "The WM model is translucent in each brain view to facilitate the visibility of anatomic details obviated in the MR volume slice displayed. See Irimia et al. (2011) for a detailed description of the neuroinformatics methodology used to generate these visualizations. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Throughout the past few decades, the ability to treat and rehabilitate traumatic brain injury (TBI) patients has become critically reliant upon the use of neuroimaging to acquire adequate knowledge of injury-related effects upon brain function and recovery. As a result, the need for TBI neuroimaging analysis methods has increased in recent years due to the recognition that spatiotemporal computational analyses of TBI evolution are useful for capturing the effects of TBI dynamics. At the same time, however, the advent of such methods has brought about the need to analyze, manage, and integrate TBI neuroimaging data using informatically inspired approaches which can take full advantage of their large dimensionality and informational complexity. Given this perspective, we here discuss the neuroinformatics challenges for TBI neuroimaging analysis in the context of structural, connectivity, and functional paradigms. Within each of these, the availability of a wide range of neuroimaging modalities can be leveraged to fully understand the heterogeneity of TBI pathology; consequently, large-scale computer hardware resources and next-generation processing software are often required for efficient data storage, management, and analysis of TBI neuroimaging data. However, each of these paradigms poses challenges in the context of informatics such that the ability to address them is critical for augmenting current capabilities to perform neuroimaging analysis of TBI and to improve therapeutic efficacy.
    Full-text · Article · Feb 2014 · Frontiers in Neuroinformatics
Show more