Review of automatic segmentation methods of multiple sclerosis white matter lesions on conventional magnetic resonance imaging

Supervised learning technique for the automated identification of white matter hyperintensities in traumatic brain injury

BACKGROUND

White matter hyperintensities (WMHs) are foci of abnormal signal intensity in white matter regions seen with magnetic resonance imaging (MRI). WMHs are associated with normal ageing and have shown prognostic value in neurological conditions such as traumatic brain injury (TBI). The impracticality of manually quantifying these lesions limits their clinical utility and motivates the utilization of machine learning techniques for automated segmentation workflows.

METHODS

This study develops a concatenated random forest framework with image features for segmenting WMHs in a TBI cohort. The framework is built upon the Advanced Normalization Tools (ANTs) and ANTsR toolkits. MR (3D FLAIR, T2- and T1-weighted) images from 24 service members and veterans scanned in the Chronic Effects of Neurotrauma Consortium's (CENC) observational study were acquired. Manual annotations were employed for both training and evaluation using a leave-one-out strategy. Performance measures include sensitivity, positive predictive value, [Formula: see text] score and relative volume difference.

RESULTS

Final average results were: sensitivity = 0.68 ± 0.38, positive predictive value = 0.51 ± 0.40, [Formula: see text] = 0.52 ± 0.36, relative volume difference = 43 ± 26%. In addition, three lesion size ranges are selected to illustrate the variation in performance with lesion size.

CONCLUSION

Paired with correlative outcome data, supervised learning methods may allow for identification of imaging features predictive of diagnosis and prognosis in individual TBI patients.

White matter hyperintensity and stroke lesion segmentation and differentiation using convolutional neural networks

White matter hyperintensities (WMH) are a feature of sporadic small vessel disease also frequently observed in magnetic resonance images (MRI) of healthy elderly subjects. The accurate assessment of WMH burden is of crucial importance for epidemiological studies to determine association between WMHs, cognitive and clinical data; their causes, and the effects of new treatments in randomized trials. The manual delineation of WMHs is a very tedious, costly and time consuming process, that needs to be carried out by an expert annotator (e.g. a trained image analyst or radiologist). The problem of WMH delineation is further complicated by the fact that other pathological features (i.e. stroke lesions) often also appear as hyperintense regions. Recently, several automated methods aiming to tackle the challenges of WMH segmentation have been proposed. Most of these methods have been specifically developed to segment WMH in MRI but cannot differentiate between WMHs and strokes. Other methods, capable of distinguishing between different pathologies in brain MRI, are not designed with simultaneous WMH and stroke segmentation in mind. Therefore, a task specific, reliable, fully automated method that can segment and differentiate between these two pathological manifestations on MRI has not yet been fully identified. In this work we propose to use a convolutional neural network (CNN) that is able to segment hyperintensities and differentiate between WMHs and stroke lesions. Specifically, we aim to distinguish between WMH pathologies from those caused by stroke lesions due to either cortical, large or small subcortical infarcts. The proposed fully convolutional CNN architecture, called uResNet, that comprised an analysis path, that gradually learns low and high level features, followed by a synthesis path, that gradually combines and up-samples the low and high level features into a class likelihood semantic segmentation. Quantitatively, the proposed CNN architecture is shown to outperform other well established and state-of-the-art algorithms in terms of overlap with manual expert annotations. Clinically, the extracted WMH volumes were found to correlate better with the Fazekas visual rating score than competing methods or the expert-annotated volumes. Additionally, a comparison of the associations found between clinical risk-factors and the WMH volumes generated by the proposed method, was found to be in line with the associations found with the expert-annotated volumes.

•

Robust, fully automatic white matter hyperintensity and stroke lesion segmentation and differentiation

•

A novel patch sampling strategy used during CNN training that avoids the introduction of erroneous locality assumptions

•

Improved segmentation accuracy in terms of Dice scores when compared to well established state-of-the-art methods

Dual-Sensitivity Multiple Sclerosis Lesion and CSF Segmentation for Multichannel 3T Brain MRI

BACKGROUND AND PURPOSE

A pipeline for fully automated segmentation of 3T brain MRI scans in multiple sclerosis (MS) is presented. This 3T morphometry (3TM) pipeline provides indicators of MS disease progression from multichannel datasets with high‐resolution 3‐dimensional T1‐weighted, T2‐weighted, and fluid‐attenuated inversion‐recovery (FLAIR) contrast. 3TM segments white (WM) and gray matter (GM) and cerebrospinal fluid (CSF) to assess atrophy and provides WM lesion (WML) volume.

METHODS

To address nonuniform distribution of noise/contrast (eg, posterior fossa in 3D‐FLAIR) of 3T magnetic resonance imaging, the method employs dual sensitivity (different sensitivities for lesion detection in predefined regions). We tested this approach by assigning different sensitivities to supratentorial and infratentorial regions, and validated the segmentation for accuracy against manual delineation, and for precision in scan‐rescans.

RESULTS

Intraclass correlation coefficients of .95, .91, and .86 were observed for WML and CSF segmentation accuracy and brain parenchymal fraction (BPF). Dual sensitivity significantly reduced infratentorial false‐positive WMLs, affording increases in global sensitivity without decreasing specificity. Scan‐rescan yielded coefficients of variation (COVs) of 8% and .4% for WMLs and BPF and COVs of .8%, 1%, and 2% for GM, WM, and CSF volumes. WML volume difference/precision was .49 ± .72 mL over a range of 0–24 mL. Correlation between BPF and age was r = .62 (P = .0004), and effect size for detecting brain atrophy was Cohen's d = 1.26 (standardized mean difference vs. healthy controls).

CONCLUSIONS

This pipeline produces probability maps for brain lesions and tissue classes, facilitating expert review/correction and may provide high throughput, efficient characterization of MS in large datasets.

An Advanced MRI Multi-Modalities Segmentation Methodology Dedicated to Multiple Sclerosis Lesions Exploration and Differentiation

Multiple sclerosis (MS) is one of the most common neurological diseases in young people. This paper dealt with an automatic biomedical aided tool involving volumetric segmentation of multiple sclerosis lesions. To meet this challenge, our proposed methodology requires one preliminary cerebral zones segmentation performed using a new Gaussian mixture model based on various databases atlases. Afterward, lesion segmentation begins with the estimation of a lesion map, which is then subjected to threshold constraints and refined by a new lesion expansion algorithm. The evaluation was carried out on four clinical databases integrating various clinical cases which had different lesion loads and were presented by a set of MRI modalities at several noise levels. The results compared with those of the existing methods proved excellent cerebral segmentation with dice averages close to 0.8 and sensitivity and specificity averages greater than 0.9. In addition, depending on the used database, the lesion segmentation recorded mean values were close to or greater than 0.8 for the different metrics. The detection error and outline error averages were about 0.3. Besides the ability to identify the lesions affecting the different parts of the brain, even those spreading in the gray matter, the proposed methodology identified the lesions cores and their surrounding vasogenic edema. This has been thoroughly tested and validated by highly qualified radiologists and neurologists. The evaluation of the resulting discriminations recorded values close to or greater than 0.9 for dice, sensitivity, and specificity. As a valuable benefit, a computer aided diagnosis tool could be offered to clinicians. It would help efficiently during the MS diagnosis and avoid several confusions. Besides, it could be used for longitudinal survey and henceforth extends to other pathologies that could be explored by MRI modalities, such as glioblastoma or alzheimer's disease.

Neuropsychology and cognitive neuroscience in the fMRI era: A recapitulation of localizationist and connectionist views

OBJECTIVE

We highlight the past 25 years of cognitive neuroscience and neuropsychology, focusing on the impact to the field of the introduction in 1992 of functional MRI (fMRI).

METHOD

We reviewed the past 25 years of literature in cognitive neuroscience and neuropsychology, focusing on the relation and interplay of fMRI studies and studies utilizing the "lesion method" in human participants with focal brain damage.

RESULTS

Our review highlights the state of localist/connectionist research debates in cognitive neuroscience and neuropsychology circa 1992, and details how the introduction of fMRI into the field at that time catalyzed a new wave of efforts to map complex human behavior to specific brain regions. This, in turn, eventually evolved into many studies that focused on networks and connections between brain areas, culminating in recent years with large-scale investigations such as the Human Connectome Project.

CONCLUSIONS

We argue that throughout the past 25 years, neuropsychology-and more precisely, the "lesion method" in humans-has continued to play a critical role in arbitrating conclusions and theories derived from inferred patterns of local brain activity or wide-spread connectivity from functional imaging approaches. We conclude by highlighting the future for neuropsychology in the context of an increasingly complex methodological armamentarium. (PsycINFO Database Record

Brain lesion segmentation through image synthesis and outlier detection

Cerebral small vessel disease (SVD) can manifest in a number of ways. Many of these result in hyperintense regions visible on T2-weighted magnetic resonance (MR) images. The automatic segmentation of these lesions has been the focus of many studies. However, previous methods tended to be limited to certain types of pathology, as a consequence of either restricting the search to the white matter, or by training on an individual pathology. Here we present an unsupervised abnormality detection method which is able to detect abnormally hyperintense regions on FLAIR regardless of the underlying pathology or location. The method uses a combination of image synthesis, Gaussian mixture models and one class support vector machines, and needs only be trained on healthy tissue. We evaluate our method by comparing segmentation results from 127 subjects with SVD with three established methods and report significantly superior performance across a number of metrics.

Volumetric Analysis from a Harmonized Multisite Brain MRI Study of a Single Subject with Multiple Sclerosis

BACKGROUND AND PURPOSE

MR imaging can be used to measure structural changes in the brains of individuals with multiple sclerosis and is essential for diagnosis, longitudinal monitoring, and therapy evaluation. The North American Imaging in Multiple Sclerosis Cooperative steering committee developed a uniform high-resolution 3T MR imaging protocol relevant to the quantification of cerebral lesions and atrophy and implemented it at 7 sites across the United States. To assess intersite variability in scan data, we imaged a volunteer with relapsing-remitting MS with a scan-rescan at each site.

MATERIALS AND METHODS

All imaging was acquired on Siemens scanners (4 Skyra, 2 Tim Trio, and 1 Verio). Expert segmentations were manually obtained for T1-hypointense and T2 (FLAIR) hyperintense lesions. Several automated lesion-detection and whole-brain, cortical, and deep gray matter volumetric pipelines were applied. Statistical analyses were conducted to assess variability across sites, as well as systematic biases in the volumetric measurements that were site-related.

RESULTS

Systematic biases due to site differences in expert-traced lesion measurements were significant (P < .01 for both T1 and T2 lesion volumes), with site explaining >90% of the variation (range, 13.0-16.4 mL in T1 and 15.9-20.1 mL in T2) in lesion volumes. Site also explained >80% of the variation in most automated volumetric measurements. Output measures clustered according to scanner models, with similar results from the Skyra versus the other 2 units.

CONCLUSIONS

Even in multicenter studies with consistent scanner field strength and manufacturer after protocol harmonization, systematic differences can lead to severe biases in volumetric analyses.

Validation of a Regression Technique for Segmentation of White Matter Hyperintensities in Alzheimer's Disease

Segmentation and volumetric quantification of white matter hyperintensities (WMHs) is essential in assessment and monitoring of the vascular burden in aging and Alzheimer's disease (AD), especially when considering their effect on cognition. Manually segmenting WMHs in large cohorts is technically unfeasible due to time and accuracy concerns. Automated tools that can detect WMHs robustly and with high accuracy are needed. Here, we present and validate a fully automatic technique for segmentation and volumetric quantification of WMHs in aging and AD. The proposed technique combines intensity and location features frommultiplemagnetic resonance imaging contrasts and manually labeled training data with a linear classifier to perform fast and robust segmentations. It provides both a continuous subject specific WMH map reflecting different levels of tissue damage and binary segmentations. Themethodwas used to detectWMHs in 80 elderly/AD brains (ADC data set) as well as 40 healthy subjects at risk of AD (PREVENT-AD data set). Robustness across different scanners was validated using ten subjects from ADNI2/GO study. Voxel-wise and volumetric agreements were evaluated using Dice similarity index (SI) and intra-class correlation (ICC), yielding ICC=0.96 , SI = 0.62±0.16 for ADC data set and ICC=0.78 , SI=0.51±0.15 for PREVENT-AD data set. The proposed method was robust in the independent sample yielding SI=0.64±0.17 with ICC=0.93 for ADNI2/GO subjects. The proposed method provides fast, accurate, and robust segmentations on previously unseen data from different models of scanners, making it ideal to study WMHs in large scale multi-site studies.

Deep Learning for Brain MRI Segmentation: State of the Art and Future Directions

Quantitative analysis of brain MRI is routine for many neurological diseases and conditions and relies on accurate segmentation of structures of interest. Deep learning-based segmentation approaches for brain MRI are gaining interest due to their self-learning and generalization ability over large amounts of data. As the deep learning architectures are becoming more mature, they gradually outperform previous state-of-the-art classical machine learning algorithms. This review aims to provide an overview of current deep learning-based segmentation approaches for quantitative brain MRI. First we review the current deep learning architectures used for segmentation of anatomical brain structures and brain lesions. Next, the performance, speed, and properties of deep learning approaches are summarized and discussed. Finally, we provide a critical assessment of the current state and identify likely future developments and trends.

Location Sensitive Deep Convolutional Neural Networks for Segmentation of White Matter Hyperintensities

The anatomical location of imaging features is of crucial importance for accurate diagnosis in many medical tasks. Convolutional neural networks (CNN) have had huge successes in computer vision, but they lack the natural ability to incorporate the anatomical location in their decision making process, hindering success in some medical image analysis tasks. In this paper, to integrate the anatomical location information into the network, we propose several deep CNN architectures that consider multi-scale patches or take explicit location features while training. We apply and compare the proposed architectures for segmentation of white matter hyperintensities in brain MR images on a large dataset. As a result, we observe that the CNNs that incorporate location information substantially outperform a conventional segmentation method with handcrafted features as well as CNNs that do not integrate location information. On a test set of 50 scans, the best configuration of our networks obtained a Dice score of 0.792, compared to 0.805 for an independent human observer. Performance levels of the machine and the independent human observer were not statistically significantly different (p-value = 0.06).

Performance comparison of 10 different classification techniques in segmenting white matter hyperintensities in aging

INTRODUCTION

White matter hyperintensities (WMHs) are areas of abnormal signal on magnetic resonance images (MRIs) that characterize various types of histopathological lesions. The load and location of WMHs are important clinical measures that may indicate the presence of small vessel disease in aging and Alzheimer's disease (AD) patients. Manually segmenting WMHs is time consuming and prone to inter-rater and intra-rater variabilities. Automated tools that can accurately and robustly detect these lesions can be used to measure the vascular burden in individuals with AD or the elderly population in general. Many WMH segmentation techniques use a classifier in combination with a set of intensity and location features to segment WMHs, however, the optimal choice of classifier is unknown.

METHODS

We compare 10 different linear and nonlinear classification techniques to identify WMHs from MRI data. Each classifier is trained and optimized based on a set of features obtained from co-registered MR images containing spatial location and intensity information. We further assess the performance of the classifiers using different combinations of MRI contrast information. The performances of the different classifiers were compared on three heterogeneous multi-site datasets, including images acquired with different scanners and different scan-parameters. These included data from the ADC study from University of California Davis, the NACC database and the ADNI study. The classifiers (naïve Bayes, logistic regression, decision trees, random forests, support vector machines, k-nearest neighbors, bagging, and boosting) were evaluated using a variety of voxel-wise and volumetric similarity measures such as Dice Kappa similarity index (SI), Intra-Class Correlation (ICC), and sensitivity as well as computational burden and processing times. These investigations enable meaningful comparisons between the performances of different classifiers to determine the most suitable classifiers for segmentation of WMHs. In the spirit of open-source science, we also make available a fully automated tool for segmentation of WMHs with pre-trained classifiers for all these techniques.

RESULTS

Random Forests yielded the best performance among all classifiers with mean Dice Kappa (SI) of 0.66±0.17 and ICC=0.99 for the ADC dataset (using T1w, T2w, PD, and FLAIR scans), SI=0.72±0.10, ICC=0.93 for the NACC dataset (using T1w and FLAIR scans), SI=0.66±0.23, ICC=0.94 for ADNI1 dataset (using T1w, T2w, and PD scans) and SI=0.72±0.19, ICC=0.96 for ADNI2/GO dataset (using T1w and FLAIR scans). Not using the T2w/PD information did not change the performance of the Random Forest classifier (SI=0.66±0.17, ICC=0.99). However, not using FLAIR information in the ADC dataset significantly decreased the Dice Kappa, but the volumetric correlation did not drastically change (SI=0.47±0.21, ICC=0.95).

CONCLUSION

Our investigations showed that with appropriate features, most off-the-shelf classifiers are able to accurately detect WMHs in presence of FLAIR scan information, while Random Forests had the best performance across all datasets. However, we observed that the performances of most linear classifiers and some nonlinear classifiers drastically decline in absence of FLAIR information, with Random Forest still retaining the best performance.

State-of-the-Art Methods for Brain Tissue Segmentation: A Review

Brain tissue segmentation is one of the most sought after research areas in medical image processing. It provides detailed quantitative brain analysis for accurate disease diagnosis, detection, and classification of abnormalities. It plays an essential role in discriminating healthy tissues from lesion tissues. Therefore, accurate disease diagnosis and treatment planning depend merely on the performance of the segmentation method used. In this review, we have studied the recent advances in brain tissue segmentation methods and their state-of-the-art in neuroscience research. The review also highlights the major challenges faced during tissue segmentation of the brain. An effective comparison is made among state-of-the-art brain tissue segmentation methods. Moreover, a study of some of the validation measures to evaluate different segmentation methods is also discussed. The brain tissue segmentation, content in terms of methodologies, and experiments presented in this review are encouraging enough to attract researchers working in this field.

Autodelineation of cervical cancers using multiparametric magnetic resonance imaging and machine learning

BACKGROUND

Tumour delineation is a challenging, time-consuming and complex part of radiotherapy planning. In this study, an automatic method for delineating locally advanced cervical cancers was developed using a machine learning approach.

MATERIALS AND METHODS

A method for tumour segmentation based on image voxel classification using Fisher?s Linear Discriminant Analysis (LDA) was developed. This was applied to magnetic resonance (MR) images of 78 patients with locally advanced cervical cancer. The segmentation was based on multiparametric MRI consisting of T2- weighted (T2w), T1-weighted (T1w) and dynamic contrast-enhanced (DCE) sequences, and included intensity and spatial information from the images. The model was trained and assessed using delineations made by two radiologists.

RESULTS

Segmentation based on T2w or T1w images resulted in mean sensitivity and specificity of 94% and 52%, respectively. Including DCE-MR images improved the segmentation model?s performance significantly, giving mean sensitivity and specificity of 85?93%. Comparisons with radiologists? tumour delineations gave Dice similarity coefficients of up to 0.44.

CONCLUSION

Voxel classification using a machine learning approach is a flexible and fully automatic method for tumour delineation. Combining all relevant MR image series resulted in high sensitivity and specificity. Moreover, the presented method can be extended to include additional imaging modalities.

Reproducible segmentation of white matter hyperintensities using a new statistical definition

OBJECTIVES

We present a method based on a proposed statistical definition of white matter hyperintensities (WMH), which can work with any combination of conventional magnetic resonance (MR) sequences without depending on manually delineated samples.

MATERIALS AND METHODS

T1-weighted, T2-weighted, FLAIR, and PD sequences acquired at 1.5 Tesla from 119 subjects from the Kings Health Partners-Dementia Case Register (healthy controls, mild cognitive impairment, Alzheimer's disease) were used. The segmentation was performed using a proposed definition for WMH based on the one-tailed Kolmogorov-Smirnov test.

RESULTS

The presented method was verified, given all possible combinations of input sequences, against manual segmentations and a high similarity (Dice 0.85-0.91) was observed. Comparing segmentations with different input sequences to one another also yielded a high similarity (Dice 0.83-0.94) that exceeded intra-rater similarity (Dice 0.75-0.91). We compared the results with those of other available methods and showed that the segmentation based on the proposed definition has better accuracy and reproducibility in the test dataset used.

CONCLUSION

Overall, the presented definition is shown to produce accurate results with higher reproducibility than manual delineation. This approach can be an alternative to other manual or automatic methods not only because of its accuracy, but also due to its good reproducibility.

Automated detection of white matter hyperintensities of all sizes in cerebral small vessel disease

PURPOSE

White matter hyperintensities (WMH) are seen on FLAIR-MRI in several neurological disorders, including multiple sclerosis, dementia, Parkinsonism, stroke and cerebral small vessel disease (SVD). WMHs are often used as biomarkers for prognosis or disease progression in these diseases, and additionally longitudinal quantification of WMHs is used to evaluate therapeutic strategies. Human readers show considerable disagreement and inconsistency on detection of small lesions. A multitude of automated detection algorithms for WMHs exists, but since most of the current automated approaches are tuned to optimize segmentation performance according to Jaccard or Dice scores, smaller WMHs often go undetected in these approaches. In this paper, the authors propose a method to accurately detect all WMHs, large as well as small.

METHODS

A two-stage learning approach was used to discriminate WMHs from normal brain tissue. Since small and larger WMHs have quite a different appearance, the authors have trained two probabilistic classifiers: one for the small WMHs (⩽3 mm effective diameter) and one for the larger WMHs (>3 mm in-plane effective diameter). For each size-specific classifier, an Adaboost is trained for five iterations, with random forests as the basic classifier. The feature sets consist of 22 features including intensities, location information, blob detectors, and second order derivatives. The outcomes of the two first-stage classifiers were combined into a single WMH likelihood by a second-stage classifier. Their method was trained and evaluated on a dataset with MRI scans of 362 SVD patients (312 subjects for training and validation annotated by one and 50 for testing annotated by two trained raters). To analyze performance on the separate test set, the authors performed a free-response receiving operating characteristic (FROC) analysis, instead of using segmentation based methods that tend to ignore the contribution of small WMHs.

RESULTS

Experimental results based on FROC analysis demonstrated a close performance of the proposed computer aided detection (CAD) system to human readers. While an independent reader had 0.78 sensitivity with 28 false positives per volume on average, their proposed CAD system reaches a sensitivity of 0.73 with the same number of false positives.

CONCLUSIONS

The authors have developed a CAD system with all its ingredients being optimized for a better detection of WMHs of all size, which shows performance close to an independent reader.

FLAIRfusion Processing with Contrast Inversion : Improving Detection and Reading Time of New Cerebral MS Lesions

PURPOSE

To evaluate the performance of an innovative image processing approach for detection of T2-weighted hyperintense multiple sclerosis (MS) lesions.

METHODS

In this study 20 consecutive patients with inflammatory demyelinating lesions were retrospectively evaluated of whom 10 patients featured progressive disease and 10 a stable lesion load. 3 mm transversal FLAIRfusion imaging was processed and archived. Image processing was performed through landmark-based 3D co-registration of the previous and current isotropic FLAIR examination followed by inversion of image contrast. Thereby, the hyperintense signals of the unchanged MS plaques extinguish each other, while newly developed lesions appear bright on FLAIRfusion. Diagnostic performance was evaluated by 4 experienced readers. Consensus reading supplied the reference standard. Sensitivity, specificity, NPV (negative predictive value), PPV (positive predictive value), interreader agreement and reading time were the outcome measures analyzed.

RESULTS

Combined sensitivity was 100% at a specificity of 88.2%, with PPV ranging from 83.3% to 90.1% and NPV at 100%. Reading time was nearly 5‑fold faster than conventional side by side comparison (35.6 s vs. 163.7 s, p < 0.001). Cohen's kappa was excellent (>0.75; p < 0.001) and Cronbach's alpha was 0.994.

CONCLUSION

FLAIRfusion provides reliable detection of newly developed MS lesions along with strong interreader agreement across all levels of expertise in 35 s of reading time.

Longitudinal multiple sclerosis lesion segmentation: Resource and challenge

In conjunction with the ISBI 2015 conference, we organized a longitudinal lesion segmentation challenge providing training and test data to registered participants. The training data consisted of five subjects with a mean of 4.4 time-points, and test data of fourteen subjects with a mean of 4.4 time-points. All 82 data sets had the white matter lesions associated with multiple sclerosis delineated by two human expert raters. Eleven teams submitted results using state-of-the-art lesion segmentation algorithms to the challenge, with ten teams presenting their results at the conference. We present a quantitative evaluation comparing the consistency of the two raters as well as exploring the performance of the eleven submitted results in addition to three other lesion segmentation algorithms. The challenge presented three unique opportunities: (1) the sharing of a rich data set; (2) collaboration and comparison of the various avenues of research being pursued in the community; and (3) a review and refinement of the evaluation metrics currently in use. We report on the performance of the challenge participants, as well as the construction and evaluation of a consensus delineation. The image data and manual delineations will continue to be available for download, through an evaluation website2 as a resource for future researchers in the area. This data resource provides a platform to compare existing methods in a fair and consistent manner to each other and multiple manual raters.

Two Time Point MS Lesion Segmentation in Brain MRI: An Expectation-Maximization Framework

Purpose: Lesion volume is a meaningful measure in multiple sclerosis (MS) prognosis. Manual lesion segmentation for computing volume in a single or multiple time points is time consuming and suffers from intra and inter-observer variability. Methods: In this paper, we present MSmetrix-long: a joint expectation-maximization (EM) framework for two time point white matter (WM) lesion segmentation. MSmetrix-long takes as input a 3D T1-weighted and a 3D FLAIR MR image and segments lesions in three steps: (1) cross-sectional lesion segmentation of the two time points; (2) creation of difference image, which is used to model the lesion evolution; (3) a joint EM lesion segmentation framework that uses output of step (1) and step (2) to provide the final lesion segmentation. The accuracy (Dice score) and reproducibility (absolute lesion volume difference) of MSmetrix-long is evaluated using two datasets. Results: On the first dataset, the median Dice score between MSmetrix-long and expert lesion segmentation was 0.63 and the Pearson correlation coefficient (PCC) was equal to 0.96. On the second dataset, the median absolute volume difference was 0.11 ml. Conclusions: MSmetrix-long is accurate and consistent in segmenting MS lesions. Also, MSmetrix-long compares favorably with the publicly available longitudinal MS lesion segmentation algorithm of Lesion Segmentation Toolbox.

MRI FLAIR lesion segmentation in multiple sclerosis: Does automated segmentation hold up with manual annotation?

Introduction

Magnetic resonance imaging (MRI) has become key in the diagnosis and disease monitoring of patients with multiple sclerosis (MS). Both, T2 lesion load and Gadolinium (Gd) enhancing T1 lesions represent important endpoints in MS clinical trials by serving as a surrogate of clinical disease activity. T2- and fluid-attenuated inversion recovery (FLAIR) lesion quantification - largely due to methodological constraints – is still being performed manually or in a semi-automated fashion, although strong efforts have been made to allow automated quantitative lesion segmentation. In 2012, Schmidt and co-workers published an algorithm to be applied on FLAIR sequences. The aim of this study was to apply the Schmidt algorithm on an independent data set and compare automated segmentation to inter-rater variability of three independent, experienced raters.

Methods

MRI data of 50 patients with RRMS were randomly selected from a larger pool of MS patients attending the MS Clinic at the Brain and Mind Centre, University of Sydney, Australia. MRIs were acquired on a 3.0T GE scanner (Discovery MR750, GE Medical Systems, Milwaukee, WI) using an 8 channel head coil. We determined T2-lesion load (total lesion volume and total lesion number) using three versions of an automated segmentation algorithm (Lesion growth algorithm (LGA) based on SPM8 or SPM12 and lesion prediction algorithm (LPA) based on SPM12) as first described by Schmidt et al. (2012). Additionally, manual segmentation was performed by three independent raters. We calculated inter-rater correlation coefficients (ICC) and dice coefficients (DC) for all possible pairwise comparisons.

Results

We found a strong correlation between manual and automated lesion segmentation based on LGA SPM8, regarding lesion volume (ICC = 0.958 and DC = 0.60) that was not statistically different from the inter-rater correlation (ICC = 0.97 and DC = 0.66). Correlation between the two other algorithms (LGA SPM12 and LPA SPM12) and manual raters was weaker but still adequate (ICC = 0.927 and DC = 0.53 for LGA SPM12 and ICC = 0.949 and DC = 0.57 for LPA SPM12). Variability of both manual and automated segmentation was significantly higher regarding lesion numbers.

Conclusion

Automated lesion volume quantification can be applied reliably on FLAIR data sets using the SPM based algorithm of Schmidt et al. and shows good agreement with manual segmentation.

•

Fully automated and manual MS lesion segmentation on FLAIR images were compared.

•

Automated FLAIR lesion volume segmentation holds up with manual annotation.

•

When using DC and ICC, SPM8 based algorithm performed better than recent updates.

A Semiautomatic Method for Multiple Sclerosis Lesion Segmentation on Dual-Echo MR Imaging: Application in a Multicenter Context

BACKGROUND AND PURPOSE

The automatic segmentation of MS lesions could reduce time required for image processing together with inter- and intraoperator variability for research and clinical trials. A multicenter validation of a proposed semiautomatic method for hyperintense MS lesion segmentation on dual-echo MR imaging is presented.

MATERIALS AND METHODS

The classification technique used is based on a region-growing approach starting from manual lesion identification by an expert observer with a final segmentation-refinement step. The method was validated in a cohort of 52 patients with relapsing-remitting MS, with dual-echo images acquired in 6 different European centers.

RESULTS

We found a mathematic expression that made the optimization of the method independent of the need for a training dataset. The automatic segmentation was in good agreement with the manual segmentation (dice similarity coefficient = 0.62 and root mean square error = 2 mL). Assessment of the segmentation errors showed no significant differences in algorithm performance between the different MR scanner manufacturers (P > .05).

CONCLUSIONS

The method proved to be robust, and no center-specific training of the algorithm was required, offering the possibility for application in a clinical setting. Adoption of the method should lead to improved reliability and less operator time required for image analysis in research and clinical trials in MS.

Abnormality Detection via Iterative Deformable Registration and Basis-Pursuit Decomposition

We present a generic method for automatic detection of abnormal regions in medical images as deviations from a normative data base. The algorithm decomposes an image, or more broadly a function defined on the image grid, into the superposition of a normal part and a residual term. A statistical model is constructed with regional sparse learning to represent normative anatomical variations among a reference population (e.g., healthy controls), in conjunction with a Markov random field regularization that ensures mutual consistency of the regional learning among partially overlapping image blocks. The decomposition is performed in a principled way so that the normal part fits well with the learned normative model, while the residual term absorbs pathological patterns, which may then be detected through a statistical significance test. The decomposition is applied to multiple image features from an individual scan, detecting abnormalities using both intensity and shape information. We form an iterative scheme that interleaves abnormality detection with deformable registration, gradually improving robustness of the spatial normalization and precision of the detection. The algorithm is evaluated with simulated images and clinical data of brain lesions, and is shown to achieve robust deformable registration and localize pathological regions simultaneously. The algorithm is also applied on images from Alzheimer's disease patients to demonstrate the generality of the method.

Reproducibility and variability of quantitative magnetic resonance imaging markers in cerebral small vessel disease

Brain imaging is essential for the diagnosis and characterization of cerebral small vessel disease. Several magnetic resonance imaging markers have therefore emerged, providing new information on the diagnosis, progression, and mechanisms of small vessel disease. Yet, the reproducibility of these small vessel disease markers has received little attention despite being widely used in cross-sectional and longitudinal studies. This review focuses on the main small vessel disease-related markers on magnetic resonance imaging including: white matter hyperintensities, lacunes, dilated perivascular spaces, microbleeds, and brain volume. The aim is to summarize, for each marker, what is currently known about: (1) its reproducibility in studies with a scan-rescan procedure either in single or multicenter settings; (2) the acquisition-related sources of variability; and, (3) the techniques used to minimize this variability. Based on the results, we discuss technical and other challenges that need to be overcome in order for these markers to be reliably used as outcome measures in future clinical trials. We also highlight the key points that need to be considered when designing multicenter magnetic resonance imaging studies of small vessel disease.

Automated Robust Image Segmentation: Level Set Method Using Nonnegative Matrix Factorization with Application to Brain MRI

We address the problem of fully automated region discovery and robust image segmentation by devising a new deformable model based on the level set method (LSM) and the probabilistic nonnegative matrix factorization (NMF). We describe the use of NMF to calculate the number of distinct regions in the image and to derive the local distribution of the regions, which is incorporated into the energy functional of the LSM. The results demonstrate that our NMF-LSM method is superior to other approaches when applied to synthetic binary and gray-scale images and to clinical magnetic resonance images (MRI) of the human brain with and without a malignant brain tumor, glioblastoma multiforme. In particular, the NMF-LSM method is fully automated, highly accurate, less sensitive to the initial selection of the contour(s) or initial conditions, more robust to noise and model parameters, and able to detect as small distinct regions as desired. These advantages stem from the fact that the proposed method relies on histogram information instead of intensity values and does not introduce nuisance model parameters. These properties provide a general approach for automated robust region discovery and segmentation in heterogeneous images. Compared with the retrospective radiological diagnoses of two patients with non-enhancing grade 2 and 3 oligodendroglioma, the NMF-LSM detects earlier progression times and appears suitable for monitoring tumor response. The NMF-LSM method fills an important need of automated segmentation of clinical MRI.

White Matter MS-Lesion Segmentation Using a Geometric Brain Model

Brain magnetic resonance imaging (MRI) in patients with Multiple Sclerosis (MS) shows regions of signal abnormalities, named plaques or lesions. The spatial lesion distribution plays a major role for MS diagnosis. In this paper we present a 3D MS-lesion segmentation method based on an adaptive geometric brain model. We model the topological properties of the lesions and brain tissues in order to constrain the lesion segmentation to the white matter. As a result, the method is independent of an MRI atlas. We tested our method on the MICCAI MS grand challenge proposed in 2008 and achieved competitive results. In addition, we used an in-house dataset of 15 MS patients, for which we achieved best results in most distances in comparison to atlas based methods. Besides classical segmentation distances, we motivate and formulate a new distance to evaluate the quality of the lesion segmentation, while being robust with respect to minor inconsistencies at the boundary level of the ground truth annotation.

MR brain image analysis in dementia: From quantitative imaging biomarkers to ageing brain models and imaging genetics

MR brain image analysis has constantly been a hot topic research area in medical image analysis over the past two decades. In this article, it is discussed how the field developed from the construction of tools for automatic quantification of brain morphology, function, connectivity and pathology, to creating models of the ageing brain in normal ageing and disease, and tools for integrated analysis of imaging and genetic data. The current and future role of the field in improved understanding of the development of neurodegenerative disease is discussed, and its potential for aiding in early and differential diagnosis and prognosis of different types of dementia. For the latter, the use of reference imaging data and reference models derived from large clinical and population imaging studies, and the application of machine learning techniques on these reference data, are expected to play a key role.

Automated, quantitative measures of grey and white matter lesion burden correlates with motor and cognitive function in children with unilateral cerebral palsy

White and grey matter lesions are the most prevalent type of injury observable in the Magnetic Resonance Images (MRIs) of children with cerebral palsy (CP). Previous studies investigating the impact of lesions in children with CP have been qualitative, limited by the lack of automated segmentation approaches in this setting. As a result, the quantitative relationship between lesion burden has yet to be established. In this study, we perform automatic lesion segmentation on a large cohort of data (107 children with unilateral CP and 18 healthy children) with a new, validated method for segmenting both white matter (WM) and grey matter (GM) lesions. The method has better accuracy (94%) than the best current methods (73%), and only requires standard structural MRI sequences. Anatomical lesion burdens most predictive of clinical scores of motor, cognitive, visual and communicative function were identified using the Least Absolute Shrinkage and Selection operator (LASSO). The improved segmentations enabled identification of significant correlations between regional lesion burden and clinical performance, which conform to known structure-function relationships. Model performance was validated in an independent test set, with significant correlations observed for both WM and GM regional lesion burden with motor function (p < 0.008), and between WM and GM lesions alone with cognitive and visual function respectively (p < 0.008). The significant correlation of GM lesions with functional outcome highlights the serious implications GM lesions, in addition to WM lesions, have for prognosis, and the utility of structural MRI alone for quantifying lesion burden and planning therapy interventions.

Deep 3D Convolutional Encoder Networks With Shortcuts for Multiscale Feature Integration Applied to Multiple Sclerosis Lesion Segmentation

We propose a novel segmentation approach based on deep 3D convolutional encoder networks with shortcut connections and apply it to the segmentation of multiple sclerosis (MS) lesions in magnetic resonance images. Our model is a neural network that consists of two interconnected pathways, a convolutional pathway, which learns increasingly more abstract and higher-level image features, and a deconvolutional pathway, which predicts the final segmentation at the voxel level. The joint training of the feature extraction and prediction pathways allows for the automatic learning of features at different scales that are optimized for accuracy for any given combination of image types and segmentation task. In addition, shortcut connections between the two pathways allow high- and low-level features to be integrated, which enables the segmentation of lesions across a wide range of sizes. We have evaluated our method on two publicly available data sets (MICCAI 2008 and ISBI 2015 challenges) with the results showing that our method performs comparably to the top-ranked state-of-the-art methods, even when only relatively small data sets are available for training. In addition, we have compared our method with five freely available and widely used MS lesion segmentation methods (EMS, LST-LPA, LST-LGA, Lesion-TOADS, and SLS) on a large data set from an MS clinical trial. The results show that our method consistently outperforms these other methods across a wide range of lesion sizes.

A robust statistics driven volume-scalable active contour for segmenting anatomical structures in volumetric medical images with complex conditions

BACKGROUND

Accurate segmentation of anatomical structures in medical images is a critical step in the development of computer assisted intervention systems. However, complex image conditions, such as intensity inhomogeneity, noise and weak object boundary, often cause considerable difficulties in medical image segmentation. To cope with these difficulties, we propose a novel robust statistics driven volume-scalable active contour framework, to extract desired object boundary from magnetic resonance (MR) and computed tomography (CT) imagery in 3D.

METHODS

We define an energy functional in terms of the initial seeded labels and two fitting functions that are derived from object local robust statistics features. This energy is then incorporated into a level set scheme which drives the active contour evolving and converging at the desired position of the object boundary. Due to the local robust statistics and the volume scaling function in the energy fitting term, the object features in local volumes are learned adaptively to guide the motion of the contours, which thereby guarantees the capability of our method to cope with intensity inhomogeneity, noise and weak boundary. In addition, the initialization of active contour is simplified by select several seeds in the object and/or background to eliminate the sensitivity to initialization.

RESULTS

The proposed method was applied to extensive public available volumetric medical images with challenging image conditions. The segmentation results of various anatomical structures, such as white matter (WM), atrium, caudate nucleus and brain tumor, were evaluated quantitatively by comparing with the corresponding ground truths. It was found that the proposed method achieves consistent and coherent segmentation accuracy of 0.9246 ± 0.0068 for WM, 0.9043 ± 0.0131 for liver tumors, 0.8725 ± 0.0374 for caudate nucleus, 0.8802 ± 0.0595 for brain tumors, etc., measured by Dice similarity coefficients value for the overlap between the algorithm one and the ground truth. Further comparative experimental results showed desirable performances of the proposed method over several well-known segmentation methods in terms of accuracy and robustness.

CONCLUSION

We proposed an approach to accurate segment volumetric medical images with complex conditions. The accuracy of segmentation, robustness to noise and contour initialization were validated on the basis of extensive MR and CT volumes.

Estimating Brain Lesion Volume Change in Multiple Sclerosis by Subtraction of Magnetic Resonance Images

BACKGROUND

Change in lesion volume over time, measured on brain magnetic resonance imaging (MRI) scans, is an important outcome measure for natural history studies and clinical trials in multiple sclerosis (MS).

PURPOSE

To develop and test image analysis methods for quantification of lesion volume change in order to improve reliability.

METHODS

The technique is based on registration and subtraction, and was evaluated in a cohort of 20 MS patients with dual-echo images acquired annually over a period of four years. The study protocol was approved by the local ethics review boards of participating centers, and all subjects gave written informed consent. The repeatability was compared to that obtained by the standard method for obtaining lesion volume change by evaluating the total volume at each time point, and then subtracting the volumes to obtain the difference.

RESULTS

Compared to the standard method, the subtraction method had improved intrarater correlation (0.95 and 0.72 for the subtraction method and the standard method, respectively) and interrater correlation (0.51 and 0.28, respectively). Furthermore, the mean time required to analyze the scans from one patient was 41 minutes for the subtraction method compared to 125 minutes for the standard method.

CONCLUSION

Use of the subtraction algorithm leads to improved reliability and lower operator fatigue in clinical trials and studies of the natural history of MS.

Automatic Detection of White Matter Hyperintensities in Healthy Aging and Pathology Using Magnetic Resonance Imaging: A Review

White matter hyperintensities (WMH) are commonly seen in the brain of healthy elderly subjects and patients with several neurological and vascular disorders. A truly reliable and fully automated method for quantitative assessment of WMH on magnetic resonance imaging (MRI) has not yet been identified. In this paper, we review and compare the large number of automated approaches proposed for segmentation of WMH in the elderly and in patients with vascular risk factors. We conclude that, in order to avoid artifacts and exclude the several sources of bias that may influence the analysis, an optimal method should comprise a careful preprocessing of the images, be based on multimodal, complementary data, take into account spatial information about the lesions and correct for false positives. All these features should not exclude computational leanness and adaptability to available data.

A MS-lesion pattern discrimination plot based on geostatistics

INTRODUCTION

A geostatistical approach to characterize MS-lesion patterns based on their geometrical properties is presented.

METHODS

A dataset of 259 binary MS-lesion masks in MNI space was subjected to directional variography. A model function was fit to express the observed spatial variability in x, y, z directions by the geostatistical parameters Range and Sill.

RESULTS

Parameters Range and Sill correlate with MS-lesion pattern surface complexity and total lesion volume. A scatter plot of ln(Range) versus ln(Sill), classified by pattern anisotropy, enables a consistent and clearly arranged presentation of MS-lesion patterns based on geometry: the so-called MS-Lesion Pattern Discrimination Plot.

CONCLUSIONS

The geostatistical approach and the graphical representation of results are considered efficient exploratory data analysis tools for cross-sectional, follow-up, and medication impact analysis.

Patch-Based Segmentation with Spatial Consistency: Application to MS Lesions in Brain MRI

This paper presents an automatic lesion segmentation method based on similarities between multichannel patches. A patch database is built using training images for which the label maps are known. For each patch in the testing image, k similar patches are retrieved from the database. The matching labels for these k patches are then combined to produce an initial segmentation map for the test case. Finally an iterative patch-based label refinement process based on the initial segmentation map is performed to ensure the spatial consistency of the detected lesions. The method was evaluated in experiments on multiple sclerosis (MS) lesion segmentation in magnetic resonance images (MRI) of the brain. An evaluation was done for each image in the MICCAI 2008 MS lesion segmentation challenge. Results are shown to compete with the state of the art in the challenge. We conclude that the proposed algorithm for segmentation of lesions provides a promising new approach for local segmentation and global detection in medical images.

Imalytics Preclinical: Interactive Analysis of Biomedical Volume Data

A software tool is presented for interactive segmentation of volumetric medical data sets. To allow interactive processing of large data sets, segmentation operations, and rendering are GPU-accelerated. Special adjustments are provided to overcome GPU-imposed constraints such as limited memory and host-device bandwidth. A general and efficient undo/redo mechanism is implemented using GPU-accelerated compression of the multiclass segmentation state. A broadly applicable set of interactive segmentation operations is provided which can be combined to solve the quantification task of many types of imaging studies. A fully GPU-accelerated ray casting method for multiclass segmentation rendering is implemented which is well-balanced with respect to delay, frame rate, worst-case memory consumption, scalability, and image quality. Performance of segmentation operations and rendering are measured using high-resolution example data sets showing that GPU-acceleration greatly improves the performance. Compared to a reference marching cubes implementation, the rendering was found to be superior with respect to rendering delay and worst-case memory consumption while providing sufficiently high frame rates for interactive visualization and comparable image quality. The fast interactive segmentation operations and the accurate rendering make our tool particularly suitable for efficient analysis of multimodal image data sets which arise in large amounts in preclinical imaging studies.

Quantifying brain tissue volume in multiple sclerosis with automated lesion segmentation and filling

Lesion filling has been successfully applied to reduce the effect of hypo-intense T1-w Multiple Sclerosis (MS) lesions on automatic brain tissue segmentation. However, a study of fully automated pipelines incorporating lesion segmentation and lesion filling on tissue volume analysis has not yet been performed. Here, we analyzed the % of error introduced by automating the lesion segmentation and filling processes in the tissue segmentation of 70 clinically isolated syndrome patient images. First of all, images were processed using the LST and SLS toolkits with different pipeline combinations that differed in either automated or manual lesion segmentation, and lesion filling or masking out lesions. Then, images processed following each of the pipelines were segmented into gray matter (GM) and white matter (WM) using SPM8, and compared with the same images where expert lesion annotations were filled before segmentation. Our results showed that fully automated lesion segmentation and filling pipelines reduced significantly the % of error in GM and WM volume on images of MS patients, and performed similarly to the images where expert lesion annotations were masked before segmentation. In all the pipelines, the amount of misclassified lesion voxels was the main cause in the observed error in GM and WM volume. However, the % of error was significantly lower when automatically estimated lesions were filled and not masked before segmentation. These results are relevant and suggest that LST and SLS toolboxes allow the performance of accurate brain tissue volume measurements without any kind of manual intervention, which can be convenient not only in terms of time and economic costs, but also to avoid the inherent intra/inter variability between manual annotations.

•

SLS and LST pipelines incorporate fully automated lesion segmentation and filling.

•

Both pipelines reduced the % of error in GM and WM volume of MS patient images.

•

Performance was similar to images with lesion annotations masked before segmentation.

•

The amount of misclassified lesion voxels was the main factor in the % of error.

•

SLS/LST can reduce time, economic costs, and the variability between manual annotations.

Bayesian model selection for pathological neuroimaging data applied to white matter lesion segmentation

In neuroimaging studies, pathologies can present themselves as abnormal intensity patterns. Thus, solutions for detecting abnormal intensities are currently under investigation. As each patient is unique, an unbiased and biologically plausible model of pathological data would have to be able to adapt to the subject's individual presentation. Such a model would provide the means for a better understanding of the underlying biological processes and improve one's ability to define pathologically meaningful imaging biomarkers. With this aim in mind, this work proposes a hierarchical fully unsupervised model selection framework for neuroimaging data which enables the distinction between different types of abnormal image patterns without pathological a priori knowledge. Its application on simulated and clinical data demonstrated the ability to detect abnormal intensity clusters, resulting in a competitive to improved behavior in white matter lesion segmentation when compared to three other freely-available automated methods.

Longitudinal Patch-Based Segmentation of Multiple Sclerosis White Matter Lesions

Segmenting T2-hyperintense white matter lesions from longitudinal MR images is essential in understanding progression of multiple sclerosis. Most lesion segmentation techniques find lesions independently at each time point, even though there are different noise and image contrast variations at each point in the time series. In this paper, we present a patch based 4D lesion segmentation method that takes advantage of the temporal component of longitudinal data. For each subject with multiple time-points, 4D patches are constructed from the T₁-w and FLAIR scans of all time-points. For every 4D patch from a subject, a few relevant matching 4D patches are found from a reference, such that their convex combination reconstructs the subject's 4D patch. Then corresponding manual segmentation patches of the reference are combined in a similar manner to generate a 4D membership of lesions of the subject patch. We compare our 4D patch-based segmentation with independent 3D voxel-based and patch-based lesion segmentation algorithms. Based on ground truth segmentations from 30 data sets, we show that the mean Dice coefficients between manual and automated segmentations improve after using the 4D approach compared to two state-of-the-art 3D segmentation algorithms.

Stratified mixture modeling for segmentation of white-matter lesions in brain MR images

Accurate characterization of white-matter lesions from magnetic resonance (MR) images has increasing importance for diagnosis and management of treatment of certain neurological diseases, and can be performed in an objective and effective way by automated lesion segmentation. This usually involves modeling the whole-brain MR intensity distribution, however, capturing various sources of MR intensity variability and lesion heterogeneity results in highly complex whole-brain MR intensity models, thus their robust estimation on a large set of MR images presents a huge challenge. We propose a novel approach employing stratified mixture modeling, where the main premise is that the otherwise complex whole-brain model can be reduced to a tractable parametric form in small brain subregions. We show on MR images of multiple sclerosis (MS) patients with different lesion loads that robust estimators enable accurate mixture modeling of MR intensity in small brain subregions even in the presence of lesions. Recombination of the mixture models across strata provided an accurate whole-brain MR intensity model. Increasing the number of subregions and, thereby, the model complexity, consistently improved the accuracy of whole-brain MR intensity modeling and segmentation of normal structures. The proposed approach was incorporated into three unsupervised lesion segmentation methods and, compared to original and three other state-of-the-art methods, the proposed modeling approach significantly improved lesion segmentation according to increased Dice similarity indices and lower number of false positives on real MR images of 30 patients with MS.

Automatic white matter lesion segmentation using contrast enhanced FLAIR intensity and Markov Random Field

White matter lesions (WMLs) are small groups of dead cells that clump together in the white matter of brain. In this paper, we propose a reliable method to automatically segment WMLs. Our method uses a novel filter to enhance the intensity of WMLs. Then a feature set containing enhanced intensity, anatomical and spatial information is used to train a random forest classifier for the initial segmentation of WMLs. Following that a reliable and robust edge potential function based Markov Random Field (MRF) is proposed to obtain the final segmentation by removing false positive WMLs. Quantitative evaluation of the proposed method is performed on 24 subjects of ENVISion study. The segmentation results are validated against the manual segmentation, performed under the supervision of an expert neuroradiologist. The results show a dice similarity index of 0.76 for severe lesion load, 0.73 for moderate lesion load and 0.61 for mild lesion load. In addition to that we have compared our method with three state of the art methods on 20 subjects of Medical Image Computing and Computer Aided Intervention Society's (MICCAI's) MS lesion challenge dataset, where our method shows better segmentation accuracy compare to the state of the art methods. These results indicate that the proposed method can assist the neuroradiologists in assessing the WMLs in clinical practice.

Improving Multiple Sclerosis Plaque Detection Using a Semiautomated Assistive Approach

BACKGROUND AND PURPOSE

Treating MS with disease-modifying drugs relies on accurate MR imaging follow-up to determine the treatment effect. We aimed to develop and validate a semiautomated software platform to facilitate detection of new lesions and improved lesions.

MATERIALS AND METHODS

We developed VisTarsier to assist manual comparison of volumetric FLAIR sequences by using interstudy registration, resectioning, and color-map overlays that highlight new lesions and improved lesions. Using the software, 2 neuroradiologists retrospectively assessed MR imaging MS comparison study pairs acquired between 2009 and 2011 (161 comparison study pairs met the study inclusion criteria). Lesion detection and reading times were recorded. We tested inter- and intraobserver agreement and comparison with original clinical reports. Feedback was obtained from referring neurologists to assess the potential clinical impact.

RESULTS

More comparison study pairs with new lesions (reader 1, n = 60; reader 2, n = 62) and improved lesions (reader 1, n = 28; reader 2, n = 39) were recorded by using the software compared with original radiology reports (new lesions, n = 20; improved lesions, n = 5); the difference reached statistical significance (P < .001). Interobserver lesion number agreement was substantial (≥1 new lesion: κ = 0.87; 95% CI, 0.79-0.95; ≥1 improved lesion: κ = 0.72; 95% CI, 0.59-0.85), and overall interobserver lesion number correlation was good (Spearman ρ: new lesion = 0.910, improved lesion = 0.774). Intraobserver agreement was very good (new lesion: κ = 1.0, improved lesion: κ = 0.94; 95% CI, 0.82-1.00). Mean reporting times were <3 minutes. Neurologists indicated retrospective management alterations in 79% of comparative study pairs with newly detected lesion changes.

CONCLUSIONS

Using software that highlights changes between study pairs can improve lesion detection. Neurologist feedback indicated a likely impact on management.