Contrast-enhanced to non-contrast-enhanced image translation to exploit a clinical data warehouse of T1-weighted brain MRI.

BACKGROUND: Clinical data warehouses provide access to massive amounts of medical images, but these images are often heterogeneous. They can for instance include images acquired both with or without the injection of a gadolinium-based contrast agent. Harmonizing such data sets is thus fundamental to guarantee unbiased results, for example when performing differential diagnosis. Furthermore, classical neuroimaging software tools for feature extraction are typically applied only to images without gadolinium. The objective of this work is to evaluate how image translation can be useful to exploit a highly heterogeneous data set containing both contrast-enhanced and non-contrast-enhanced images from a clinical data warehouse. METHODS: We propose and compare different 3D U-Net and conditional GAN models to convert contrast-enhanced T1-weighted (T1ce) into non-contrast-enhanced (T1nce) brain MRI. These models were trained using 230 image pairs and tested on 77 image pairs from the clinical data warehouse of the Greater Paris area. RESULTS: Validation using standard image similarity measures demonstrated that the similarity between real and synthetic T1nce images was higher than between real T1nce and T1ce images for all the models compared. The best performing models were further validated on a segmentation task. We showed that tissue volumes extracted from synthetic T1nce images were closer to those of real T1nce images than volumes extracted from T1ce images. CONCLUSION: We showed that deep learning models initially developed with research quality data could synthesize T1nce from T1ce images of clinical quality and that reliable features could be extracted from the synthetic images, thus demonstrating the ability of such methods to help exploit a data set coming from a clinical data warehouse.

Automatic motion artefact detection in brain T1-weighted magnetic resonance images from a clinical data warehouse using synthetic data.

Containing the medical data of millions of patients, clinical data warehouses (CDWs) represent a great opportunity to develop computational tools. Magnetic resonance images (MRIs) are particularly sensitive to patient movements during image acquisition, which will result in artefacts (blurring, ghosting and ringing) in the reconstructed image. As a result, a significant number of MRIs in CDWs are corrupted by these artefacts and may be unusable. Since their manual detection is impossible due to the large number of scans, it is necessary to develop tools to automatically exclude (or at least identify) images with motion in order to fully exploit CDWs. In this paper, we propose a novel transfer learning method from research to clinical data for the automatic detection of motion in 3D T1-weighted brain MRI. The method consists of two steps: a pre-training on research data using synthetic motion, followed by a fine-tuning step to generalise our pre-trained model to clinical data, relying on the labelling of 4045 images. The objectives were both (1) to be able to exclude images with severe motion, (2) to detect mild motion artefacts. Our approach achieved excellent accuracy for the first objective with a balanced accuracy nearly similar to that of the annotators (balanced accuracy>80 %). However, for the second objective, the performance was weaker and substantially lower than that of human raters. Overall, our framework will be useful to take advantage of CDWs in medical imaging and highlight the importance of a clinical validation of models trained on research data.

Evaluation of MRI-based machine learning approaches for computer-aided diagnosis of dementia in a clinical data warehouse.

A variety of algorithms have been proposed for computer-aided diagnosis of dementia from anatomical brain MRI. These approaches achieve high accuracy when applied to research data sets but their performance on real-life clinical routine data has not been evaluated yet. The aim of this work was to study the performance of such approaches on clinical routine data, based on a hospital data warehouse, and to compare the results to those obtained on a research data set. The clinical data set was extracted from the hospital data warehouse of the Greater Paris area, which includes 39 different hospitals. The research set was composed of data from the Alzheimer's Disease Neuroimaging Initiative data set. In the clinical set, the population of interest was identified by exploiting the diagnostic codes from the 10th revision of the International Classification of Diseases that are assigned to each patient. We studied how the imbalance of the training sets, in terms of contrast agent injection and image quality, may bias the results. We demonstrated that computer-aided diagnosis performance was strongly biased upwards (over 17 percent points of balanced accuracy) by the confounders of image quality and contrast agent injection, a phenomenon known as the Clever Hans effect or shortcut learning. When these biases were removed, the performance was very poor. In any case, the performance was considerably lower than on the research data set. Our study highlights that there are still considerable challenges for translating dementia computer-aided diagnosis systems to clinical routine.

Persistent perfusion abnormalities at day 1 correspond to different clinical trajectories after stroke.

BACKGROUND: Perfusion abnormalities after thrombolysis are frequent within and surrounding ischemic lesions, but their relative frequency is not well known. OBJECTIVE: To describe the different patterns of perfusion abnormalities observed at 24 hours and compare the characteristics of the patients according to their perfusion pattern. METHODS: From our thrombolysis registry, we included 226 consecutive patients with an available arterial spin labeling (ASL) perfusion sequence at day 1. We performed a blinded assessment of the perfusion status (hypoperfusion-h, hyperperfusion-H, or normal-N) in the ischemic lesion and in the surrounding tissue. We compared the time course of clinical recovery, the rate of arterial recanalization, and hemorrhagic transformations in the different perfusion profiles. RESULTS: We identified seven different perfusion profiles at day 1. Four of these (h/h, h/H, H/H, and H/N) represented the majority of the population (84.1%). The H/H profile was the most frequent (34.5%) and associated with 3-month good outcome (modified Rankin Scale (mRS): 63.5%). Patients with persistent hypoperfusion within and outside the lesion (h/h, 12.4%) exhibited worse outcomes after treatment (mRS score 0-2: 23.8%) than other patients, were less frequently recanalized (40.7%), and had more parenchymal hematoma (17.8%). The h/H profile had an intermediate clinical trajectory between the h/h profile and the hyperperfused profiles. CONCLUSION: ASL hypoperfusion within the infarct and the surrounding tissue was associated with poor outcome. A more comprehensive view of the mechanisms in the hypoperfused surrounding tissue could help to design new therapeutic approaches during and after reperfusion therapies.

Erdheim-Chester disease: look it in the eye. An orbital magnetic resonance imaging study.

Erdheim-Chester disease (ECD) is a rare L-group histiocytosis. Orbital involvement is found in a third of cases, but few data are available concerning the radiological features of ECD-related orbital disease (ECD-ROD). Our aim was to characterize the initial radiological phenotype and outcome of patients with ECD-ROD. Initial and follow-up orbital magnetic resonance imaging (MRI) from the patients with histologically proven ECD at a national reference center were reviewed. Pathological orbital findings were recorded for 45 (33%) of the 137 patients included, with bilateral involvement in 38/45 (84%) cases. The mean age (± standard deviation) of these patients was 60 (±11.3) years and 78% were men. Intraconal fat infiltration around the optic nerve sheath adjacent to the eye globe (52%), with intense gadolinium uptake and a fibrous component was the most frequent phenotype described. Optic nerve signal abnormalities were observed in 47% of cases. Two patients had bilateral homogeneous extraocular muscle enlargement suggestive of a myositis-like involvement of ECD-ROD. None had isolated dacryoadenitis but in 17 eyes dacryodenitis was described in association with other types of orbital lesions. Only seven patients (15%) had normal brain MRI findings. ECD-associated paranasal sinus involvement and post-pituitary involvement were detected in 56% and 53% of patients, respectively. A decrease/disappearance of the lesions was observed in 17/24 (71%) of the patients undergoing late (>12 months) followups. Interestingly, ECD-ROD only rarely (7/45; 16%) revealed the disease, with exophthalmos being the most frequently identified feature in this subgroup (3/45; 6%). Even though ECD-ROD can be clinically silent, it comprises a broad array of lesions often resulting in optic nerve signal abnormalities, the functional outcome of which remains to be established. ECD-ROD should thus be assessed initially and subsequently monitored by orbital MRI and ophthalmological follow-up.

ClinicaDL: An open-source deep learning software for reproducible neuroimaging processing.

BACKGROUND AND OBJECTIVE: As deep learning faces a reproducibility crisis and studies on deep learning applied to neuroimaging are contaminated by methodological flaws, there is an urgent need to provide a safe environment for deep learning users to help them avoid common pitfalls that will bias and discredit their results. Several tools have been proposed to help deep learning users design their framework for neuroimaging data sets. Software overview: We present here ClinicaDL, one of these software tools. ClinicaDL interacts with BIDS, a standard format in the neuroimaging field, and its derivatives, so it can be used with a large variety of data sets. Moreover, it checks the absence of data leakage when inferring the results of new data with trained networks, and saves all necessary information to guarantee the reproducibility of results. The combination of ClinicaDL and its companion project Clinica allows performing an end-to-end neuroimaging analysis, from the download of raw data sets to the interpretation of trained networks, including neuroimaging preprocessing, quality check, label definition, architecture search, and network training and evaluation. CONCLUSIONS: We implemented ClinicaDL to bring answers to three common issues encountered by deep learning users who are not always familiar with neuroimaging data: (1) the format and preprocessing of neuroimaging data sets, (2) the contamination of the evaluation procedure by data leakage and (3) a lack of reproducibility. We hope that its use by researchers will allow producing more reliable and thus valuable scientific studies in our field.

Benefit of mechanical thrombectomy in acute ischemic stroke related to calcified cerebral embolus.

PURPOSE: Mechanical thrombectomies (MT) in patients with large vessel occlusion (LVO) related to calcified cerebral embolus (CCE) have been reported, through small case series, being associated with low reperfusion rate and worse outcome, compared to regular MT. The purpose of the MASC (Mechanical Thrombectomy in Acute Ischemic Stroke Related to Calcified Cerebral Embolus) study was to evaluate the incidence of CCEs treated by MT and the effectiveness of MT in this indication. METHODS: The MASC study is a retrospective multicentric (n = 37) national study gathering the cases of adult patients who underwent MT for acute ischemic stroke with LVO related to a CCE in France from January 2015 to November 2019. Reperfusion rate (mTICI ≥ 2B), complication rate and 90-day mRS were systematically collected. We then conducted a systematic review by searching for articles in PubMed, Cochrane Library, Embase and Google Scholar from January 2015 to March 2020. A meta-analysis was performed to estimate clinical outcome at 90 days, reperfusion rate and complications. RESULTS: We gathered data from 35 patients. Reperfusion was obtained in 57% of the cases. Good clinical outcome was observed in 28% of the patients. The meta-analysis retrieved 136 patients. Reperfusion and good clinical outcome were obtained in 50% and 29% of the cases, respectively. CONCLUSION: The MASC study found worse angiographic and clinical outcomes compared to regular thrombectomies. Individual patient-based meta-analysis including the MASC findings shows a 50% reperfusion rate and a 29% of good clinical outcome.

Validation of an automatic tool for the rapid measurement of brain atrophy and white matter hyperintensity: QyScore®.

OBJECTIVES: QyScore® is an imaging analysis tool certified in Europe (CE marked) and the US (FDA cleared) for the automatic volumetry of grey and white matter (GM and WM respectively), hippocampus (HP), amygdala (AM), and white matter hyperintensity (WMH). Here we compare QyScore® performances with the consensus of expert neuroradiologists. METHODS: Dice similarity coefficient (DSC) and the relative volume difference (RVD) for GM, WM volumes were calculated on 50 3DT1 images. DSC and the F1 metrics were calculated for WMH on 130 3DT1 and FLAIR images. For each index, we identified thresholds of reliability based on current literature review results. We hypothesized that DSC/F1 scores obtained using QyScore® markers would be higher than the threshold. In contrast, RVD scores would be lower. Regression analysis and Bland-Altman plots were obtained to evaluate QyScore® performance in comparison to the consensus of three expert neuroradiologists. RESULTS: The lower bound of the DSC/F1 confidence intervals was higher than the threshold for the GM, WM, HP, AM, and WMH, and the higher bounds of the RVD confidence interval were below the threshold for the WM, GM, HP, and AM. QyScore®, compared with the consensus of three expert neuroradiologists, provides reliable performance for the automatic segmentation of the GM and WM volumes, and HP and AM volumes, as well as WMH volumes. CONCLUSIONS: QyScore® represents a reliable medical device in comparison with the consensus of expert neuroradiologists. Therefore, QyScore® could be implemented in clinical trials and clinical routine to support the diagnosis and longitudinal monitoring of neurological diseases. KEY POINTS: • QyScore® provides reliable automatic segmentation of brain structures in comparison with the consensus of three expert neuroradiologists. • QyScore® automatic segmentation could be performed on MRI images using different vendors and protocols of acquisition. In addition, the fast segmentation process saves time over manual and semi-automatic methods. • QyScore® could be implemented in clinical trials and clinical routine to support the diagnosis and longitudinal monitoring of neurological diseases.

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

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.

Pathomechanisms behind cognitive disorders following ruptured anterior communicating aneurysms: A diffusion tensor imaging study.

INTRODUCTION: After the rupture of anterior communicating aneurysms, most patients experience debilitating cognitive disorders; and sometimes even without showing morphological anomaly on MRI examinations. Diffusion Tensor Imaging (DTI) may help understanding the pathomechanisms leading to such disorders in this subset of patients. METHODS: After independent assessment, we constituted a population of patients with normal morphological imaging (ACOM group). Then, a case-control study comparing volumetric and voxel-based DTI parameters between the ACOM group and a control population was performed. All patients underwent the full imaging and neuropsychological assessments at 6 months after the aneurysm rupture. Results were considered significant when p<2.02.10-4. RESULTS: Twelve patients were included in the ACOM group: 75% had at least one disabled cognitive domain. Significant differences in DTI parameters of global white matter were noted (average Fractional Anisotropy: 0.915 [±0.05] in ACOM group versus 0.943 (±0.03); p = 1.10-5) and in frontal white matter tracts (superior fronto-occipital fasciculus and anterior parts of the corona radiata) as well as in the fornix. CONCLUSION: Cognitive disorders are under-estimated, and DTI confirmed that, even when conventional MRI examinations were normal, there were still signs of diffuse neuronal injuries that seemed to dominate in frontal areas, close to the site of rupture.

Non-ischemic cerebral enhancing lesions after intracranial aneurysm endovascular repair: a retrospective French national registry.

BACKGROUND: Non-ischemic cerebral enhancing (NICE) lesions are exceptionally rare following aneurysm endovascular therapy (EVT). OBJECTIVE: To investigate the presenting features and longitudinal follow-up of patients with NICE lesions following aneurysm EVT. METHODS: Patients included in a retrospective national multicentre inception cohort were analysed. NICE lesions were defined, using MRI, as delayed onset punctate, nodular or annular foci enhancements with peri-lesion edema, distributed in the vascular territory of the aneurysm EVT, with no other confounding disease. RESULTS: From a pool of 58 815 aneurysm endovascular treatment procedures during the study sampling period (2006-2019), 21/37 centres identified 31 patients with 32 aneurysms of the anterior circulation who developed NICE lesions (mean age 45±10 years). Mean delay to diagnosis was 5±9 months, with onset occurring a month or less after the index EVT procedure in 10 out of 31 patients (32%). NICE lesions were symptomatic at time of onset in 23 of 31 patients (74%). After a mean follow-up of 25±26 months, 25 patients (81%) were asymptomatic or minimally symptomatic without disability (modified Rankin Scale (mRS) score 0-1) at last follow-up while 4 (13%) presented with mild disability (mRS score 2). Clinical follow-up data were unavailable for two patients. Follow-up MRI (available in 27 patients; mean time interval after onset of 22±22 months) demonstrated persistent enhancement in 71% of cases. CONCLUSIONS: The clinical spectrum of NICE lesions following aneurysm EVT therapy spans a wide range of neurological symptoms. Clinical course is most commonly benign, although persistent long-term enhancement is frequent.

Automatic quality control of brain T1-weighted magnetic resonance images for a clinical data warehouse.

Many studies on machine learning (ML) for computer-aided diagnosis have so far been mostly restricted to high-quality research data. Clinical data warehouses, gathering routine examinations from hospitals, offer great promises for training and validation of ML models in a realistic setting. However, the use of such clinical data warehouses requires quality control (QC) tools. Visual QC by experts is time-consuming and does not scale to large datasets. In this paper, we propose a convolutional neural network (CNN) for the automatic QC of 3D T1-weighted brain MRI for a large heterogeneous clinical data warehouse. To that purpose, we used the data warehouse of the hospitals of the Greater Paris area (Assistance Publique-Hôpitaux de Paris [AP-HP]). Specifically, the objectives were: 1) to identify images which are not proper T1-weighted brain MRIs; 2) to identify acquisitions for which gadolinium was injected; 3) to rate the overall image quality. We used 5000 images for training and validation and a separate set of 500 images for testing. In order to train/validate the CNN, the data were annotated by two trained raters according to a visual QC protocol that we specifically designed for application in the setting of a data warehouse. For objectives 1 and 2, our approach achieved excellent accuracy (balanced accuracy and F1-score >90%), similar to the human raters. For objective 3, the performance was good but substantially lower than that of human raters. Nevertheless, the automatic approach accurately identified (balanced accuracy and F1-score >80%) low quality images, which would typically need to be excluded. Overall, our approach shall be useful for exploiting hospital data warehouses in medical image computing.

Changes in the use of psychotropic drugs during the course of Alzheimer's disease: A large-scale longitudinal study of French medical records.

INTRODUCTION: We aim to understand how patients with Alzheimer's disease (AD) are treated by identifying in a longitudinal fashion the late-life changes in patients' medical history that precede and follow AD diagnosis. METHODS: We use prescription history of 34,782 patients followed between 1996 and 2019 by French general practitioners. We compare patients with an AD diagnosis, patients with mild cognitive impairment (MCI), and patients free of mental disorders. We use a generalized mixed-effects model to study the longitudinal changes in the prescription of eight drug types for a period 15 years before diagnosis and 10 years after. RESULTS: In the decades preceding diagnosis, we find that future AD patients are treated significantly more than MCI patients with most psychotropic drugs and that most studied drugs are increasingly prescribed with age. At the time of diagnosis, all psychotropic drugs except benzodiazepines show a significant increase in prescription, while other drugs are significantly less prescribed. In the 10 years after diagnosis, nearly all categories of drugs are less and less prescribed including antidementia drugs. DISCUSSION: Pre-diagnosis differences between future AD patients and MCI patients may indicate that subtle cognitive changes are recognized and treated as psychiatric symptoms. The disclosure of AD diagnosis drastically changes patients' care, priority being given to the management of psychiatric symptoms. The decrease of all prescriptions in the late stages may reflect treatment discontinuation and simplification of therapeutic procedures. This study therefore provides new insights into the medical practices for management of AD.

Simultaneously acquired PET and ASL imaging biomarkers may be helpful in differentiating progression from pseudo-progression in treated gliomas.

OBJECTIVES: The aim of this work was investigating the methods based on coupling cerebral perfusion (ASL) and amino acid metabolism ([18F]DOPA-PET) measurements to evaluate the diagnostic performance of PET/MRI in glioma follow-up. METHODS: Images were acquired using a 3-T PET/MR system, on a prospective cohort of patients addressed for possible glioma progression. Data were preprocessed with statistical parametric mapping (SPM), including registration on T1-weighted images, spatial and intensity normalization, and tumor segmentation. As index tests, tumor isocontour maps of [18F]DOPA-PET and ASL T-maps were created and metabolic/perfusion abnormalities were evaluated with the asymmetry index z-score. SPM map analysis of significant size clusters and semi-quantitative PET and ASL map evaluation were performed and compared to the gold standard diagnosis. Lastly, ASL and PET topography of significant clusters was compared to that of the initial tumor. RESULTS: Fifty-eight patients with unilateral treated glioma were included (34 progressions and 24 pseudo-progressions). The tumor isocontour maps and T-maps showed the highest specificity (100%) and sensitivity (94.1%) for ASL and [18F]DOPA analysis, respectively. The sensitivity of qualitative SPM maps and semi-quantitative rCBF and rSUV analyses were the highest for glioblastoma. CONCLUSION: Tumor isocontour T-maps and combined analysis of CBF and [18F]DOPA-PET uptake allow achieving high diagnostic performance in differentiating between progression and pseudo-progression in treated gliomas. The sensitivity is particularly high for glioblastomas. KEY POINTS: • Applied separately, MRI and PET imaging modalities may be insufficient to characterize the brain glioma post-therapeutic profile. • Combined ASL and [18F]DOPA-PET map analysis allows differentiating between tumor progression and pseudo-progression.

Reliability and accuracy of time-resolved contrast-enhanced magnetic resonance angiography in hypervascular spinal metastases prior embolization.

OBJECTIVES: Preoperative embolization of hypervascular spinal metastases (HSM) is efficient to reduce perioperative bleeding. However, intra-arterial digital subtraction angiography (IA-DSA) must confirm the hypervascular nature and rule out spinal cord arterial feeders. This study aimed to evaluate the reliability and accuracy of time-resolved contrast-enhanced magnetic resonance angiography (TR-CE-MRA) in assessing HSM prior to embolization. METHODS: All consecutive patients referred for preoperative embolization of an HSM were prospectively included. TR-CE-MRA sequences and selective IA-DSA were performed prior to embolization. Two readers independently reviewed imaging data to grade tumor vascularity (using a 3-grade and a dichotomized "yes vs no" scale) and identify the arterial supply of the spinal cord. Interobserver and intermodality agreements were estimated using kappa statistics. RESULTS: Thirty patients included between 2016 and 2019 were assessed for 55 levels. Interobserver agreement was moderate (κ = 0.52; 95% CI [0.09-0.81]) for TR-CE-MRA. Intermodality agreement between TR-CE-MRA and IA-DSA was good (κ = 0.74; 95% CI [0.37-1.00]). TR-CE-MRA had a sensitivity of 97.9%, a specificity of 71.4%, a positive predictive value of 95.9%, a negative predictive value of 83.3%, and an overall accuracy of 94.6%, for differentiating hypervascular from non-hypervascular SM. The arterial supply of the spine was assessable in 2/30 (6.7%) cases with no interobserver agreement (κ < 0). CONCLUSIONS: TR-CE-MRA can reliably differentiate hypervascular from non-hypervascular SM and thereby avoid futile IA-DSAs. However, TR-CE-MRA was not able to evaluate the vascular supply of the spinal cord at the target levels, thus limiting its scope as a pretherapeutic assessment tool. KEY POINTS: • TR-CE-MRA aids in distinguishing hypervascular from non-hypervascular spinal metastases. • TR-CE-MRA could avoid one-quarter of patients referred for HSM embolization to undergo futile conventional angiography. • TR-CE-MRA's spatial resolution is insufficient to replace IA-DSA in the pretherapeutic assessment of the spinal cord vascular anatomy.

Predicting the progression of mild cognitive impairment using machine learning: A systematic, quantitative and critical review.

We performed a systematic review of studies focusing on the automatic prediction of the progression of mild cognitive impairment to Alzheimer's disease (AD) dementia, and a quantitative analysis of the methodological choices impacting performance. This review included 172 articles, from which 234 experiments were extracted. For each of them, we reported the used data set, the feature types, the algorithm type, performance and potential methodological issues. The impact of these characteristics on the performance was evaluated using a multivariate mixed effect linear regressions. We found that using cognitive, fluorodeoxyglucose-positron emission tomography or potentially electroencephalography and magnetoencephalography variables significantly improved predictive performance compared to not including them, whereas including other modalities, in particular T1 magnetic resonance imaging, did not show a significant effect. The good performance of cognitive assessments questions the wide use of imaging for predicting the progression to AD and advocates for exploring further fine domain-specific cognitive assessments. We also identified several methodological issues, including the absence of a test set, or its use for feature selection or parameter tuning in nearly a fourth of the papers. Other issues, found in 15% of the studies, cast doubts on the relevance of the method to clinical practice. We also highlight that short-term predictions are likely not to be better than predicting that subjects stay stable over time. These issues highlight the importance of adhering to good practices for the use of machine learning as a decision support system for the clinical practice.

Radiological classification of dementia from anatomical MRI assisted by machine learning-derived maps.

BACKGROUND AND PURPOSE: Many artificial intelligence tools are currently being developed to assist diagnosis of dementia from magnetic resonance imaging (MRI). However, these tools have so far been difficult to integrate in the clinical routine workflow. In this work, we propose a new simple way to use them and assess their utility for improving diagnostic accuracy. MATERIALS AND METHODS: We studied 34 patients with early-onset Alzheimer's disease (EOAD), 49 with late-onset AD (LOAD), 39 with frontotemporal dementia (FTD) and 24 with depression from the pre-existing cohort CLIN-AD. Support vector machine (SVM) automatic classifiers using 3D T1 MRI were trained to distinguish: LOAD vs. Depression, FTD vs. LOAD, EOAD vs. Depression, EOAD vs. FTD. We extracted SVM weight maps, which are tridimensional representations of discriminant atrophy patterns used by the classifier to take its decisions and we printed posters of these maps. Four radiologists (2 senior neuroradiologists and 2 unspecialized junior radiologists) performed a visual classification of the 4 diagnostic pairs using 3D T1 MRI. Classifications were performed twice: first with standard radiological reading and then using SVM weight maps as a guide. RESULTS: Diagnostic performance was significantly improved by the use of the weight maps for the two junior radiologists in the case of FTD vs. EOAD. Improvement was over 10 points of diagnostic accuracy. CONCLUSION: This tool can improve the diagnostic accuracy of junior radiologists and could be integrated in the clinical routine workflow.

Comparison and validation of seven white matter hyperintensities segmentation software in elderly patients.

BACKGROUND: Manual segmentation is currently the gold standard to assess white matter hyperintensities (WMH), but it is time consuming and subject to intra and inter-operator variability. PURPOSE: To compare automatic methods to segment white matter hyperintensities (WMH) in the elderly in order to assist radiologist and researchers in selecting the most relevant method for application on clinical or research data. MATERIAL AND METHODS: We studied a research dataset composed of 147 patients, including 97 patients from the Alzheimer's Disease Neuroimaging Initiative (ADNI) 2 database and 50 patients from ADNI 3 and a clinical routine dataset comprising 60 patients referred for cognitive impairment at the Pitié-Salpêtrière hospital (imaged using four different MRI machines). We used manual segmentation as the gold standard reference. Both manual and automatic segmentations were performed using FLAIR MRI. We compared seven freely available methods that produce segmentation mask and are usable by a radiologist without a strong knowledge of computer programming: LGA (Schmidt et al., 2012), LPA (Schmidt, 2017), BIANCA (Griffanti et al., 2016), UBO detector (Jiang et al., 2018), W2MHS (Ithapu et al., 2014), nicMSlesion (with and without retraining) (Valverde et al., 2019, 2017). The primary outcome for assessing segmentation accuracy was the Dice similarity coefficient (DSC) between the manual and the automatic segmentation software. Secondary outcomes included five other metrics. RESULTS: A deep learning approach, NicMSlesion, retrained on data from the research dataset ADNI, performed best on this research dataset (DSC: 0.595) and its DSC was significantly higher than that of all others. However, it ranked fifth on the clinical routine dataset and its performance severely dropped on data with artifacts. On the clinical routine dataset, the three top-ranked methods were LPA, SLS and BIANCA. Their performance did not differ significantly but was significantly higher than that of other methods. CONCLUSION: This work provides an objective comparison of methods for WMH segmentation. Results can be used by radiologists to select a tool.

Retrospective Observational Study of Brain MRI Findings in Patients with Acute SARS-CoV-2 Infection and Neurologic Manifestations.

Background This study provides a detailed imaging assessment in a large series of patients infected with coronavirus disease 2019 (COVID-19) and presenting with neurologic manifestations. Purpose To review the MRI findings associated with acute neurologic manifestations in patients with COVID-19. Materials and Methods This was a cross-sectional study conducted between March 23 and May 7, 2020, at the Pitié-Salpêtrière Hospital, a reference center for COVID-19 in the Paris area. Adult patients were included if they had a diagnosis of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection with acute neurologic manifestations and referral for brain MRI. Patients with a prior history of neurologic disease were excluded. The characteristics and frequency of different MRI features were investigated. The findings were analyzed separately in patients in intensive care units (ICUs) and other departments (non-ICU). Results During the inclusion period, 1176 patients suspected of having COVID-19 were hospitalized. Of 308 patients with acute neurologic symptoms, 73 met the inclusion criteria and were included (23.7%): thirty-five patients were in the ICU (47.9%) and 38 were not (52.1%). The mean age was 58.5 years ± 15.6 [standard deviation], with a male predominance (65.8% vs 34.2%). Forty-three patients had abnormal MRI findings 2-4 weeks after symptom onset (58.9%), including 17 with acute ischemic infarct (23.3%), one with a deep venous thrombosis (1.4%), eight with multiple microhemorrhages (11.3%), 22 with perfusion abnormalities (47.7%), and three with restricted diffusion foci within the corpus callosum consistent with cytotoxic lesions of the corpus callosum (4.1%). Multifocal white matter-enhancing lesions were seen in four patients in the ICU (5%). Basal ganglia abnormalities were seen in four other patients (5%). Cerebrospinal fluid analyses were negative for SARS-CoV-2 in all patients tested (n = 39). Conclusion In addition to cerebrovascular lesions, perfusion abnormalities, cytotoxic lesions of the corpus callosum, and intensive care unit-related complications, we identified two patterns including white matter-enhancing lesions and basal ganglia abnormalities that could be related to severe acute respiratory syndrome coronavirus 2 infection. © RSNA, 2020 Online supplemental material is available for this article.