Fast refacing of MR images with a generative neural network lowers re-identification risk and preserves volumetric consistency.

With the rise of open data, identifiability of individuals based on 3D renderings obtained from routine structural magnetic resonance imaging (MRI) scans of the head has become a growing privacy concern. To protect subject privacy, several algorithms have been developed to de-identify imaging data using blurring, defacing or refacing. Completely removing facial structures provides the best re-identification protection but can significantly impact post-processing steps, like brain morphometry. As an alternative, refacing methods that replace individual facial structures with generic templates have a lower effect on the geometry and intensity distribution of original scans, and are able to provide more consistent post-processing results by the price of higher re-identification risk and computational complexity. In the current study, we propose a novel method for anonymized face generation for defaced 3D T1-weighted scans based on a 3D conditional generative adversarial network. To evaluate the performance of the proposed de-identification tool, a comparative study was conducted between several existing defacing and refacing tools, with two different segmentation algorithms (FAST and Morphobox). The aim was to evaluate (i) impact on brain morphometry reproducibility, (ii) re-identification risk, (iii) balance between (i) and (ii), and (iv) the processing time. The proposed method takes 9 s for face generation and is suitable for recovering consistent post-processing results after defacing.

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.

Systematic Estimation of Treatment Effect on Hospitalization Risk as a Drug Repurposing Screening Method.

Drug repurposing (DR) intends to identify new uses for approved medications outside their original indication. Computational methods for finding DR candidates usually rely on prior biological and chemical information on a specific drug or target but rarely utilize real-world observations. In this work, we propose a simple and effective systematic screening approach to measure medication impact on hospitalization risk based on large-scale observational data. We use common classification systems to group drugs and diseases into broader functional categories and test for non-zero effects in each drug-disease category pair. Treatment effects on the hospitalization risk of an individual disease are obtained by combining widely used methods for causal inference and time-to-event modelling. 6468 drug-disease pairs were tested using data from the UK Biobank, focusing on cardiovascular, metabolic, and respiratory diseases. We determined key parameters to reduce the number of spurious correlations and identified 7 statistically significant associations of reduced hospitalization risk after correcting for multiple testing. Some of these associations were already reported in other studies, including new potential applications for cardioselective beta-blockers and thiazides. We also found evidence for proton pump inhibitor side effects and multiple possible associations for anti-diabetic drugs. Our work demonstrates the applicability of the present screening approach and the utility of real-world data for identifying potential DR candidates.

Inter-regional correlation estimators for functional magnetic resonance imaging.

Functional magnetic resonance imaging (fMRI) functional connectivity between brain regions is often computed using parcellations defined by functional or structural atlases. Typically, some kind of voxel averaging is performed to obtain a single temporal correlation estimate per region pair. However, several estimators can be defined for this task, with various assumptions and degrees of robustness to local noise, global noise, and region size. In this paper, we systematically present and study the properties of 9 different functional connectivity estimators taking into account the spatial structure of fMRI data, based on a simple fMRI data spatial model. These include 3 existing estimators and 6 novel estimators. We demonstrate the empirical properties of the estimators using synthetic, animal, and human data, in terms of graph structure, repeatability and reproducibility, discriminability, dependence on region size, as well as local and global noise robustness. We prove analytically the link between regional intra-correlation and inter-region correlation, and show that the choice of estimator has a strong influence on inter-correlation values. Some estimators, including the commonly used correlation of averages (ca), are positively biased, and have more dependence to region size and intra-correlation than robust alternatives, resulting in spatially-dependent bias. We define the new local correlation of averages estimator with better theoretical guarantees, lower bias, significantly lower dependence on region size (Spearman correlation 0.40 vs 0.55, paired t-test T=27.2, p=1.1e-47), at negligible cost to discriminative power, compared to the ca estimator. The difference in connectivity pattern between the estimators is not distributed uniformly throughout the brain, but rather shows a clear ventral-dorsal gradient, suggesting that region size and intra-correlation plays an important role in shaping functional networks defined using the ca estimator, and leading to non-trivial differences in their connectivity structure. We provide an open source R package and equivalent Python implementation to facilitate the use of the new estimators, together with preprocessed rat time-series.

Towards Automated Brain Aneurysm Detection in TOF-MRA: Open Data, Weak Labels, and Anatomical Knowledge.

Brain aneurysm detection in Time-Of-Flight Magnetic Resonance Angiography (TOF-MRA) has undergone drastic improvements with the advent of Deep Learning (DL). However, performances of supervised DL models heavily rely on the quantity of labeled samples, which are extremely costly to obtain. Here, we present a DL model for aneurysm detection that overcomes the issue with "weak" labels: oversized annotations which are considerably faster to create. Our weak labels resulted to be four times faster to generate than their voxel-wise counterparts. In addition, our model leverages prior anatomical knowledge by focusing only on plausible locations for aneurysm occurrence. We first train and evaluate our model through cross-validation on an in-house TOF-MRA dataset comprising 284 subjects (170 females / 127 healthy controls / 157 patients with 198 aneurysms). On this dataset, our best model achieved a sensitivity of 83%, with False Positive (FP) rate of 0.8 per patient. To assess model generalizability, we then participated in a challenge for aneurysm detection with TOF-MRA data (93 patients, 20 controls, 125 aneurysms). On the public challenge, sensitivity was 68% (FP rate = 2.5), ranking 4th/18 on the open leaderboard. We found no significant difference in sensitivity between aneurysm risk-of-rupture groups (p = 0.75), locations (p = 0.72), or sizes (p = 0.15). Data, code and model weights are released under permissive licenses. We demonstrate that weak labels and anatomical knowledge can alleviate the necessity for prohibitively expensive voxel-wise annotations.

Cerebrospinal Fluid Cortisol and Dehydroepiandrosterone Sulfate, Alzheimer's Disease Pathology, and Cognitive Decline.

Introduction: Elevated cortisol levels have been reported in Alzheimer's disease (AD) and may accelerate the development of brain pathology and cognitive decline. Dehydroepiandrosterone sulfate (DHEAS) has anti-glucocorticoid effects and it may be involved in the AD pathophysiology. Objectives: To investigate associations of cerebrospinal fluid (CSF) cortisol and DHEAS levels with (1) cognitive performance at baseline; (2) CSF biomarkers of amyloid pathology (as assessed by CSF Aß levels), neuronal injury (as assessed by CSF tau), and tau hyperphosphorylation (as assessed by CSF p-tau); (3) regional brain volumes; and (4) clinical disease progression. Materials and Methods: Individuals between 49 and 88 years (n = 145) with mild cognitive impairment or dementia or with normal cognition were included. Clinical scores, AD biomarkers, brain MRI volumetry along with CSF cortisol and DHEAS were obtained at baseline. Cognitive and functional performance was re-assessed at 18 and 36 months from baseline. We also assessed the following covariates: apolipoprotein E (APOE) genotype, BMI, and education. We used linear regression and mixed models to address associations of interest. Results: Higher CSF cortisol was associated with poorer global cognitive performance and higher disease severity at baseline. Cortisol and cortisol/DHEAS ratio were positively associated with tau and p-tau CSF levels, and negatively associated with the amygdala and insula volumes at baseline. Higher CSF cortisol predicted more pronounced cognitive decline and clinical disease progression over 36 months. Higher CSF DHEAS predicted more pronounced disease progression over 36 months. Conclusion: Increased cortisol in the CNS is associated with tau pathology and neurodegeneration, and with decreased insula and amygdala volume. Both CSF cortisol and DHEAS levels predict faster clinical disease progression. These results have implications for the identification of patients at risk of rapid decline as well as for the development of interventions targeting both neurodegeneration and clinical manifestations of AD.

Systemic and central nervous system neuroinflammatory signatures of neuropsychiatric symptoms and related cognitive decline in older people.

BACKGROUND: Neuroinflammation may contribute to psychiatric symptoms in older people, in particular in the context of Alzheimer's disease (AD). We sought to identify systemic and central nervous system (CNS) inflammatory alterations associated with neuropsychiatric symptoms (NPS); and to investigate their relationships with AD pathology and clinical disease progression. METHODS: We quantified a panel of 38 neuroinflammation and vascular injury markers in paired serum and cerebrospinal fluid (CSF) samples in a cohort of cognitively normal and impaired older subjects. We performed neuropsychiatric and cognitive evaluations and measured CSF biomarkers of AD pathology. Multivariate analysis determined serum and CSF neuroinflammatory alterations associated with NPS, considering cognitive status, AD pathology, and cognitive decline at follow-up visits. RESULTS: NPS were associated with distinct inflammatory profiles in serum, involving eotaxin-3, interleukin (IL)-6 and C-reactive protein (CRP); and in CSF, including soluble intracellular cell adhesion molecule-1 (sICAM-1), IL-8, 10-kDa interferon-γ-induced protein, and CRP. AD pathology interacted with CSF sICAM-1 in association with NPS. Presenting NPS was associated with subsequent cognitive decline which was mediated by CSF sICAM-1. CONCLUSIONS: Distinct systemic and CNS inflammatory processes are involved in the pathophysiology of NPS in older people. Neuroinflammation may explain the link between NPS and more rapid clinical disease progression.

Machine learning algorithms on eye tracking trajectories to classify patients with spatial neglect.

BACKGROUND AND OBJECTIVE: Eye-movement trajectories are rich behavioral data, providing a window on how the brain processes information. We address the challenge of characterizing signs of visuo-spatial neglect from saccadic eye trajectories recorded in brain-damaged patients with spatial neglect as well as in healthy controls during a visual search task. METHODS: We establish a standardized pre-processing pipeline adaptable to other task-based eye-tracker measurements. We use traditional machine learning algorithms together with deep convolutional networks (both 1D and 2D) to automatically analyze eye trajectories. RESULTS: Our top-performing machine learning models classified neglect patients vs. healthy individuals with an Area Under the ROC curve (AUC) ranging from 0.83 to 0.86. Moreover, the 1D convolutional neural network scores correlated with the degree of severity of neglect behavior as estimated with standardized paper-and-pencil tests and with the integrity of white matter tracts measured from Diffusion Tensor Imaging (DTI). Interestingly, the latter showed a clear correlation with the third branch of the superior longitudinal fasciculus (SLF), especially damaged in neglect. CONCLUSIONS: The study introduces new methods for both the pre-processing and the classification of eye-movement trajectories in patients with neglect syndrome. The proposed methods can likely be applied to other types of neurological diseases opening the possibility of new computer-aided, precise, sensitive and non-invasive diagnostic tools.

Multi-centre classification of functional neurological disorders based on resting-state functional connectivity.

BACKGROUND: Patients suffering from functional neurological disorder (FND) experience disabling neurological symptoms not caused by an underlying classical neurological disease (such as stroke or multiple sclerosis). The diagnosis is made based on reliable positive clinical signs, but clinicians often require additional time- and cost consuming medical tests and examinations. Resting-state functional connectivity (RS FC) showed its potential as an imaging-based adjunctive biomarker to help distinguish patients from healthy controls and could represent a "rule-in" procedure to assist in the diagnostic process. However, the use of RS FC depends on its applicability in a multi-centre setting, which is particularly susceptible to inter-scanner variability. The aim of this study was to test the robustness of a classification approach based on RS FC in a multi-centre setting. METHODS: This study aimed to distinguish 86 FND patients from 86 healthy controls acquired in four different centres using a multivariate machine learning approach based on whole-brain resting-state functional connectivity. First, previously published results were replicated in each centre individually (intra-centre cross-validation) and its robustness across inter-scanner variability was assessed by pooling all the data (pooled cross-validation). Second, we evaluated the generalizability of the method by using data from each centre once as a test set, and the data from the remaining centres as a training set (inter-centre cross-validation). RESULTS: FND patients were successfully distinguished from healthy controls in the replication step (accuracy of 74%) as well as in each individual additional centre (accuracies of 73%, 71% and 70%). The pooled cross validation confirmed that the classifier was robust with an accuracy of 72%. The results survived post-hoc adjustment for anxiety, depression, psychotropic medication intake, and symptom severity. The most discriminant features involved the angular- and supramarginal gyri, sensorimotor cortex, cingular- and insular cortex, and hippocampal regions. The inter-centre validation step did not exceed chance level (accuracy below 50%). CONCLUSIONS: The results demonstrate the applicability of RS FC to correctly distinguish FND patients from healthy controls in different centres and its robustness against inter-scanner variability. In order to generalize its use across different centres and aim for clinical application, future studies should work towards optimization of acquisition parameters and include neurological and psychiatric control groups presenting with similar symptoms.

Preparing for a Second Attack: A Lesion Simulation Study on Network Resilience After Stroke.

BACKGROUND: Does the brain become more resilient after a first stroke to reduce the consequences of a new lesion? Although recurrent strokes are a major clinical issue, whether and how the brain prepares for a second attack is unknown. This is due to the difficulties to obtain an appropriate dataset of stroke patients with comparable lesions, imaged at the same interval after onset. Furthermore, timing of the recurrent event remains unpredictable. METHODS: Here, we used a novel clinical lesion simulation approach to test the hypothesis that resilience in brain networks increases during stroke recovery. Sixteen highly selected patients with a lesion restricted to the primary motor cortex were recruited. At 3 time points of the index event (10 days, 3 weeks, 3 months), we mimicked recurrent infarcts by deletion of nodes in brain networks (resting-state functional magnetic resonance imaging). Graph measures were applied to determine resilience (global efficiency after attack) and wiring cost (mean degree) of the network. RESULTS: At 10 days and 3 weeks after stroke, resilience was similar in patients and controls. However, at 3 months, although motor function had fully recovered, resilience to clinically representative simulated lesions was higher compared to controls (cortical lesion P=0.012; subcortical: P=0.009; cortico-subcortical: P=0.009). Similar results were found after random (P=0.012) and targeted (P=0.015) attacks. CONCLUSIONS: Our results suggest that, in this highly selected cohort of patients with lesions restricted to the primary motor cortex, brain networks reconfigure to increase resilience to future insults. Lesion simulation is an innovative approach, which may have major implications for stroke therapy. Individualized neuromodulation strategies could be developed to foster resilient network reconfigurations after a first stroke to limit the consequences of future attacks.

The Cardiomyocyte in Heart Failure with Preserved Ejection Fraction-Victim of Its Environment?

Heart failure (HF) with preserved left ventricular ejection fraction (HFpEF) is becoming the predominant form of HF. However, medical therapy that improves cardiovascular outcome in HF patients with almost normal and normal systolic left ventricular function, but diastolic dysfunction is missing. The cause of this unmet need is incomplete understanding of HFpEF pathophysiology, the heterogeneity of the patient population, and poor matching of therapeutic mechanisms and primary pathophysiological processes. Recently, animal models improved understanding of the pathophysiological role of highly prevalent and often concomitantly presenting comorbidity in HFpEF patients. Evidence from these animal models provide first insight into cellular pathophysiology not considered so far in HFpEF disease, promising that improved understanding may provide new therapeutical targets. This review merges observation from animal models and human HFpEF disease with the intention to converge cardiomyocytes pathophysiological aspects and clinical knowledge.

Data-driven myelin water imaging based on T1 and T2 relaxometry.

Long acquisition times preclude the application of multiecho spin echo (MESE) sequences for myelin water fraction (MWF) mapping in daily clinical practice. In search of alternative methods, previous studies of interest explored the biophysical modeling of MWF from measurements of different tissue properties that can be obtained in scan times shorter than those required for the MESE. In this work, a novel data-driven estimation of MWF maps from fast relaxometry measurements is proposed and investigated. T1 and T2 relaxometry maps were acquired in a cohort of 20 healthy subjects along with a conventional MESE sequence. Whole-brain quantitative mapping was achieved with a fast protocol in 6 min 24 s. Reference MWF maps were derived from the MESE sequence (TA = 11 min 17 s) and their data-driven estimation from relaxometry measurements was investigated using three different modeling strategies: two general linear models (GLMs) with linear and quadratic regressors, respectively; a random forest regression model; and two deep neural network architectures, a U-Net and a conditional generative adversarial network (cGAN). Models were validated using a 10-fold crossvalidation. The resulting maps were visually and quantitatively compared by computing the root mean squared error (RMSE) between the estimated and reference MWF maps, the intraclass correlation coefficients (ICCs) between corresponding MWF values in different brain regions, and by performing Bland-Altman analysis. Qualitatively, the estimated maps appear to generally provide a similar, yet more blurred MWF contrast in comparison with the reference, with the cGAN model best capturing MWF variabilities in small structures. By estimating the average adjusted coefficient of determination of the GLM with quadratic regressors, we showed that 87% of the variability in the MWF values can be explained by relaxation times alone. Further quantitative analysis showed an average RMSE smaller than 0.1% for all methods. The ICC was greater than 0.81 for all methods, and the bias smaller than 2.19%. It was concluded that this work confirms the notion that relaxometry parameters contain a large part of the information on myelin water and that MWF maps can be generated from T1 /T2 data with minimal error. Among the investigated modeling approaches, the cGAN provided maps with the best trade-off between accuracy and blurriness. Fast relaxometry, like the 6 min 24 s whole-brain protocol used in this work in conjunction with machine learning, may thus have the potential to replace time-consuming MESE acquisitions.

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

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.

Comparing methods of detecting and segmenting unruptured intracranial aneurysms on TOF-MRAS: The ADAM challenge.

Accurate detection and quantification of unruptured intracranial aneurysms (UIAs) is important for rupture risk assessment and to allow an informed treatment decision to be made. Currently, 2D manual measures used to assess UIAs on Time-of-Flight magnetic resonance angiographies (TOF-MRAs) lack 3D information and there is substantial inter-observer variability for both aneurysm detection and assessment of aneurysm size and growth. 3D measures could be helpful to improve aneurysm detection and quantification but are time-consuming and would therefore benefit from a reliable automatic UIA detection and segmentation method. The Aneurysm Detection and segMentation (ADAM) challenge was organised in which methods for automatic UIA detection and segmentation were developed and submitted to be evaluated on a diverse clinical TOF-MRA dataset. A training set (113 cases with a total of 129 UIAs) was released, each case including a TOF-MRA, a structural MR image (T1, T2 or FLAIR), annotation of any present UIA(s) and the centre voxel of the UIA(s). A test set of 141 cases (with 153 UIAs) was used for evaluation. Two tasks were proposed: (1) detection and (2) segmentation of UIAs on TOF-MRAs. Teams developed and submitted containerised methods to be evaluated on the test set. Task 1 was evaluated using metrics of sensitivity and false positive count. Task 2 was evaluated using dice similarity coefficient, modified hausdorff distance (95th percentile) and volumetric similarity. For each task, a ranking was made based on the average of the metrics. In total, eleven teams participated in task 1 and nine of those teams participated in task 2. Task 1 was won by a method specifically designed for the detection task (i.e. not participating in task 2). Based on segmentation metrics, the top two methods for task 2 performed statistically significantly better than all other methods. The detection performance of the top-ranking methods was comparable to visual inspection for larger aneurysms. Segmentation performance of the top ranking method, after selection of true UIAs, was similar to interobserver performance. The ADAM challenge remains open for future submissions and improved submissions, with a live leaderboard to provide benchmarking for method developments at https://adam.isi.uu.nl/.

Periinfarct rewiring supports recovery after primary motor cortex stroke.

After stroke restricted to the primary motor cortex (M1), it is uncertain whether network reorganization associated with recovery involves the periinfarct or more remote regions. We studied 16 patients with focal M1 stroke and hand paresis. Motor function and resting-state MRI functional connectivity (FC) were assessed at three time points: acute (<10 days), early subacute (3 weeks), and late subacute (3 months). FC correlates of recovery were investigated at three spatial scales, (i) ipsilesional non-infarcted M1, (ii) core motor network (M1, premotor cortex (PMC), supplementary motor area (SMA), and primary somatosensory cortex), and (iii) extended motor network including all regions structurally connected to the upper limb representation of M1. Hand dexterity was impaired only in the acute phase (P = 0.036). At a small spatial scale, clinical recovery was more frequently associated with connections involving ipsilesional non-infarcted M1 (Odds Ratio = 6.29; P = 0.036). At a larger scale, recovery correlated with increased FC strength in the core network compared to the extended motor network (rho = 0.71;P = 0.006). These results suggest that FC changes associated with motor improvement involve the perilesional M1 and do not extend beyond the core motor network. Core motor regions, and more specifically ipsilesional non-infarcted M1, could hence become primary targets for restorative therapies.

Plasma neurofilament light and phosphorylated tau 181 as biomarkers of Alzheimer's disease pathology and clinical disease progression.

BACKGROUND: To assess the performance of plasma neurofilament light (NfL) and phosphorylated tau 181 (p-tau181) to inform about cerebral Alzheimer's disease (AD) pathology and predict clinical progression in a memory clinic setting. METHODS: Plasma NfL and p-tau181, along with established cerebrospinal fluid (CSF) biomarkers of AD pathology, were measured in participants with normal cognition (CN) and memory clinic patients with cognitive impairment (mild cognitive impairment and dementia, CI). Clinical and neuropsychological assessments were performed at inclusion and follow-up visits at 18 and 36 months. Multivariate analysis assessed associations of plasma NfL and p-tau181 levels with AD, single CSF biomarkers, hippocampal volume, and clinical measures of disease progression. RESULTS: Plasma NfL levels were higher in CN participants with an AD CSF profile (defined by a CSF p-tau181/Aß1-42 > 0.0779) as compared with CN non-AD, while p-tau181 plasma levels were higher in CI patients with AD. Plasma NfL levels correlated with CSF tau and p-tau181 in CN, and with CSF tau in CI patients. Plasma p-tau181 correlated with CSF p-tau181 in CN and with CSF tau, p-tau181, Aß1-42, and Aß1-42/Aß1-40 in CI participants. Compared with a reference model, adding plasma p-tau181 improved the prediction of AD in CI patients while adding NfL did not. Adding p-tau181, but not NfL levels, to a reference model improved prediction of cognitive decline in CI participants. CONCLUSION: Plasma NfL indicates neurodegeneration while plasma p-tau181 levels can serve as a biomarker of cerebral AD pathology and cognitive decline. Their predictive performance depends on the presence of cognitive impairment.

To buy or not to buy-evaluating commercial AI solutions in radiology (the ECLAIR guidelines).

Artificial intelligence (AI) has made impressive progress over the past few years, including many applications in medical imaging. Numerous commercial solutions based on AI techniques are now available for sale, forcing radiology practices to learn how to properly assess these tools. While several guidelines describing good practices for conducting and reporting AI-based research in medicine and radiology have been published, fewer efforts have focused on recommendations addressing the key questions to consider when critically assessing AI solutions before purchase. Commercial AI solutions are typically complicated software products, for the evaluation of which many factors are to be considered. In this work, authors from academia and industry have joined efforts to propose a practical framework that will help stakeholders evaluate commercial AI solutions in radiology (the ECLAIR guidelines) and reach an informed decision. Topics to consider in the evaluation include the relevance of the solution from the point of view of each stakeholder, issues regarding performance and validation, usability and integration, regulatory and legal aspects, and financial and support services. KEY POINTS: • Numerous commercial solutions based on artificial intelligence techniques are now available for sale, and radiology practices have to learn how to properly assess these tools. • We propose a framework focusing on practical points to consider when assessing an AI solution in medical imaging, allowing all stakeholders to conduct relevant discussions with manufacturers and reach an informed decision as to whether to purchase an AI commercial solution for imaging applications. • Topics to consider in the evaluation include the relevance of the solution from the point of view of each stakeholder, issues regarding performance and validation, usability and integration, regulatory and legal aspects, and financial and support services.

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.