Automated detection of white matter and cortical lesions in early stages of multiple sclerosis

Emerging MRI biomarkers for the diagnosis of multiple sclerosis

The need to improve diagnostic precision in multiple sclerosis (MS) is widely recognized. In recent years, several novel magnetic resonance imaging (MRI) biomarkers have been proposed to enhance diagnostic specificity and reduce misdiagnosis. Some of these imaging biomarkers are deemed highly specific for MS and are likely ready to enter the MS diagnostic work-up, while others are still in their exploratory phase. In addition, new synthetic MRI contrasts and artificial intelligence-based diagnostic algorithms are being tested to reduce the time burden related to imaging data acquisition and analysis. In this review, we summarize the most recent advancement in the field, focusing on the adoption of these novel MRI biomarkers-whether used alone or in combination-for the differential diagnosis of MS.

Automated Quantitative Susceptibility and Morphometry MR Study: Feasibility and Interrelation Between Clinical Score, Lesion Load, Deep Grey Matter and Normal-Appearing White Matter in Multiple Sclerosis

INTRODUCTION

Lesion load (LL), deep gray matter (DGM) and normal-appearing white matter (NAWM) susceptibility and morphometry may help in monitoring brain changes in multiple sclerosis (MS) patients. We aimed at evaluating the feasibility of a fully automated segmentation and the potential interrelation between these biomarkers and clinical disability.

METHODS

Sixty-six patients with brain MRIs and clinical evaluations (Expanded Disability Status Scale [EDSS]) were retrospectively included. Automated prototypes were used for the segmentation and morphometry of brain regions (MorphoBox) and MS lesions (LeManPV). Susceptibility maps were estimated using standard post-processing (RESHARP and TVSB). Spearman's rho was computed to evaluate the interrelation between biomarkers and EDSS.

RESULTS

We found (i) anticorrelations between the LL and right thalamus susceptibility (rho = -0.46, p < 0.001) and between the LL and NAWM susceptibility (rho = [-0.68 to -0.25], p ≤ 0.05); (ii) an anticorrelation between LL and DGM (rho = [-0.71 to -0.36], p < 0.04) and WM morphometry (rho = [-0.64 to -0.28], p ≤ 0.01); and (iii) a positive correlation between EDSS and LL (rho = [0.28 to 0.5], p ≤ 0.03) and anticorrelation between EDSS and NAWM susceptibility (rho = [-0.29 to -0.38], p < 0.014).

CONCLUSIONS

Fully automated brain morphometry and susceptibility monitoring is feasible in MS patients. The lesion load, thalamus and NAWM susceptibility values and trophicity are interrelated and correlate with disability.

Microstructural characterization of multiple sclerosis lesion phenotypes using multiparametric longitudinal analysis

BACKGROUND AND OBJECTIVES

In multiple sclerosis (MS), slowly expanding lesions were shown to be associated with worse disability and prognosis. Their timely detection from cross-sectional data at early disease stages could be clinically relevant to inform treatment planning. Here, we propose to use multiparametric, quantitative MRI to allow a better cross-sectional characterization of lesions with different longitudinal phenotypes.

METHODS

We analysed T1 and T2 relaxometry maps from a longitudinal cohort of MS patients. Lesions were classified as enlarging, shrinking, new or stable based on their longitudinal volumetric change using a newly developed automated technique. Voxelwise deviations were computed as z-scores by comparing individual patient data to T1, T2 and T2/T1 normative values from healthy subjects. We studied the distribution of microstructural properties inside lesions and within perilesional tissue.

RESULTS AND CONCLUSIONS

Stable lesions exhibited the highest T1 and T2 z-scores in lesion tissue, while the lowest values were observed for new lesions. Shrinking lesions presented the highest T1 z-scores in the first perilesional ring while enlarging lesions showed the highest T2 z-scores in the same region. Finally, a classification model was trained to predict the longitudinal lesion type based on microstructural metrics and feature importance was assessed. Z-scores estimated in lesion and perilesional tissue from T1, T2 and T2/T1 quantitative maps carry discriminative and complementary information to classify longitudinal lesion phenotypes, hence suggesting that multiparametric MRI approaches are essential for a better understanding of the pathophysiological mechanisms underlying disease activity in MS lesions.

Tract-wise microstructural analysis informs on current and future disability in early multiple sclerosis

OBJECTIVES

Microstructural characterization of patients with multiple sclerosis (MS) has been shown to correlate better with disability compared to conventional radiological biomarkers. Quantitative MRI provides effective means to characterize microstructural brain tissue changes both in lesions and normal-appearing brain tissue. However, the impact of the location of microstructural alterations in terms of neuronal pathways has not been thoroughly explored so far. Here, we study the extent and the location of tissue changes probed using quantitative MRI along white matter (WM) tracts extracted from a connectivity atlas.

METHODS

We quantified voxel-wise T1 tissue alterations compared to normative values in a cohort of 99 MS patients. For each WM tract, we extracted metrics reflecting tissue alterations both in lesions and normal-appearing WM and correlated these with cross-sectional disability and disability evolution after 2 years.

RESULTS

In early MS patients, T1 alterations in normal-appearing WM correlated better with disability evolution compared to cross-sectional disability. Further, the presence of lesions in supratentorial tracts was more strongly associated with cross-sectional disability, while microstructural alterations in infratentorial pathways yielded higher correlations with disability evolution. In progressive patients, all major WM pathways contributed similarly to explaining disability, and correlations with disability evolution were generally poor.

CONCLUSIONS

We showed that microstructural changes evaluated in specific WM pathways contribute to explaining future disability in early MS, hence highlighting the potential of tract-wise analyses in monitoring disease progression. Further, the proposed technique allows to estimate WM tract-specific microstructural characteristics in clinically compatible acquisition times, without the need for advanced diffusion imaging.

A systematic review of the application of machine-learning algorithms in multiple sclerosis

INTRODUCTION

The applications of artificial intelligence, and in particular automatic learning or "machine learning" (ML), constitute both a challenge and a great opportunity in numerous scientific, technical, and clinical disciplines. Specific applications in the study of multiple sclerosis (MS) have been no exception, and constitute an area of increasing interest in recent years.

OBJECTIVE

We present a systematic review of the application of ML algorithms in MS.

MATERIALS AND METHODS

We used the PubMed search engine, which allows free access to the MEDLINE medical database, to identify studies including the keywords "machine learning" and "multiple sclerosis." We excluded review articles, studies written in languages other than English or Spanish, and studies that were mainly technical and did not specifically apply to MS. The final selection included 76 articles, and 38 were rejected.

CONCLUSIONS

After the review process, we established 4 main applications of ML in MS: 1) classifying MS subtypes; 2) distinguishing patients with MS from healthy controls and individuals with other diseases; 3) predicting progression and response to therapeutic interventions; and 4) other applications. Results found to date have shown that ML algorithms may offer great support for health professionals both in clinical settings and in research into MS.

Diagnostic delay of multiple sclerosis: prevalence, determinants and consequences

BACKGROUND

Early diagnosis and treatment of patients with multiple sclerosis (MS) are associated with better outcomes; however, diagnostic delays remain a major problem.

OBJECTIVE

Describe the prevalence, determinants and consequences of delayed diagnoses.

METHODS

This single-centre ambispective study analysed 146 adult relapsing-remitting MS patients (2016-2021) for frequency and determinants of diagnostic delays and their associations with clinical, cognitive, imaging and biochemical measures.

RESULTS

Diagnostic delays were identified in 77 patients (52.7%), including 42 (28.7%) physician-dependent cases and 35 (24.0%) patient-dependent cases. Diagnosis was delayed in 22 (15.1%) patients because of misdiagnosis by a neurologist. A longer diagnostic delay was associated with trends towards greater Expanded Disability Status Scale (EDSS) scores (B = 0.03; p = 0.034) and greater z-score of the blood neurofilament light chain (B = 0.35; p = 0.031) at the time of diagnosis. Compared with patients diagnosed at their first clinical relapse, patients with a history of >1 relapse at diagnosis (n = 63; 43.2%) had a trend towards greater EDSS scores (B = 0.06; p = 0.006) and number of total (B = 0.13; p = 0.040) and periventricular (B = 0.06; p = 0.039) brain lesions.

CONCLUSION

Diagnostic delays in MS are common, often determined by early misdiagnosis and associated with greater disease burden.

A Multicenter Longitudinal MRI Study Assessing LeMan-PV Software Accuracy in the Detection of White Matter Lesions in Multiple Sclerosis Patients

BACKGROUND

Detecting new and enlarged lesions in multiple sclerosis (MS) patients is needed to determine their disease activity. LeMan-PV is a software embedded in the scanner reconstruction system of one vendor, which automatically assesses new and enlarged white matter lesions (NELs) in the follow-up of MS patients; however, multicenter validation studies are lacking.

PURPOSE

To assess the accuracy of LeMan-PV for the longitudinal detection NEL white-matter MS lesions in a multicenter clinical setting.

STUDY TYPE

Retrospective, longitudinal.

SUBJECTS

A total of 206 patients with a definitive MS diagnosis and at least two follow-up MRI studies from five centers participating in the Swiss Multiple Sclerosis Cohort study. Mean age at first follow-up = 45.2 years (range: 36.9-52.8 years); 70 males.

FIELD STRENGTH/SEQUENCE

Fluid attenuated inversion recovery (FLAIR) and T1-weighted magnetization prepared rapid gradient echo (T1-MPRAGE) sequences at 1.5 T and 3 T.

ASSESSMENT

The study included 313 MRI pairs of datasets. Data were analyzed with LeMan-PV and compared with a manual "reference standard" provided by a neuroradiologist. A second rater (neurologist) performed the same analysis in a subset of MRI pairs to evaluate the rating-accuracy. The Sensitivity (Se), Specificity (Sp), Accuracy (Acc), F1-score, lesion-wise False-Positive-Rate (aFPR), and other measures were used to assess LeMan-PV performance for the detection of NEL at 1.5 T and 3 T. The performance was also evaluated in the subgroup of 123 MRI pairs at 3 T.

STATISTICAL TESTS

Intraclass correlation coefficient (ICC) and Cohen's kappa (CK) were used to evaluate the agreement between readers.

RESULTS

The interreader agreement was high for detecting new lesions (ICC = 0.97, Pvalue < 10-20 , CK = 0.82, P value = 0) and good (ICC = 0.75, P value < 10-12 , CK = 0.68, P value = 0) for detecting enlarged lesions. Across all centers, scanner field strengths (1.5 T, 3 T), and for NEL, LeMan-PV achieved: Acc = 61%, Se = 65%, Sp = 60%, F1-score = 0.44, aFPR = 1.31. When both follow-ups were acquired at 3 T, LeMan-PV accuracy was higher (Acc = 66%, Se = 66%, Sp = 66%, F1-score = 0.28, aFPR = 3.03).

DATA CONCLUSION

In this multicenter study using clinical data settings acquired at 1.5 T and 3 T, and variations in MRI protocols, LeMan-PV showed similar sensitivity in detecting NEL with respect to other recent 3 T multicentric studies based on neural networks. While LeMan-PV performance is not optimal, its main advantage is that it provides automated clinical decision support integrated into the radiological-routine flow.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: Stage 2.

Imaging cortical lesions in multiple sclerosis

PURPOSE OF REVIEW

Cortical lesions are an established pathological feature of multiple sclerosis, develop from the earliest disease stages and contribute to disease progression. Here, we discuss current imaging approaches for detecting cortical lesions in vivo and their contribution for improving our understanding of cortical lesion pathogenesis as well as their clinical significance.

RECENT FINDINGS

Although a variable portion of cortical lesions goes undetected at clinical field strength and even at ultra-high field MRI, their evaluation is still clinically relevant. Cortical lesions are important for differential multiple sclerosis (MS) diagnosis, have relevant prognostic value and independently predict disease progression. Some studies also show that cortical lesion assessment could be used as a therapeutic outcome target in clinical trials. Advances in ultra-high field MRI not only allow increased cortical lesion detection in vivo but also the disclosing of some interesting features of cortical lesions related to their pattern of development and evolution as well to the nature of associated pathological changes, which might prove relevant for better understanding the pathogenesis of these lesions.

SUMMARY

Despite some limitations, imaging of cortical lesions is of paramount importance in MS for elucidating disease mechanisms as well as for improving patient management in clinic.

Using Deep-Learning-Based Artificial Intelligence Technique to Automatically Evaluate the Collateral Status of Multiphase CTA in Acute Ischemic Stroke

BACKGROUND

Collateral status is an important predictor for the outcome of acute ischemic stroke with large vessel occlusion. Multiphase computed-tomography angiography (mCTA) is useful to evaluate the collateral status, but visual evaluation of this examination is time-consuming. This study aims to use an artificial intelligence (AI) technique to develop an automatic AI prediction model for the collateral status of mCTA.

METHODS

This retrospective study enrolled subjects with acute ischemic stroke receiving endovascular thrombectomy between January 2015 and June 2020 in a tertiary referral hospital. The demographic data and images of mCTA were collected. The collateral status of all mCTA was visually evaluated. Images at the basal ganglion and supraganglion levels of mCTA were selected to produce AI models using the convolutional neural network (CNN) technique to automatically predict the collateral status of mCTA.

RESULTS

A total of 82 subjects were enrolled. There were 57 cases randomly selected for the training group and 25 cases for the validation group. In the training group, there were 40 cases with a positive collateral result (good or intermediate) and 17 cases with a negative collateral result (poor). In the validation group, there were 21 cases with a positive collateral result and 4 cases with a negative collateral result. During training for the CNN prediction model, the accuracy of the training group could reach 0.999 ± 0.015, whereas the prediction model had a performance of 0.746 ± 0.008 accuracy on the validation group. The area under the ROC curve was 0.7.

CONCLUSIONS

This study suggests that the application of the AI model derived from mCTA images to automatically evaluate the collateral status is feasible.

Single-timepoint low-dimensional characterization and classification of acute versus chronic multiple sclerosis lesions using machine learning

Multiple sclerosis (MS) is a chronic inflammatory and neurodegenerative disease characterized by the appearance of focal lesions across the central nervous system. The discrimination of acute from chronic MS lesions may yield novel biomarkers of inflammatory disease activity which may support patient management in the clinical setting and provide endpoints in clinical trials. On a single timepoint and in the absence of a prior reference scan, existing methods for acute lesion detection rely on the segmentation of hyperintense foci on post-gadolinium T1-weighted magnetic resonance imaging (MRI), which may underestimate recent acute lesion activity. In this paper, we aim to improve the sensitivity of acute MS lesion detection in the single-timepoint setting, by developing a novel machine learning approach for the automatic detection of acute MS lesions, using single-timepoint conventional non-contrast T1- and T2-weighted brain MRI. The MRI input data are supplemented via the use of a convolutional neural network generating "lesion-free" reconstructions from original "lesion-present" scans using image inpainting. A multi-objective statistical ranking module evaluates the relevance of textural radiomic features from the core and periphery of lesion sites, compared within "lesion-free" versus "lesion-present" image pairs. Then, an ensemble classifier is optimized through a recursive loop seeking consensus both in the feature space (via a greedy feature-pruning approach) and in the classifier space (via model selection repeated after each pruning operation). This leads to the identification of a compact textural signature characterizing lesion phenotype. On the patch-level task of acute versus chronic MS lesion classification, our method achieves a balanced accuracy in the range of 74.3-74.6% on fully external validation cohorts.

Cortical lesions, central vein sign, and paramagnetic rim lesions in multiple sclerosis: Emerging machine learning techniques and future avenues

The current diagnostic criteria for multiple sclerosis (MS) lack specificity, and this may lead to misdiagnosis, which remains an issue in present-day clinical practice. In addition, conventional biomarkers only moderately correlate with MS disease progression. Recently, some MS lesional imaging biomarkers such as cortical lesions (CL), the central vein sign (CVS), and paramagnetic rim lesions (PRL), visible in specialized magnetic resonance imaging (MRI) sequences, have shown higher specificity in differential diagnosis. Moreover, studies have shown that CL and PRL are potential prognostic biomarkers, the former correlating with cognitive impairments and the latter with early disability progression. As machine learning-based methods have achieved extraordinary performance in the assessment of conventional imaging biomarkers, such as white matter lesion segmentation, several automated or semi-automated methods have been proposed as well for CL, PRL, and CVS. In the present review, we first introduce these MS biomarkers and their imaging methods. Subsequently, we describe the corresponding machine learning-based methods that were proposed to tackle these clinical questions, putting them into context with respect to the challenges they are facing, including non-standardized MRI protocols, limited datasets, and moderate inter-rater variability. We conclude by presenting the current limitations that prevent their broader deployment and suggesting future research directions.

Multiple sclerosis cortical lesion detection with deep learning at ultra-high-field MRI

Manually segmenting multiple sclerosis (MS) cortical lesions (CLs) is extremely time consuming, and past studies have shown only moderate inter-rater reliability. To accelerate this task, we developed a deep-learning-based framework (CLAIMS: Cortical Lesion AI-Based Assessment in Multiple Sclerosis) for the automated detection and classification of MS CLs with 7 T MRI. Two 7 T datasets, acquired at different sites, were considered. The first consisted of 60 scans that include 0.5 mm isotropic MP2RAGE acquired four times (MP2RAGE×4), 0.7 mm MP2RAGE, 0.5 mm T₂ *-weighted GRE, and 0.5 mm T₂ *-weighted EPI. The second dataset consisted of 20 scans including only 0.75 × 0.75 × 0.9 mm³ MP2RAGE. CLAIMS was first evaluated using sixfold cross-validation with single and multi-contrast 0.5 mm MRI input. Second, the performance of the model was tested on 0.7 mm MP2RAGE images after training with either 0.5 mm MP2RAGE×4, 0.7 mm MP2RAGE, or alternating the two. Third, its generalizability was evaluated on the second external dataset and compared with a state-of-the-art technique based on partial volume estimation and topological constraints (MSLAST). CLAIMS trained only with MP2RAGE×4 achieved results comparable to those of the multi-contrast model, reaching a CL true positive rate of 74% with a false positive rate of 30%. Detection rate was excellent for leukocortical and subpial lesions (83%, and 70%, respectively), whereas it reached 53% for intracortical lesions. The correlation between disability measures and CL count was similar for manual and CLAIMS lesion counts. Applying a domain-scanner adaptation approach and testing CLAIMS on the second dataset, the performance was superior to MSLAST when considering a minimum lesion volume of 6 μL (lesion-wise detection rate of 71% versus 48%). The proposed framework outperforms previous state-of-the-art methods for automated CL detection across scanners and protocols. In the future, CLAIMS may be useful to support clinical decisions at 7 T MRI, especially in the field of diagnosis and differential diagnosis of MS patients.

Periventricular gradient of T1 tissue alterations in multiple sclerosis

•

T₁ relaxation times alterations were assessed based on a study-specific atlas.

•

T₁ alterations depend on distance from lateral ventricles (“periventricular gradient”).

•

Gradient parameters correlate better with disability compared to conventional MRI.

Objective

Pathology in multiple sclerosis is not homogenously distributed. Recently, it has been shown that structures adjacent to CSF are more severely affected. A gradient of brain tissue involvement was shown with more severe pathology in periventricular areas and in proximity to brain surfaces such as the subarachnoid spaces and ependyma, and hence termed the “surface–in” gradient. Here, we study whether (i) the surface-in gradient of periventricular tissue alteration measured by T₁ relaxometry is already present in early multiple sclerosis patients, (ii) how it differs between early and progressive multiple sclerosis patients, and (iii) whether the gradient-derived metrics in normal-appearing white matter and lesions correlate better with physical disability than conventional MRI-based metrics.

Methods

Forty-seven patients with early multiple sclerosis, 52 with progressive multiple sclerosis, and 92 healthy controls were included in the study. Isotropic 3D T₁ relaxometry maps were obtained using the Magnetization-Prepared 2 Rapid Acquisition Gradient Echoes sequence at 3 T. After spatially normalizing the T₁ maps into a study-specific common space, T₁ inter-subject variability within the healthy cohort was modelled voxel-wise, yielding a normative T₁ atlas. Individual comparisons of each multiple sclerosis patient against the atlas were performed by computing z-scores. Equidistant bands of voxels were defined around the ventricles in the supratentorial white matter; the z-scores in these bands were analysed and compared between the early and progressive multiple sclerosis cohorts. Correlations between both conventional and z-score-gradient-derived MRI metrics and the Expanded Disability Status Scale were assessed.

Results

Patients with early and progressive multiple sclerosis demonstrated a periventricular gradient of T₁ relaxation time z-scores. In progressive multiple sclerosis, z-score-derived metrics reflecting the gradient of tissue abnormality in normal-appearing white matter were more strongly correlated with disability (maximal rho = 0.374) than the conventional lesion volume and count (maximal rho = 0.189 and 0.21 respectively). In early multiple sclerosis, the gradient of normal-appearing white matter volume with z-scores > 2 at baseline correlated with clinical disability assessed at two years follow-up.

Conclusion

Our results suggest that the surface-in white matter gradient of tissue alteration is detectable with T1 relaxometry and is already present at clinical disease onset. The periventricular gradients correlate with clinical disability. The periventricular gradient in normal-appearing white matter may thus qualify as a promising biomarker for monitoring of disease activity from an early stage in all phenotypes of multiple sclerosis.

Higher-resolution quantification of white matter hypointensities by large-scale transfer learning from 2D images on the JPSC-AD cohort

White matter lesions (WML) commonly occur in older brains and are quantifiable on MRI, often used as a biomarker in Aging research. Although algorithms are regularly proposed that identify these lesions from T2‐fluid‐attenuated inversion recovery (FLAIR) sequences, none so far can estimate lesions directly from T1‐weighted images with acceptable accuracy. Since 3D T1 is a polyvalent and higher‐resolution sequence, it could be beneficial to obtain the distribution of WML directly from it. However a serious difficulty, both for algorithms and human, can be found in the ambiguities of brain signal intensity in T1 images. This manuscript shows that a cross‐domain ConvNet (Convolutional Neural Network) approach can help solve this problem. Still, this is non‐trivial, as it would appear to require a large and varied dataset (for robustness) labelled at the same high resolution (for spatial accuracy). Instead, our model was taught from two‐dimensional FLAIR images with a loss function designed to handle the super‐resolution need. And crucially, we leveraged a very large training set for this task, the recently assembled, multi‐sites Japan Prospective Studies Collaboration for Aging and Dementia (JPSC‐AD) cohort. We describe the two‐step procedure that we followed to handle such a large number of imperfectly labeled samples. A large‐scale accuracy evaluation conducted against FreeSurfer 7, and a further visual expert rating revealed that WML segmentation from our ConvNet was consistently better. Finally, we made a directly usable software program based on that trained ConvNet model, available at https://github.com/bthyreau/deep-T1-WMH.

Multiple Sclerosis Lesions Segmentation Using Attention-Based CNNs in FLAIR Images

Objective: Multiple Sclerosis (MS) is an autoimmune and demyelinating disease that leads to lesions in the central nervous system. This disease can be tracked and diagnosed using Magnetic Resonance Imaging (MRI). A multitude of multimodality automatic biomedical approaches are used to segment lesions that are not beneficial for patients in terms of cost, time, and usability. The authors of the present paper propose a method employing just one modality (FLAIR image) to segment MS lesions accurately. Methods: A patch-based Convolutional Neural Network (CNN) is designed, inspired by 3D-ResNet and spatial-channel attention module, to segment MS lesions. The proposed method consists of three stages: (1) the Contrast-Limited Adaptive Histogram Equalization (CLAHE) is applied to the original images and concatenated to the extracted edges to create 4D images; (2) the patches of size [Formula: see text] are randomly selected from the 4D images; and (3) the extracted patches are passed into an attention-based CNN which is used to segment the lesions. Finally, the proposed method was compared to previous studies of the same dataset. Results: The current study evaluates the model with a test set of ISIB challenge data. Experimental results illustrate that the proposed approach significantly surpasses existing methods of Dice similarity and Absolute Volume Difference while the proposed method uses just one modality (FLAIR) to segment the lesions. Conclusion: The authors have introduced an automated approach to segment the lesions, which is based on, at most, two modalities as an input. The proposed architecture comprises convolution, deconvolution, and an SCA-VoxRes module as an attention module. The results show, that the proposed method outperforms well compared to other methods.

Automatic segmentation of white matter hyperintensities: validation and comparison with state-of-the-art methods on both Multiple Sclerosis and elderly subjects

•

Automatic segmentation of MS lesions and age-related WMH from 3D T1 and T2-FLAIR.

•

Comparison to consensus show improved performance of WHASA-3D compared to WHASA.

•

WHASA-3D outperforms available state-of-the-art methods with their default settings.

•

WHASA-3D could be a useful tool for clinical practice and clinical trials.

Different types of white matter hyperintensities (WMH) can be observed through MRI in the brain and spinal cord, especially Multiple Sclerosis (MS) lesions for patients suffering from MS and age-related WMH for subjects with cognitive disorders and/or elderly people. To better diagnose and monitor the disease progression, the quantitative evaluation of WMH load has proven to be useful for clinical routine and trials. Since manual delineation for WMH segmentation is highly time-consuming and suffers from intra and inter observer variability, several methods have been proposed to automatically segment either MS lesions or age-related WMH, but none is validated on both WMH types. Here, we aim at proposing the White matter Hyperintensities Automatic Segmentation Algorithm adapted to 3D T2-FLAIR datasets (WHASA-3D), a fast and robust automatic segmentation tool designed to be implemented in clinical practice for the detection of both MS lesions and age-related WMH in the brain, using both 3D T1-weighted and T2-FLAIR images. In order to increase its robustness for MS lesions, WHASA-3D expands the original WHASA method, which relies on the coupling of non-linear diffusion framework and watershed parcellation, where regions considered as WMH are selected based on intensity and location characteristics, and finally refined with geodesic dilation. The previous validation was performed on 2D T2-FLAIR and subjects with cognitive disorders and elderly subjects. 60 subjects from a heterogeneous database of dementia patients, multiple sclerosis patients and elderly subjects with multiple MRI scanners and a wide range of lesion loads were used to evaluate WHASA and WHASA-3D through volume and spatial agreement in comparison with consensus reference segmentations. In addition, a direct comparison on the MS database with six available supervised and unsupervised state-of-the-art WMH segmentation methods (LST-LGA and LPA, Lesion-TOADS, lesionBrain, BIANCA and nicMSlesions) with default and optimised settings (when feasible) was conducted. WHASA-3D confirmed an improved performance with respect to WHASA, achieving a better spatial overlap (Dice) (0.67 vs 0.63), a reduced absolute volume error (AVE) (3.11 vs 6.2 mL) and an increased volume agreement (intraclass correlation coefficient, ICC) (0.96 vs 0.78). Compared to available state-of-the-art algorithms on the MS database, WHASA-3D outperformed both unsupervised and supervised methods when used with their default settings, showing the highest volume agreement (ICC = 0.95) as well as the highest average Dice (0.58). Optimising and/or retraining LST-LGA, BIANCA and nicMSlesions, using a subset of the MS database as training set, resulted in improved performances on the remaining testing set (average Dice: LST-LGA default/optimized = 0.41/0.51, BIANCA default/optimized = 0.22/0.39, nicMSlesions default/optimized = 0.17/0.63, WHASA-3D = 0.58). Evaluation and comparison results suggest that WHASA-3D is a reliable and easy-to-use method for the automated segmentation of white matter hyperintensities, for both MS lesions and age-related WMH. Further validation on larger datasets would be useful to confirm these first findings.

Artificial double inversion recovery images can substitute conventionally acquired images: an MRI-histology study

Cortical multiple sclerosis lesions are disease-specific, yet inconspicuous on magnetic resonance images (MRI). Double inversion recovery (DIR) images are sensitive, but often unavailable in clinical routine and clinical trials. Artificially generated images can mitigate this issue, but lack histopathological validation. In this work, artificial DIR images were generated from postmortem 3D-T1 and proton-density (PD)/T2 or 3D-T1 and 3D fluid-inversion recovery (FLAIR) images, using a generative adversarial network. All sequences were scored for cortical lesions, blinded to histopathology. Subsequently, tissue samples were stained for proteolipid protein (myelin) and scored for cortical lesions type I-IV (leukocortical, intracortical, subpial and cortex-spanning, respectively). Histopathological scorings were then (unblinded) compared to MRI using linear mixed models. Images from 38 patients (26 female, mean age 64.3 ± 10.7) were included. A total of 142 cortical lesions were detected, predominantly subpial. Histopathology-blinded/unblinded sensitivity was 13.4/35.2% for artificial DIR generated from T1-PD/T2, 14.1/41.5% for artificial DIR from T1-FLAIR, 17.6/49.3% for conventional DIR and 10.6/34.5% for 3D-T1. When blinded to histopathology, there were no differences; with histopathological feedback at hand, conventional DIR and artificial DIR from T1-FLAIR outperformed the other sequences. Differences between histopathology-blinded/unblinded sensitivity could be minified through adjustment of the scoring criteria. In conclusion, artificial DIR images, particularly generated from T1-FLAIR could potentially substitute conventional DIR images when these are unavailable.

Validating atlas-based lesion disconnectomics in multiple sclerosis: A retrospective multi-centric study

•

Structural disconnectomes can be modelled without diffusion using tractography atlases.

•

Atlas-based and DTI-derived disconnectome topological metrics correlate strongly.

•

MS patient disconnectomes relate to clinical scores.

The translational potential of MR-based connectivity modelling is limited by the need for advanced diffusion imaging, which is not part of clinical protocols for many diseases. In addition, where diffusion data is available, brain connectivity analyses rely on tractography algorithms which imply two major limitations. First, tracking algorithms are known to be sensitive to the presence of white matter lesions and therefore leading to interpretation pitfalls and poor inter-subject comparability in clinical applications such as multiple sclerosis. Second, tractography quality is highly dependent on the acquisition parameters of diffusion sequences, leading to a trade-off between acquisition time and tractography precision.

Here, we propose an atlas-based approach to study the interplay between structural disconnectivity and lesions without requiring individual diffusion imaging. In a multi-centric setting involving three distinct multiple sclerosis datasets (containing both 1.5 T and 3 T data), we compare our atlas-based structural disconnectome computation pipeline to disconnectomes extracted from individual tractography and explore its clinical utility for reducing the gap between radiological findings and clinical symptoms in multiple sclerosis. Results using topological graph properties showed that overall, our atlas-based disconnectomes were suitable approximations of individual disconnectomes from diffusion imaging. Small-worldness was found to decrease for larger total lesion volumes thereby suggesting a loss of efficiency in brain connectivity of MS patients. Finally, the global efficiency of the created brain graph, combined with total lesion volume, allowed to stratify patients into subgroups with different clinical scores in all three cohorts.

Improved particle swarm optimized deep convolutional neural network with super-pixel clustering for multiple sclerosis lesion segmentation in brain MRI imaging

A central nervous system (CNS) disease affecting the insulating myelin sheaths around the brain axons is called multiple sclerosis (MS). In today's world, MS is extensively diagnosed and monitored using the MRI, because of the structural MRI sensitivity in dissemination of white matter lesions with respect to space and time. The main aim of this study is to propose Multiple Sclerosis Lesion Segmentation in Brain MRI imaging using Optimized Deep Convolutional Neural Network and Super-pixel Clustering. Three stages included in the proposed methodology are: (a) preprocessing, (b) segmentation of super-pixel, and (c) classification of super-pixel. In the first stage, image enhancement and skull stripping is done through performing a preprocessing step. In the second stage, the MS lesion and Non-MS lesion regions are segmented through applying SLICO algorithm over each slice of the volume. In the fourth stage, a CNN training and classification is performed using this segmented lesion and non-lesion regions. To handle this complex task, a newly developed Improved Particle Swarm Optimization (IPSO) based optimized convolutional neural network classifier is applied. On clinical MS data, the approach exhibits a significant increase in the accuracy segmenting of WM lesions when compared with the rest of evaluated methods.

Machine Learning Approaches in Study of Multiple Sclerosis Disease Through Magnetic Resonance Images

Multiple sclerosis (MS) is one of the most common autoimmune diseases which is commonly diagnosed and monitored using magnetic resonance imaging (MRI) with a combination of clinical manifestations. The purpose of this review is to highlight the main applications of Machine Learning (ML) models and their performance in the MS field using MRI. We reviewed the articles of the last decade and grouped them based on the applications of ML in MS using MRI data into four categories: 1) Automated diagnosis of MS, 2) Prediction of MS disease progression, 3) Differentiation of MS stages, 4) Differentiation of MS from similar disorders.

Clinical correlates of white matter lesions in Parkinson's disease using automated multi-modal segmentation measures

BACKGROUND

Age-related white matter lesions (WML) are common, impact neuronal connectivity, and affect motor function and cognition. In addition to pathological nigrostriatal losses, WML are also common co-morbidities in Parkinson's disease (PD) that affect postural stability and gait. Automated brain volume measures are increasingly incorporated into the clinical reporting workflow to facilitate precision in medicine. Recently, multi-modal segmentation algorithms have been developed to overcome challenges with WML quantification based on single-modality input.

OBJECTIVE

We evaluated WML volumes and their distribution in a case-control cohort of PD patients to predict the domain-specific clinical severity using a fully automated multi-modal segmentation algorithm.

METHODS

Fifty-five subjects comprising of twenty PD patients and thirty-five age- and gender-matched control subjects underwent standardized motor/gait and cognitive assessments and brain MRI. Spatially differentiated WML obtained using automated segmentation algorithms on multi-modal MPRAGE and FLAIR images were used to predict domain-specific clinical severity. Preliminary statistical analysis focused on describing the relationship between WML and clinical scores, and the distribution of WML by brain regions. Subsequent stepwise regressions were performed to predict each clinical score using WML volumes in different brain regions, while controlling for age.

RESULTS

WML volume strongly correlates with both motor and cognitive dysfunctions in PD patients (p < 0.05), with differential impact in the frontal lobe and periventricular regions on cognitive domains (p < 0.01) and severity of motor deficits (p < 0.01), respectively.

CONCLUSION

Automated multi-modal segmentation algorithms may facilitate precision medicine through regional WML load quantification, which show potential as imaging biomarkers for predicting domain-specific disease severity in PD.

Deep-Learning Generated Synthetic Double Inversion Recovery Images Improve Multiple Sclerosis Lesion Detection

OBJECTIVES

The aim of the study was to implement a deep-learning tool to produce synthetic double inversion recovery (synthDIR) images and compare their diagnostic performance to conventional sequences in patients with multiple sclerosis (MS).

MATERIALS AND METHODS

For this retrospective analysis, 100 MS patients (65 female, 37 [22-68] years) were randomly selected from a prospective observational cohort between 2014 and 2016. In a subset of 50 patients, an artificial neural network (DiamondGAN) was trained to generate a synthetic DIR (synthDIR) from standard acquisitions (T1, T2, and fluid-attenuated inversion recovery [FLAIR]). With the resulting network, synthDIR was generated for the remaining 50 subjects. These images as well as conventionally acquired DIR (trueDIR) and FLAIR images were assessed for MS lesions by 2 independent readers, blinded to the source of the DIR image. Lesion counts in the different modalities were compared using a Wilcoxon signed-rank test, and interrater analysis was performed. Contrast-to-noise ratios were compared for objective image quality.

RESULTS

Utilization of synthDIR allowed to detect significantly more lesions compared with the use of FLAIR images (31.4 ± 20.7 vs 22.8 ± 12.7, P < 0.001). This improvement was mainly attributable to an improved depiction of juxtacortical lesions (12.3 ± 10.8 vs 7.2 ± 5.6, P < 0.001). Interrater reliability was excellent in FLAIR 0.92 (95% confidence interval [CI], 0.85-0.95), synthDIR 0.93 (95% CI, 0.87-0.96), and trueDIR 0.95 (95% CI, 0.85-0.98).Contrast-to-noise ratio in synthDIR exceeded that of FLAIR (22.0 ± 6.4 vs 16.7 ± 3.6, P = 0.009); no significant difference was seen in comparison to trueDIR (22.0 ± 6.4 vs 22.4 ± 7.9, P = 0.87).

CONCLUSIONS

Computationally generated DIR images improve lesion depiction compared with the use of standard modalities. This method demonstrates how artificial intelligence can help improving imaging in specific pathologies.

Current review and next steps for artificial intelligence in multiple sclerosis risk research

In the last few decades, the prevalence of multiple sclerosis (MS), a chronic inflammatory disease of the nervous system, has increased, particularly in Northern European countries, the United States, and United Kingdom. The promise of artificial intelligence (AI) and machine learning (ML) as tools to address problems in MS research has attracted increasing interest in these methods. Bayesian networks offer a clear advantage since they can integrate data and causal knowledge allowing for visualizing interactions between dependent variables and potential confounding factors. A review of AI/ML research methods applied to MS found 216 papers using terms "Multiple Sclerosis", "machine learning", "artificial intelligence", "Bayes", and "Bayesian", of which 90 were relevant and recently published. More than half of these involve the detection and segmentation of MS lesions for quantitative analysis; however clinical and lifestyle risk factor assessment and prediction have largely been ignored. Of those that address risk factors, most provide only association studies for some factors and often fail to include the potential impact of confounding factors and bias (especially where these have causal explanations) that could affect data interpretation, such as reporting quality and medical care access in various countries. To address these gaps in the literature, we propose a causal Bayesian network approach to assessing risk factors for MS, which can address deficiencies in current epidemiological methods of producing risk measurements and makes better use of observational data.

MPRAGE to MP2RAGE UNI translation via generative adversarial network improves the automatic tissue and lesion segmentation in multiple sclerosis patients

BACKGROUND AND OBJECTIVE

Compared to the conventional magnetization-prepared rapid gradient-echo imaging (MPRAGE) MRI sequence, the specialized magnetization prepared 2 rapid acquisition gradient echoes (MP2RAGE) shows a higher brain tissue and lesion contrast in multiple sclerosis (MS) patients. The goal of this work is to retrospectively generate realistic-looking MP2RAGE uniform images (UNI) from already acquired MPRAGE images in order to improve the automatic lesion and tissue segmentation.

METHODS

For this task we propose a generative adversarial network (GAN). Multi-contrast MRI data of 12 healthy controls and 44 patients diagnosed with MS was retrospectively analyzed. Imaging was acquired at 3T using a SIEMENS scanner with MPRAGE, MP2RAGE, FLAIR, and DIR sequences. We train the GAN with both healthy controls and MS patients to generate synthetic MP2RAGE UNI images. These images were then compared to the real MP2RAGE UNI (considered as ground truth) analyzing the output of automatic brain tissue and lesion segmentation tools. Reference-based metrics as well as the lesion-wise true and false positives, Dice coefficient, and volume difference were considered for the evaluation. Statistical differences were assessed with the Wilcoxon signed-rank test.

RESULTS

The synthetic MP2RAGE UNI significantly improves the lesion and tissue segmentation masks in terms of Dice coefficient and volume difference (p-values < 0.001) compared to the MPRAGE. For the segmentation metrics analyzed no statistically significant differences are found between the synthetic and acquired MP2RAGE UNI.

CONCLUSION

Synthesized MP2RAGE UNI images are visually realistic and improve the output of automatic segmentation tools.

A systematic review of the application of machine-learning algorithms in multiple sclerosis

INTRODUCTION

The applications of artificial intelligence, and in particular automatic learning or "machine learning" (ML), constitute both a challenge and a great opportunity in numerous scientific, technical, and clinical disciplines. Specific applications in the study of multiple sclerosis (MS) have been no exception, and constitute an area of increasing interest in recent years.

OBJECTIVE

We present a systematic review of the application of ML algorithms in MS.

MATERIALS AND METHODS

We used the PubMed search engine, which allows free access to the MEDLINE medical database, to identify studies including the keywords "machine learning" and "multiple sclerosis." We excluded review articles, studies written in languages other than English or Spanish, and studies that were mainly technical and did not specifically apply to MS. The final selection included 76 articles, and 38 were rejected.

CONCLUSIONS

After the review process, we established 4 main applications of ML in MS: 1) classifying MS subtypes; 2) distinguishing patients with MS from healthy controls and individuals with other diseases; 3) predicting progression and response to therapeutic interventions; and 4) other applications. Results found to date have shown that ML algorithms may offer great support for health professionals both in clinical settings and in research into MS.

Automatic brain lesion segmentation on standard magnetic resonance images: a scoping review

OBJECTIVES

Medical image analysis practices face challenges that can potentially be addressed with algorithm-based segmentation tools. In this study, we map the field of automatic MR brain lesion segmentation to understand the clinical applicability of prevalent methods and study designs, as well as challenges and limitations in the field.

DESIGN

Scoping review.

SETTING

Three databases (PubMed, IEEE Xplore and Scopus) were searched with tailored queries. Studies were included based on predefined criteria. Emerging themes during consecutive title, abstract, methods and whole-text screening were identified. The full-text analysis focused on materials, preprocessing, performance evaluation and comparison.

RESULTS

Out of 2990 unique articles identified through the search, 441 articles met the eligibility criteria, with an estimated growth rate of 10% per year. We present a general overview and trends in the field with regard to publication sources, segmentation principles used and types of lesions. Algorithms are predominantly evaluated by measuring the agreement of segmentation results with a trusted reference. Few articles describe measures of clinical validity.

CONCLUSIONS

The observed reporting practices leave room for improvement with a view to studying replication, method comparison and clinical applicability. To promote this improvement, we propose a list of recommendations for future studies in the field.

[Artificial Intelligence in Radiology - Definition, Potential and Challenges]

Artificial Intelligence in Radiology - Definition, Potential and Challenges Abstract. Artificial Intelligence (AI) is omnipresent. It has neatly permeated our daily life, even if we are not always fully aware of its ubiquitous presence. The healthcare sector in particular is experiencing a revolution which will change our daily routine considerably in the near future. Due to its advanced digitization and its historical technical affinity radiology is especially prone to these developments. But what exactly is AI and what makes AI so potent that established medical disciplines such as radiology worry about their future job perspectives? What are the assets of AI in radiology today - and what are the major challenges? This review article tries to give some answers to these questions.

Biomimetic phantom with anatomical accuracy for evaluating brain volumetric measurements with magnetic resonance imaging

Purpose: Brain image volumetric measurements (BVM) methods have been used to quantify brain tissue volumes using magnetic resonance imaging (MRI) when investigating abnormalities. Although BVM methods are widely used, they need to be evaluated to quantify their reliability. Currently, the gold-standard reference to evaluate a BVM is usually manual labeling measurement. Manual volume labeling is a time-consuming and expensive task, but the confidence level ascribed to this method is not absolute. We describe and evaluate a biomimetic brain phantom as an alternative for the manual validation of BVM. Methods: We printed a three-dimensional (3D) brain mold using an MRI of a three-year-old boy diagnosed with Sturge-Weber syndrome. Then we prepared three different mixtures of styrene-ethylene/butylene-styrene gel and paraffin to mimic white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF). The mold was filled by these three mixtures with known volumes. We scanned the brain phantom using two MRI scanners, 1.5 and 3.0 Tesla. Our suggestion is a new challenging model to evaluate the BVM which includes the measured volumes of the phantom compartments and its MRI. We investigated the performance of an automatic BVM, i.e., the expectation-maximization (EM) method, to estimate its accuracy in BVM. Results: The automatic BVM results using the EM method showed a relative error (regarding the phantom volume) of 0.08, 0.03, and 0.13 ( ± 0.03 uncertainty) percentages of the GM, CSF, and WM volume, respectively, which was in good agreement with the results reported using manual segmentation. Conclusions: The phantom can be a potential quantifier for a wide range of segmentation methods.

Diverse Applications of Artificial Intelligence in Neuroradiology

Recent advances in artificial intelligence (AI) and deep learning (DL) hold promise to augment neuroimaging diagnosis for patients with brain tumors and stroke. Here, the authors review the diverse landscape of emerging neuroimaging applications of AI, including workflow optimization, lesion segmentation, and precision education. Given the many modalities used in diagnosing neurologic diseases, AI may be deployed to integrate across modalities (MR imaging, computed tomography, PET, electroencephalography, clinical and laboratory findings), facilitate crosstalk among specialists, and potentially improve diagnosis in patients with trauma, multiple sclerosis, epilepsy, and neurodegeneration. Together, there are myriad applications of AI for neuroradiology."

Serum GFAP in multiple sclerosis: correlation with disease type and MRI markers of disease severity

Neurofilament light chain (NfL) has been demonstrated to correlate with multiple sclerosis disease severity as well as treatment response. Nevertheless, additional serum biomarkers are still needed to better differentiate disease activity from disease progression. The aim of our study was to assess serum glial fibrillary acid protein (s-GFAP) and neurofilament light chain (s-NfL) in a cohort of 129 multiple sclerosis (MS) patients. Eighteen primary progressive multiple sclerosis (PPMS) and 111 relapsing remitting MS (RRMS) were included. We showed that these 2 biomarkers were significantly correlated with each other (R = 0.72, p < 0.001). Moreover, both biomarkers were higher in PPMS than in RRMS even if multivariate analysis only confirmed this difference for s-GFAP (130.3 ± 72.8 pg/ml vs 83.4 ± 41.1 pg/ml, p = 0.008). Finally, s-GFAP was correlated with white matter lesion load and inversely correlated with WM and GM volume. Our results seem to confirm the added value of s-GFAP in the context of multiple sclerosis.

Multiple sclerosis cortical and WM lesion segmentation at 3T MRI: a deep learning method based on FLAIR and MP2RAGE

The presence of cortical lesions in multiple sclerosis patients has emerged as an important biomarker of the disease. They appear in the earliest stages of the illness and have been shown to correlate with the severity of clinical symptoms. However, cortical lesions are hardly visible in conventional magnetic resonance imaging (MRI) at 3T, and thus their automated detection has been so far little explored. In this study, we propose a fully-convolutional deep learning approach, based on the 3D U-Net, for the automated segmentation of cortical and white matter lesions at 3T. For this purpose, we consider a clinically plausible MRI setting consisting of two MRI contrasts only: one conventional T2-weighted sequence (FLAIR), and one specialized T1-weighted sequence (MP2RAGE). We include 90 patients from two different centers with a total of 728 and 3856 gray and white matter lesions, respectively. We show that two reference methods developed for white matter lesion segmentation are inadequate to detect small cortical lesions, whereas our proposed framework is able to achieve a detection rate of 76% for both cortical and white matter lesions with a false positive rate of 29% in comparison to manual segmentation. Further results suggest that our framework generalizes well for both types of lesion in subjects acquired in two hospitals with different scanners.

Improved Cervical Cord Lesion Detection with 3D-MP2RAGE Sequence in Patients with Multiple Sclerosis

Spinal cord lesions have a real diagnostic and prognostic role in multiple sclerosis. Thus, optimizing their detection on MR imaging has become a central issue with direct therapeutic impact. In this study, we compared the 3D-MP2RAGE sequence with the conventional Magnetic Resonance Imaging in Multiple Sclerosis (MAGNIMS) set for cervical cord lesion detection in 28 patients with multiple sclerosis. 3D-MP2RAGE allowed better detection of cervical lesions (+62%) in this population, with better confidence, due to optimized contrast and high spatial resolution.

CVSnet: A machine learning approach for automated central vein sign assessment in multiple sclerosis

The central vein sign (CVS) is an efficient imaging biomarker for multiple sclerosis (MS) diagnosis, but its application in clinical routine is limited by inter-rater variability and the expenditure of time associated with manual assessment. We describe a deep learning-based prototype for automated assessment of the CVS in white matter MS lesions using data from three different imaging centers. We retrospectively analyzed data from 3 T magnetic resonance images acquired on four scanners from two different vendors, including adults with MS (n = 42), MS mimics (n = 33, encompassing 12 distinct neurological diseases mimicking MS) and uncertain diagnosis (n = 5). Brain white matter lesions were manually segmented on FLAIR* images. Perivenular assessment was performed according to consensus guidelines and used as ground truth, yielding 539 CVS-positive (CVS⁺ ) and 448 CVS-negative (CVS^- ) lesions. A 3D convolutional neural network ("CVSnet") was designed and trained on 47 datasets, keeping 33 for testing. FLAIR* lesion patches of CVS⁺ /CVS^- lesions were used for training and validation (n = 375/298) and for testing (n = 164/150). Performance was evaluated lesion-wise and subject-wise and compared with a state-of-the-art vesselness filtering approach through McNemar's test. The proposed CVSnet approached human performance, with lesion-wise median balanced accuracy of 81%, and subject-wise balanced accuracy of 89% on the validation set, and 91% on the test set. The process of CVS assessment, in previously manually segmented lesions, was ~ 600-fold faster using the proposed CVSnet compared with human visual assessment (test set: 4 seconds vs. 40 minutes). On the validation and test sets, the lesion-wise performance outperformed the vesselness filter method (P < 0.001). The proposed deep learning prototype shows promising performance in differentiating MS from its mimics. Our approach was evaluated using data from different hospitals, enabling larger multicenter trials to evaluate the benefit of introducing the CVS marker into MS diagnostic criteria.

Automated Detection and Segmentation of Multiple Sclerosis Lesions Using Ultra-High-Field MP2RAGE

OBJECTIVES

The aim of this study was to develop a new automated segmentation method of white matter (WM) and cortical multiple sclerosis (MS) lesions visible on magnetization-prepared 2 inversion-contrast rapid gradient echo (MP2RAGE) images acquired at 7 T MRI.

MATERIALS AND METHODS

The proposed prototype (MSLAST [Multiple Sclerosis Lesion Analysis at Seven Tesla]) takes as input a single image contrast derived from the 7T MP2RAGE prototype sequence and is based on partial volume estimation and topological constraints. First, MSLAST performs a skull-strip of MP2RAGE images and computes tissue concentration maps for WM, gray matter (GM), and cerebrospinal fluid (CSF) using a partial volume model of tissues within each voxel. Second, MSLAST performs (1) connected-component analysis to GM and CSF concentration maps to classify small isolated components as MS lesions; (2) hole-filling in the WM concentration map to classify areas with low WM concentration surrounded by WM (ie, MS lesions); and (3) outlier rejection to the WM mask to improve the classification of small WM lesions. Third, MSLAST unifies the 3 maps obtained from 1, 2, and 3 processing steps to generate a global lesion mask.

RESULTS

Quantitative and qualitative assessments were performed using MSLAST in 25 MS patients from 2 research centers. Overall, MSLAST detected a median of 71% of MS lesions, specifically 74% of WM and 58% of cortical lesions, when a minimum lesion size of 6 μL was considered. The median false-positive rate was 40%. When a 15 μL minimal lesions size was applied, which is the approximation of the minimal size recommended for 1.5/3 T images, the median detection rate was 80% for WM and 63% for cortical lesions, respectively, and the median false-positive rate was 33%. We observed high correlation between MSLAST and manual segmentations (Spearman rank correlation coefficient, ρ = 0.91), although MSLAST underestimated the total lesion volume (average difference of 1.1 mL), especially in patients with high lesion loads. MSLAST also showed good scan-rescan repeatability within the same session with an average absolute volume difference and F1 score of 0.38 ± 0.32 mL and 84%, respectively.

CONCLUSIONS

We propose a new methodology to facilitate the segmentation of WM and cortical MS lesions at 7 T MRI, our approach uses a single MP2RAGE scan and may be of special interest to clinicians and researchers.

Automated volumetric assessment with artificial neural networks might enable a more accurate assessment of disease burden in patients with multiple sclerosis

OBJECTIVES

Patients with multiple sclerosis (MS) regularly undergo MRI for assessment of disease burden. However, interpretation may be time consuming and prone to intra- and interobserver variability. Here, we evaluate the potential of artificial neural networks (ANN) for automated volumetric assessment of MS disease burden and activity on MRI.

METHODS

A single-institutional dataset with 334 MS patients (334 MRI exams) was used to develop and train an ANN for automated identification and volumetric segmentation of T2/FLAIR-hyperintense and contrast-enhancing (CE) lesions. Independent testing was performed in a single-institutional longitudinal dataset with 82 patients (266 MRI exams). We evaluated lesion detection performance (F1 scores), lesion segmentation agreement (DICE coefficients), and lesion volume agreement (concordance correlation coefficients [CCC]). Independent evaluation was performed on the public ISBI-2015 challenge dataset.

RESULTS

The F1 score was maximized in the training set at a detection threshold of 7 mm³ for T2/FLAIR lesions and 14 mm³ for CE lesions. In the training set, mean F1 scores were 0.867 for T2/FLAIR lesions and 0.636 for CE lesions, as compared to 0.878 for T2/FLAIR lesions and 0.715 for CE lesions in the test set. Using these thresholds, the ANN yielded mean DICE coefficients of 0.834 and 0.878 for segmentation of T2/FLAIR and CE lesions in the training set (fivefold cross-validation). Corresponding DICE coefficients in the test set were 0.846 for T2/FLAIR lesions and 0.908 for CE lesions, and the CCC was ≥ 0.960 in each dataset.

CONCLUSIONS

Our results highlight the capability of ANN for quantitative state-of-the-art assessment of volumetric lesion load on MRI and potentially enable a more accurate assessment of disease burden in patients with MS.

KEY POINTS

• Artificial neural networks (ANN) can accurately detect and segment both T2/FLAIR and contrast-enhancing MS lesions in MRI data. • Performance of the ANN was consistent in a clinically derived dataset, with patients presenting all possible disease stages in MRI scans acquired from standard clinical routine rather than with high-quality research sequences. • Computer-aided evaluation of MS with ANN could streamline both clinical and research procedures in the volumetric assessment of MS disease burden as well as in lesion detection.

Are multi-contrast magnetic resonance images necessary for segmenting multiple sclerosis brains? A large cohort study based on deep learning

BACKGROUND

Magnetic resonance images with multiple contrasts or sequences are commonly used for segmenting brain tissues, including lesions, in multiple sclerosis (MS). However, acquisition of images with multiple contrasts increases the scan time and complexity of the analysis, possibly introducing factors that could compromise segmentation quality.

OBJECTIVE

To investigate the effect of various combinations of multi-contrast images as input on the segmented volumes of gray (GM) and white matter (WM), cerebrospinal fluid (CSF), and lesions using a deep neural network.

METHODS

U-net, a fully convolutional neural network was used to automatically segment GM, WM, CSF, and lesions in 1000 MS patients. The input to the network consisted of 15 combinations of FLAIR, T1-, T2-, and proton density-weighted images. The Dice similarity coefficient (DSC) was evaluated to assess the segmentation performance. For lesions, true positive rate (TPR) and false positive rate (FPR) were also evaluated. In addition, the effect of lesion size on lesion segmentation was investigated.

RESULTS

Highest DSC was observed for all the tissue volumes, including lesions, when the input was combination of all four image contrasts. All other input combinations that included FLAIR also provided high DSC for all tissue classes. However, the quality of lesion segmentation showed strong dependence on the input images. The DSC and TPR values for inputs with the four contrast combination and FLAIR alone were very similar, but FLAIR showed a moderately higher FPR for lesion size <100 μl. For lesions smaller than 20 μl all image combinations resulted in poor performance. The segmentation quality improved with lesion size.

CONCLUSIONS

Best performance for segmented tissue volumes was obtained with all four image contrasts as the input, and comparable performance was attainable with FLAIR only as the input, albeit with a moderate increase in FPR for small lesions. This implies that acquisition of only FLAIR images provides satisfactory tissue segmentation. Lesion segmentation was poor for very small lesions and improved rapidly with lesion size.

Episodic memory decline in Parkinson' s disease: relation with white matter hyperintense lesions and influence of quantification method

The relation of white matter hyperintense lesions to episodic memory impairment in patients with Parkinson's disease (PD) is still controversial. We aimed at evaluating the relation between white matter hyperintense lesions and episodic memory decline in patients with PD. In this multicentric prospective study, twenty-one normal controls, 15 PD patients without mild cognitive impairment (MCI) and 13 PD patients with MCI were selected to conduct a clinico-radiological correlation analysis. Performance during episodic memory testing, age-related white matter changes score, total manual and automated white matter hyperintense lesions volume and lobar white matter hyperintense lesions volumes were compared between groups using the Kruskal-Wallis and Wilcoxon signed-rank tests, and correlations were assessed using the Spearman test. MCI PD patients had impaired free recall. They also had higher total, left prefrontal and left temporal white matter hyperintense lesions volumes than normal controls. Free recall performance was negatively correlated with the total white matter hyperintense lesions volume, either manually or automatically delineated, but not with the age-related white matter changes score. Using automated segmentation, both the left prefrontal and temporal white matter hyperintense lesions volumes were negatively correlated with the free recall performance. Early episodic memory impairment in MCI PD patients may be related to white matter hyperintense lesions, mainly in the prefrontal and temporal lobes. This relation is influenced by the method used for white matter hyperintense lesions quantification. Automated volumetry allows for detecting those changes.

Convolutional Neural Network for Automated FLAIR Lesion Segmentation on Clinical Brain MR Imaging

BACKGROUND AND PURPOSE

Most brain lesions are characterized by hyperintense signal on FLAIR. We sought to develop an automated deep learning-based method for segmentation of abnormalities on FLAIR and volumetric quantification on clinical brain MRIs across many pathologic entities and scanning parameters. We evaluated the performance of the algorithm compared with manual segmentation and existing automated methods.

MATERIALS AND METHODS

We adapted a U-Net convolutional neural network architecture for brain MRIs using 3D volumes. This network was retrospectively trained on 295 brain MRIs to perform automated FLAIR lesion segmentation. Performance was evaluated on 92 validation cases using Dice scores and voxelwise sensitivity and specificity, compared with radiologists' manual segmentations. The algorithm was also evaluated on measuring total lesion volume.

RESULTS

Our model demonstrated accurate FLAIR lesion segmentation performance (median Dice score, 0.79) on the validation dataset across a large range of lesion characteristics. Across 19 neurologic diseases, performance was significantly higher than existing methods (Dice, 0.56 and 0.41) and approached human performance (Dice, 0.81). There was a strong correlation between the predictions of lesion volume of the algorithm compared with true lesion volume (ρ = 0.99). Lesion segmentations were accurate across a large range of image-acquisition parameters on >30 different MR imaging scanners.

CONCLUSIONS

A 3D convolutional neural network adapted from a U-Net architecture can achieve high automated FLAIR segmentation performance on clinical brain MR imaging across a variety of underlying pathologies and image acquisition parameters. The method provides accurate volumetric lesion data that can be incorporated into assessments of disease burden or into radiologic reports.

Longitudinal analysis of white matter and cortical lesions in multiple sclerosis

PURPOSE

The goals of this study were to assess the performance of a novel lesion segmentation tool for longitudinal analyses, as well as to validate the generated lesion progression map between two time points using conventional and non-conventional MR sequences.

MATERIAL AND METHODS

The lesion segmentation approach was evaluated with (LeMan-PV) and without (LeMan) the partial volume framework using "conventional" and "non-conventional" MR imaging in a two-year follow-up prospective study of 32 early RRMS patients. Manual segmentations of new, enlarged, shrunken, and stable lesions were used to evaluate the performance of the method variants. The true positive rate was estimated for those lesion evolutions in both white matter and cortex. The number of false positives was compared with two strategies for longitudinal analyses. New lesion tissue volume estimation was evaluated using Bland-Altman plots. Wilcoxon signed-rank test was used to evaluate the different setups.

RESULTS

The best median of the true positive rate was obtained using LeMan-PV with non-conventional sequences (P < .05): 87%, 87%, 100%, 83%, for new, enlarged, shrunken, and stable WM lesions, and 50%, 60%, 50%, 80%, for new, enlarged, shrunken, and stable cortical lesions, respectively. Most of the missed lesions were below the mean lesion size in each category. Lesion progression maps presented a median of 0 false positives (range:0-9) and the partial volume framework improved the volume estimation of new lesion tissue.

CONCLUSION

LeMan-PV exhibited the best performance in the detection of new, enlarged, shrunken and stable WM lesions. The method showed lower performance in the detection of cortical lesions, likely due to their low occurrence, small size and low contrast with respect to surrounding tissues. The proposed lesion progression map might be useful in clinical trials or clinical routine.

What the radiologist should know about artificial intelligence - an ESR white paper

This paper aims to provide a review of the basis for application of AI in radiology, to discuss the immediate ethical and professional impact in radiology, and to consider possible future evolution.Even if AI does add significant value to image interpretation, there are implications outside the traditional radiology activities of lesion detection and characterisation. In radiomics, AI can foster the analysis of the features and help in the correlation with other omics data. Imaging biobanks would become a necessary infrastructure to organise and share the image data from which AI models can be trained. AI can be used as an optimising tool to assist the technologist and radiologist in choosing a personalised patient's protocol, tracking the patient's dose parameters, providing an estimate of the radiation risks. AI can also aid the reporting workflow and help the linking between words, images, and quantitative data. Finally, AI coupled with CDS can improve the decision process and thereby optimise clinical and radiological workflow.

Iron related changes in MS lesions and their validity to characterize MS lesion types and dynamics with Ultra-high field magnetic resonance imaging

Iron accumulates with age in the normal human brain. This process is altered at several levels in the brain of multiple sclerosis (MS) patients. Since iron is mainly stored in oligodendrocytes and myelin in the normal brain, its liberation in demyelinating lesions may amplify tissue damage in demyelinating lesions and its uptake in macrophages and microglia may help to more precisely define activity stages of the lesions. In addition, glia cells change their iron import, export and storage properties in MS lesions, which is reflected by alterations in the expression of iron transport molecules. Changes of iron distribution in the brain can be reliably detected by MRI, particularly upon application of Ultra‐high magnetic field (7 Tesla). Iron‐sensitive MRI allows to more accurately distinguish the lesions in MS from those in other inflammatory brain diseases, to visualize a subset of slowly expanding lesions in the progressive stage of MS and to increase the sensitivity for lesion detection in the gray matter, such as the cerebral cortex or deep gray matter nuclei.

Asymmetric Loss Functions and Deep Densely Connected Networks for Highly Imbalanced Medical Image Segmentation: Application to Multiple Sclerosis Lesion Detection

Fully convolutional deep neural networks have been asserted to be fast and precise frameworks with great potential in image segmentation. One of the major challenges in training such networks raises when data is unbalanced, which is common in many medical imaging applications such as lesion segmentation where lesion class voxels are often much lower in numbers than non-lesion voxels. A trained network with unbalanced data may make predictions with high precision and low recall, being severely biased towards the non-lesion class which is particularly undesired in most medical applications where false negatives are actually more important than false positives. Various methods have been proposed to address this problem including two step training, sample re-weighting, balanced sampling, and more recently similarity loss functions, and focal loss. In this work we trained fully convolutional deep neural networks using an asymmetric similarity loss function to mitigate the issue of data imbalance and achieve much better trade-off between precision and recall. To this end, we developed a 3D fully convolutional densely connected network (FC-DenseNet) with large overlapping image patches as input and an asymmetric similarity loss layer based on Tversky index (using F β scores). We used large overlapping image patches as inputs for intrinsic and extrinsic data augmentation, a patch selection algorithm, and a patch prediction fusion strategy using B-spline weighted soft voting to account for the uncertainty of prediction in patch borders. We applied this method to multiple sclerosis (MS) lesion segmentation based on two different datasets of MSSEG 2016 and ISBI longitudinal MS lesion segmentation challenge, where we achieved average Dice similarity coefficients of 69.9% and 65.74%, respectively, achieving top performance in both challenges. We compared the performance of our network trained with F β loss, focal loss, and generalized Dice loss (GDL) functions. Through September 2018 our network trained with focal loss ranked first according to the ISBI challenge overall score and resulted in the lowest reported lesion false positive rate among all submitted methods. Our network trained with the asymmetric similarity loss led to the lowest surface distance and the best lesion true positive rate that is arguably the most important performance metric in a clinical decision support system for lesion detection. The asymmetric similarity loss function based on F β scores allows training networks that make a better balance between precision and recall in highly unbalanced image segmentation. We achieved superior performance in MS lesion segmentation using a patchwise 3D FC-DenseNet with a patch prediction fusion strategy, trained with asymmetric similarity loss functions.

3D double inversion recovery MR imaging: Clinical applications and usefulness in a wide spectrum of central nervous system diseases

Double inversion recovery (DIR) imaging provides two inversion pulses that attenuate signals from cerebrospinal fluid and normal white matter. This review was undertaken to describe the principle of the DIR sequence, the clinical applications of 3D DIR in various central nervous system diseases and the clinical benefits of the 3D DIR compared with those of other MR sequences. 3D DIR imaging provides better lesion conspicuity and topography than other MR techniques. It is particularly useful for diagnosing the following disease entities: cortical and subcortical abnormalities such as multiple sclerosis, cortical microinfarcts and cortical development anomalies; sulcal abnormalities such as meningitis and subacute/chronic subarachnoid hemorrhage; and optic neuritis caused by multiple sclerosis or neuromyelitis optica.

Magnetic Resonance Imaging in Multiple Sclerosis

Since its technical development in the early 1980s, magnetic resonance imaging (MRI) has quickly been adopted as an essential tool in supporting the diagnosis, longitudinal monitoring, evaluation of therapeutic response, and scientific investigations in multiple sclerosis (MS). The clinical usage of MRI has increased in parallel with technical innovations in the technique itself; the widespread adoption of clinically routine MRI at 1.5T has allowed sensitive qualitative and quantitative assessments of macroscopic central nervous system (CNS) inflammatory demyelinating lesions and tissue atrophy. However, conventional MRI lesion measures lack specificity for the underlying MS pathology and only weakly correlate with clinical status. Higher field strength units and newer, advanced MRI techniques offer increased sensitivity and specificity in the detection of disease activity and disease severity. This review summarizes the current status and future prospects regarding the role of MRI in the characterization of MS-related brain and spinal cord involvement.

Partial volume-aware assessment of multiple sclerosis lesions

White-matter lesion count and volume estimation are key to the diagnosis and monitoring of multiple sclerosis (MS). Automated MS lesion segmentation methods that have been proposed in the past 20 years reach their limits when applied to patients in early disease stages characterized by low lesion load and small lesions. We propose an algorithm to automatically assess MS lesion load (number and volume) while taking into account the mixing of healthy and lesional tissue in the image voxels due to partial volume effects. The proposed method works on 3D MPRAGE and 3D FLAIR images as obtained from current routine MS clinical protocols. The method was evaluated and compared with manual segmentation on a cohort of 39 early-stage MS patients with low disability, and showed higher Dice similarity coefficients (median DSC = 0.55) and higher detection rate (median DR = 61%) than two widely used methods (median DSC = 0.50, median DR < 45%) for automated MS lesion segmentation. We argue that this is due to the higher performance in segmentation of small lesions, which are inherently prone to partial volume effects.

•

Modeling the partial volume improves lesion volumetric measurements.

•

Higher detection of small lesions inherently prone to partial volume effects.

•

Partial volume effects should be taken into account in early stages of MS.