IMPA-Net: Interpretable Multi-Part Attention Network for Trustworthy Brain Tumor Classification from MRI.

Deep learning (DL) networks have shown attractive performance in medical image processing tasks such as brain tumor classification. However, they are often criticized as mysterious "black boxes". The opaqueness of the model and the reasoning process make it difficult for health workers to decide whether to trust the prediction outcomes. In this study, we develop an interpretable multi-part attention network (IMPA-Net) for brain tumor classification to enhance the interpretability and trustworthiness of classification outcomes. The proposed model not only predicts the tumor grade but also provides a global explanation for the model interpretability and a local explanation as justification for the proffered prediction. Global explanation is represented as a group of feature patterns that the model learns to distinguish high-grade glioma (HGG) and low-grade glioma (LGG) classes. Local explanation interprets the reasoning process of an individual prediction by calculating the similarity between the prototypical parts of the image and a group of pre-learned task-related features. Experiments conducted on the BraTS2017 dataset demonstrate that IMPA-Net is a verifiable model for the classification task. A percentage of 86% of feature patterns were assessed by two radiologists to be valid for representing task-relevant medical features. The model shows a classification accuracy of 92.12%, of which 81.17% were evaluated as trustworthy based on local explanations. Our interpretable model is a trustworthy model that can be used for decision aids for glioma classification. Compared with black-box CNNs, it allows health workers and patients to understand the reasoning process and trust the prediction outcomes.

Generalizing the Enhanced-Deep-Super-Resolution Neural Network to Brain MR Images: A Retrospective Study on the Cam-CAN Dataset.

The Enhanced-Deep-Super-Resolution (EDSR) model is a state-of-the-art convolutional neural network suitable for improving image spatial resolution. It was previously trained with general-purpose pictures and then, in this work, tested on biomedical magnetic resonance (MR) images, comparing the network outcomes with traditional up-sampling techniques. We explored possible changes in the model response when different MR sequences were analyzed. T1w and T2w MR brain images of 70 human healthy subjects (F:M, 40:30) from the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) repository were down-sampled and then up-sampled using EDSR model and BiCubic (BC) interpolation. Several reference metrics were used to quantitatively assess the performance of up-sampling operations (RMSE, pSNR, SSIM, and HFEN). Two-dimensional and three-dimensional reconstructions were evaluated. Different brain tissues were analyzed individually. The EDSR model was superior to BC interpolation on the selected metrics, both for two- and three- dimensional reconstructions. The reference metrics showed higher quality of EDSR over BC reconstructions for all the analyzed images, with a significant difference of all the criteria in T1w images and of the perception-based SSIM and HFEN in T2w images. The analysis per tissue highlights differences in EDSR performance related to the gray-level values, showing a relative lack of outperformance in reconstructing hyperintense areas. The EDSR model, trained on general-purpose images, better reconstructs MR T1w and T2w images than BC, without any retraining or fine-tuning. These results highlight the excellent generalization ability of the network and lead to possible applications on other MR measurements.

Sulcal Morphometry Predicts Mild Cognitive Impairment Conversion to Alzheimer's Disease.

Background: Being able to differentiate mild cognitive impairment (MCI) patients who would eventually convert (MCIc) to Alzheimer's disease (AD) from those who would not (MCInc) is a key challenge for prognosis. Objective: This study aimed to investigate the ability of sulcal morphometry to predict MCI progression to AD, dedicating special attention to an accurate identification of sulci. Methods: Twenty-five AD patients, thirty-seven MCI and twenty-five healthy controls (HC) underwent a brain-MR protocol (1.5T scanner) including a high-resolution T1-weighted sequence. MCI patients underwent a neuropsychological assessment at baseline and were clinically re-evaluated after a mean of 2.3 years. At follow-up, 12 MCI were classified as MCInc and 25 as MCIc. Sulcal morphometry was investigated using the BrainVISA framework. Consistency of sulci across subjects was ensured by visual inspection and manual correction of the automatic labelling in each subject. Sulcal surface, depth, length, and width were retrieved from 106 sulci. Features were compared across groups and their classification accuracy in predicting MCI conversion was tested. Potential relationships between sulcal features and cognitive scores were explored using Spearman's correlation. Results: The width of sulci in the temporo-occipital region strongly differentiated between each pair of groups. Comparing MCIc and MCInc, the width of several sulci in the bilateral temporo-occipital and left frontal areas was significantly altered. Higher width of frontal sulci was associated with worse performances in short-term verbal memory and phonemic fluency. Conclusions: Sulcal morphometry emerged as a strong tool for differentiating HC, MCI, and AD, demonstrating its potential prognostic value for the MCI population.

Optimizing Screening for Intrastriatal Interventions in Huntington's Disease Using Predictive Models.

BACKGROUND: Intrastriatal delivery of potential therapeutics in Huntington's disease (HD) requires sufficient caudate and putamen volumes. Currently, volumetric magnetic resonance imaging is rarely done in clinical practice, and these data are not available in large research cohorts such as Enroll-HD. OBJECTIVE: The objective of this study was to investigate whether predictive models can accurately classify HD patients who exceed caudate and putamen volume thresholds required for intrastriatal therapeutic interventions. METHODS: We obtained and merged data for 1374 individuals across three HD cohorts: IMAGE-HD, PREDICT-HD, and TRACK-HD/TRACK-ON. We imputed missing data for clinical variables with >72% non-missing values and used the model-building algorithm BORUTA to identify the 10 most important variables. A random forest algorithm was applied to build a predictive model for putamen volume >2500 mm3 and caudate volume >2000 mm3 bilaterally. Using the same 10 predictors, we constructed a logistic regression model with predictors significant at P < 0.05. RESULTS: The random forest model with 1000 trees and minimal terminal node size of 5 resulted in 83% area under the curve (AUC). The logistic regression model retaining age, CAG repeat size, and symbol digit modalities test-correct had 85.1% AUC. A probability cutoff of 0.8 resulted in 5.4% false positive and 66.7% false negative rates. CONCLUSIONS: Using easily obtainable clinical data and machine learning-identified initial predictor variables, random forest, and logistic regression models can successfully identify people with sufficient striatal volumes for inclusion cutoffs. Adopting these models in prescreening could accelerate clinical trial enrollment in HD and other neurodegenerative disorders when volume cutoffs are necessary enrollment criteria. © 2024 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Fracture strength of 3-units fixed partial dentures fabricated with metal-ceramic, graphene doped PMMA and PMMA before and after ageing: An in-vitro study.

OBJECTIVES: To evaluate the fracture strength and linear elongation at break of three-units fixed partial dentures (FPDs) fabricated with traditional and new materials for fixed prosthodontics before and after ageing. METHODS: Sixty models of three-units FPDs were fabricated and cemented onto a Co-Cr model simulating the replacement of a maxillary second premolar. The samples were randomly divided into 3 groups: metal-ceramic (MCR), graphene-doped polymethylmethacrylate (PMMA-GR) and polymethylmethacrylate (PMMA). Half of the samples were directly subjected to fracture test, while the remaining half underwent an ageing process and then a fracture loading test using an electrodynamic testing machine. Fracture load and elongation at break values were taken and statistically analysed. RESULTS: Significant differences were detected between the different materials (p<0.05). All groups showed a reduction of the fracture load and elongation at break values after ageing, but not statistically significant, except for PMMA group (p = 2.012e-19) (p = 3.8e-11). CONCLUSIONS: MCR and PMMA-GR three-units FPDs showed higher fracture strength and lower elongation at break compared to PMMA. MCR and PMMA-GR had higher resistance to ageing processes compared to PMMA. CLINICAL SIGNIFICANCE: PMMA-GR could be considered a material for long-term provisional restorations as its mechanical behaviour and ageing resistance are more like MCR than PMMA.

Assessing robustness of quantitative susceptibility-based MRI radiomic features in patients with multiple sclerosis.

Multiple Sclerosis (MS) is an autoimmune demyelinating disease characterised by changes in iron and myelin content. These biomarkers are detectable by Quantitative Susceptibility Mapping (QSM), an advanced Magnetic Resonance Imaging technique detecting magnetic properties. When analysed with radiomic techniques that exploit its intrinsic quantitative nature, QSM may furnish biomarkers to facilitate early diagnosis of MS and timely assessment of progression. In this work, we explore the robustness of QSM radiomic features by varying the number of grey levels (GLs) and echo times (TEs), in a sample of healthy controls and patients with MS. We analysed the white matter in total and within six clinically relevant tracts, including the cortico-spinal tract and the optic radiation. After optimising the number of GLs (n = 64), at least 65% of features were robust for each Volume of Interest (VOI), with no difference (p > .05) between left and right hemispheres. Different outcomes in feature robustness among the VOIs depend on their characteristics, such as volume and variance of susceptibility values. This study validated the processing pipeline for robustness analysis and established the reliability of QSM-based radiomics features against GLs and TEs. Our results provide important insights for future radiomics studies using QSM in clinical applications.

Methods of diffusion MRI tractography for localization of the anterior optic pathway: A systematic review of validated methods.

The anterior optic pathway (AOP) is a system of three structures (optic nerves, optic chiasma, and optic tracts) that convey visual stimuli from the retina to the lateral geniculate nuclei. A successful reconstruction of the AOP using tractography could be helpful in several clinical scenarios, from presurgical planning and neuronavigation of sellar and parasellar surgery to monitoring the stage of fiber degeneration both in acute (e.g., traumatic optic neuropathy) or chronic conditions that affect AOP structures (e.g., amblyopia, glaucoma, demyelinating disorders or genetic optic nerve atrophies). However, its peculiar anatomy and course, as well as its surroundings, pose a serious challenge to obtaining successful tractographic reconstructions. Several AOP tractography strategies have been adopted but no standard procedure has been agreed upon. We performed a systematic review of the literature according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) 2020 guidelines in order to find the combinations of acquisition and reconstruction parameters that have been performed previously and have provided the highest rate of successful reconstruction of the AOP, in order to promote their routine implementation in clinical practice. For this purpose, we reviewed data regarding how the process of anatomical validation of the tractographies was performed. The Cochrane Handbook for Systematic Reviews of Interventions was used to assess the risk of bias and thus the study quality We identified thirty-nine studies that met our inclusion criteria, and only five were considered at low risk of bias and achieved over 80% of successful reconstructions. We found a high degree of heterogeneity in the acquisition and analysis parameters used to perform AOP tractography and different combinations of them can achieve satisfactory levels of anterior optic tractographic reconstruction both in real-life research and clinical scenarios. One thousand s/mm2 was the most frequently used b value, while both deterministic and probabilistic tractography algorithms performed morphological reconstruction of the tract satisfactorily, although probabilistic algorithms estimated a more realistic percentage of crossing fibers (45.6%) in healthy subjects. A wide heterogeneity was also found regarding the method used to assess the anatomical fidelity of the AOP reconstructions. Three main strategies can be found: direct visual direct visual assessment of the tractography superimposed to a conventional MR image, surgical evaluation, and computational methods. Because the latter is less dependent on a priori knowledge of the anatomy by the operator, computational methods of validation of the anatomy should be considered whenever possible.

Cognitive and functional connectivity impairment in post-COVID-19 olfactory dysfunction.

OBJECTIVES: To explore the neuropsychological profile and the integrity of the olfactory network in patients with COVID-19-related persistent olfactory dysfunction (OD). METHODS: Patients with persistent COVID-19-related OD underwent olfactory assessment with Sniffin' Sticks and neuropsychological evaluation. Additionally, both patients and a control group underwent brain MRI, including T1-weighted and resting-state functional MRI (rs-fMRI) sequences on a 3 T scanner. Morphometrical properties were evaluated in olfaction-associated regions; the rs-fMRI data were analysed using graph theory at the whole-brain level and within a standard parcellation of the olfactory functional network. All the MR-derived quantities were compared between the two groups and their correlation with clinical scores in patients were explored. RESULTS: We included 23 patients (mean age 37 ± 14 years, 12 females) with persistent (mean duration 11 ± 5 months, range 2-19 months) COVID-19-related OD (mean score 23.63 ± 5.32/48, hyposmia cut-off: 30.75) and 26 sex- and age-matched healthy controls. Applying population-derived cut-off values, the two cognitive domains mainly impaired were visuospatial memory and executive functions (17 % and 13 % of patients). Brain MRI did not show gross morphological abnormalities. The lateral orbital cortex, hippocampus, and amygdala volumes exhibited a reduction trend in patients, not significant after the correction for multiple comparisons. The olfactory bulb volumes did not differ between patients and controls. Graph analysis of the functional olfactory network showed altered global and local properties in the patients' group (n = 19, 4 excluded due to artifacts) compared to controls. Specifically, we detected a reduction in the global modularity coefficient, positively correlated with hyposmia severity, and an increase of the degree and strength of the right thalamus functional connections, negatively correlated with short-term verbal memory scores. DISCUSSION: Patients with persistent COVID-19-related OD showed an altered olfactory network connectivity correlated with hyposmia severity and neuropsychological performance. No significant morphological alterations were found in patients compared with controls.

Multimodal investigation of melanopsin retinal ganglion cells in Alzheimer's disease.

OBJECTIVE: In Alzheimer's disease (AD), the presence of circadian dysfunction is well-known and may occur early in the disease course. The melanopsin retinal ganglion cell (mRGC) system may play a relevant role in contributing to circadian dysfunction. In this study, we aimed at evaluating, through a multimodal approach, the mRGC system in AD at an early stage of disease. METHODS: We included 29 mild-moderate AD (70.9 ± 11 years) and 26 (70.5 ± 8 years) control subjects. We performed an extensive neurophtalmological evaluation including optical coherence tomography with ganglion cell layer segmentation, actigraphic evaluation of the rest-activity rhythm, chromatic pupillometry analyzed with a new data-fitting approach, and brain functional MRI combined with light stimuli assessing the mRGC system. RESULTS: We demonstrated a significant thinning of the infero-temporal sector of the ganglion cell layer in AD compared to controls. Moreover, we documented by actigraphy the presence of a circadian-impaired AD subgroup. Overall, circadian measurements worsened by age. Chromatic pupillometry evaluation highlighted the presence of a pupil-light response reduction in the rod condition pointing to mRGC dendropathy. Finally, brain fMRI showed a reduced occipital cortex activation with blue light particularly for the sustained responses. INTERPRETATION: Overall, the results of this multimodal innovative approach clearly document a dysfunctional mRGC system at early stages of disease as a relevant contributing factor for circadian impairment in AD providing also support to the use of light therapy in AD.

The Safety of Deutetrabenazine for Chorea in Huntington Disease: An Open-Label Extension Study.

BACKGROUND: Deutetrabenazine is approved in the USA, China, Australia, Israel, Brazil, and South Korea for the treatment of chorea associated with Huntington disease. OBJECTIVE: We aimed to evaluate the long-term safety and tolerability of deutetrabenazine for the treatment of Huntington disease. METHODS: This open-label, single-arm, multi-center study included patients who completed a double-blind study (Rollover) and patients who converted overnight from a stable tetrabenazine dose (Switch). Exposure-adjusted incidence rates (adverse events per person-year) were calculated. Efficacy was analyzed using a stable post-titration timepoint (8 weeks). Changes in the Unified Huntington's Disease Rating Scale total motor score and total maximal chorea score from baseline to week 8, as well as those from week 8 to week 145 (or the last visit on the study drug if that occurred earlier), were evaluated as both efficacy and safety endpoints during the study. RESULTS: Of 119 patients (Rollover, n = 82; Switch, n = 37), 100 (84%) completed ≥ 1 year of treatment. End-of-study exposure-adjusted incidence rates for adverse events in Rollover and Switch, respectively, were: any, 2.57 and 4.02; serious, 0.11 and 0.14; leading to dose suspension, 0.05 and 0.04. Common adverse events (≥ 4% either cohort) included somnolence (Rollover, 20%; Switch, 30%), depression (32%; 22%), anxiety (27%; 35%), insomnia (23%; 16%), and akathisia (6%; 11%). Adverse events of interest included suicidality (9%; 5%) and parkinsonism (4%; 8%). Mean dose at week 8 was 38.1 mg (Rollover) and 36.5 mg (Switch). Mean dose across cohorts after titration was 37.6 mg; at the final visit, mean dose across cohorts was 45.7 mg. Patients showed minimal change in the Unified Huntington's Disease Rating Scale total maximal chorea scores with stable dosing from weeks 8-145 or at the end of treatment, but total motor score increased versus week 8 (mean change [standard deviation]: 8.2 [11.9]). There were no unexpected adverse events upon drug withdrawal, and mean (standard deviation) total maximal chorea scores increased 4.7 (4.6) units from week 8 to 1-week follow-up. CONCLUSIONS: Adverse events observed with long-term deutetrabenazine exposure were consistent with previous studies. Reductions in chorea persisted over time. Upon treatment cessation, there was no unexpected worsening of chorea. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT01897896.

The contribution of the 1H-MRS lipid signal to cervical cancer prognosis: a preliminary study.

BACKGROUND: The aim of this study was to investigate the role of the lipid peak derived from 1H magnetic resonance (MR) spectroscopy in assessing cervical cancer prognosis, particularly in assessing response to neoadjuvant chemotherapy (NACT) of locally advanced cervical cancer (LACC). METHODS: We enrolled 17 patients with histologically proven cervical cancer who underwent 3-T MR imaging at baseline. In addition to conventional imaging sequences for pelvic assessment, the protocol included a single-voxel point-resolved spectroscopy (PRESS) sequence, with repetition time of 1,500 ms and echo times of 28 and 144 ms. Spectra were analysed using the LCModel fitting routine, thus extracting multiple metabolites, including lipids (Lip) and total choline (tCho). Patients with LACC were treated with NACT and reassessed by MRI at term. Based on tumour volume reduction, patients were classified as good responder (GR; tumour volume reduction > 50%) and poor responder or nonresponder (PR-or-NR; tumour volume reduction ≤ 50%). RESULTS: Of 17 patients, 11 were LACC. Of these 11, only 6 had both completed NACT and had good-quality 1H-MR spectra; 3 GR and 3 PR-or-NR. A significant difference in lipid values was observed in the two groups of patients, particularly with higher Lip values and higher Lip/tCho ratio in PR-NR patients (p =0.040). A significant difference was also observed in choline distribution (tCho), with higher values in GR patients (p = 0.040). CONCLUSIONS: Assessment of lipid peak at 1H-MR spectroscopy could be an additional quantitative parameter in predicting the response to NACT in patients with LACC.

In Vivo Parieto-Occipital White Matter Metabolism Is Correlated with Visuospatial Deficits in Adult DM1 Patients.

Myotonic dystrophy type 1 (DM1) is a genetic disorder caused by a (CTG) expansion in the DM protein kinase (DMPK) gene, representing the most common adult muscular dystrophy, characterized by a multisystem involvement with predominantly skeletal muscle and brain affection. Neuroimaging studies showed widespread white matter changes and brain atrophy in DM1, but only a few studies investigated the role of white matter metabolism in the pathophysiology of central nervous system impairment. We aim to reveal the relationship between the metabolic profile of parieto-occipital white matter (POWM) as evaluated with proton MR spectroscopy technique, with the visuoperceptual and visuoconstructional dysfunctions in DM1 patients. MR spectroscopy (3 Tesla) and neuropsychological evaluations were performed in 34 DM1 patients (19 F, age: 46.4 ± 12.1 years, disease duration: 18.7 ± 11.6 years). The content of neuro-axonal marker N-acetyl-aspartate, both relative to Creatine (NAA/Cr) and to myo-Inositol (NAA/mI) resulted significantly lower in DM1 patients compared to HC (p-values < 0.0001). NAA/Cr and NAA/mI correlated with the copy of the Rey-Osterrieth complex figure (r = 0.366, p = 0.033; r = 0.401, p = 0.019, respectively) and with Street's completion tests scores (r = 0.409, p = 0.016; r = 0.341, p = 0.048 respectively). The proportion of white matter hyperintensities within the MR spectroscopy voxel did not correlate with the metabolite content. In this study, POWM metabolic alterations in DM1 patients were not associated with the white matter morphological changes and correlated with specific neuropsychological deficits.

Convolutional Neural Network Techniques for Brain Tumor Classification (from 2015 to 2022): Review, Challenges, and Future Perspectives.

Convolutional neural networks (CNNs) constitute a widely used deep learning approach that has frequently been applied to the problem of brain tumor diagnosis. Such techniques still face some critical challenges in moving towards clinic application. The main objective of this work is to present a comprehensive review of studies using CNN architectures to classify brain tumors using MR images with the aim of identifying useful strategies for and possible impediments in the development of this technology. Relevant articles were identified using a predefined, systematic procedure. For each article, data were extracted regarding training data, target problems, the network architecture, validation methods, and the reported quantitative performance criteria. The clinical relevance of the studies was then evaluated to identify limitations by considering the merits of convolutional neural networks and the remaining challenges that need to be solved to promote the clinical application and development of CNN algorithms. Finally, possible directions for future research are discussed for researchers in the biomedical and machine learning communities. A total of 83 studies were identified and reviewed. They differed in terms of the precise classification problem targeted and the strategies used to construct and train the chosen CNN. Consequently, the reported performance varied widely, with accuracies of 91.63-100% in differentiating meningiomas, gliomas, and pituitary tumors (26 articles) and of 60.0-99.46% in distinguishing low-grade from high-grade gliomas (13 articles). The review provides a survey of the state of the art in CNN-based deep learning methods for brain tumor classification. Many networks demonstrated good performance, and it is not evident that any specific methodological choice greatly outperforms the alternatives, especially given the inconsistencies in the reporting of validation methods, performance metrics, and training data encountered. Few studies have focused on clinical usability.

Neuroplasticity Mechanisms in Frontal Brain Gliomas: A Preliminary Study.

Background: Pathological brain processes may induce adaptive cortical reorganization, however, the mechanisms underlying neuroplasticity that occurs in the presence of lesions in eloquent areas are not fully explained. The aim of this study was to evaluate functional compensatory cortical activations in patients with frontal brain gliomas during a phonemic fluency task and to explore correlations with cognitive performance, white matter tracts microstructural alterations, and tumor histopathological and molecular characterization. Methods: Fifteen patients with frontal glioma were preoperatively investigated with an MRI study on a 3T scanner and a subgroup underwent an extensive neuropsychological assessment. The hemispheric laterality index (LI) was calculated through phonemic fluency task functional MRI (fMRI) activations in the frontal, parietal, and temporal lobe parcellations. Diffusion-weighted images were acquired for all patients and for a group of 24 matched healthy volunteers. Arcuate Fasciculus (AF) and Frontal Aslant Tract (FAT) tractography was performed using constrained spherical deconvolution diffusivity modeling and probabilistic fiber tracking. All patients were operated on with a resective aim and underwent adjuvant therapies, depending on the final diagnosis. Results: All patients during the phonemic fluency task fMRI showed left hemispheric dominance in temporal and parietal regions. Regarding frontal regions (i.e., frontal operculum) we found right hemispheric dominance that increases when considering only those patients with tumors located on the left side. These latter activations positively correlate with verbal and visuo-spatial short-term memory, and executive functions. No correlations were found between the left frontal operculum and cognitive performance. Furthermore, patients with IDH-1 mutation and without TERT mutation, showed higher rightward frontal operculum fMRI activations and better cognitive performance in tests measuring general cognitive abilities, semantic fluency, verbal short-term memory, and executive functions. As for white matter tracts, we found left and right AF and FAT microstructural alterations in patients with, respectively, left-sided and right-side glioma compared to controls. Conclusions: Compensatory cortical activation of the corresponding region in the non-dominant hemisphere and its association with better cognitive performance and more favorable histopathological and molecular tumor characteristics shed light on the neuroplasticity mechanisms that occur in the presence of a tumor, helping to predict the rate of post-operative deficit, with the final goal of improving patients'quality of life.

What Can Resting-State fMRI Data Analysis Explain about the Functional Brain Connectivity in Glioma Patients?

Resting-state functional MRI has been increasingly implemented in imaging protocols for the study of functional connectivity in glioma patients as a sequence able to capture the activity of brain networks and to investigate their properties without requiring the patients' cooperation. The present review aims at describing the most recent results obtained through the analysis of resting-state fMRI data in different contexts of interest for brain gliomas: the identification and localization of functional networks, the characterization of altered functional connectivity, and the evaluation of functional plasticity in relation to the resection of the glioma. An analysis of the literature showed that significant and promising results could be achieved through this technique in all the aspects under investigation. Nevertheless, there is room for improvement, especially in terms of stability and generalizability of the outcomes. Further research should be conducted on homogeneous samples of glioma patients and at fixed time points to reduce the considerable variability in the results obtained across and within studies. Future works should also aim at establishing robust metrics for the assessment of the disruption of functional connectivity and its recovery at the single-subject level.

Association of Essential Tremor With Novel Risk Loci: A Genome-Wide Association Study and Meta-analysis.

Importance: Essential tremor (ET) is one of the most common movement disorders, affecting 5% of the general population older than 65 years. Common variants are thought to contribute toward susceptibility to ET, but no variants have been robustly identified. Objective: To identify common genetic factors associated with risk of ET. Design, Setting, and Participants: Case-control genome-wide association study. Inverse-variance meta-analysis was used to combine cohorts. Multicenter samples collected from European populations were collected from January 2010 to September 2019 as part of an ongoing study. Included patients were clinically diagnosed with or reported having ET. Control individuals were not diagnosed with or reported to have ET. Of 485 250 individuals, data for 483 054 passed data quality control and were used. Main Outcomes and Measures: Genotypes of common variants associated with risk of ET. Results: Of the 483 054 individuals included, there were 7177 with ET (3693 [51.46%] female; mean [SD] age, 62.66 [15.12] years), and 475 877 control individuals (253 785 [53.33%] female; mean [SD] age, 56.40 [17.6] years). Five independent genome-wide significant loci and were identified and were associated with approximately 18% of ET heritability. Functional analyses found significant enrichment in the cerebellar hemisphere, cerebellum, and axonogenesis pathways. Genetic correlation (r), which measures the degree of genetic overlap, revealed significant common variant overlap with Parkinson disease (r, 0.28; P = 2.38 × 10-8) and depression (r, 0.12; P = 9.78 × 10-4). A separate fine-mapping of transcriptome-wide association hits identified genes such as BACE2, LRRN2, DHRS13, and LINC00323 in disease-relevant brain regions, such as the cerebellum. Conclusions and Relevance: The results of this genome-wide association study suggest that a portion of ET heritability can be explained by common genetic variation and can help identify new common genetic risk factors for ET.

Circumventing the curse of dimensionality in magnetic resonance fingerprinting through a deep learning approach.

Magnetic resonance fingerprinting (MRF) is a rapidly developing approach for fast quantitative MRI. A typical drawback of dictionary-based MRF is an explosion of the dictionary size as a function of the number of reconstructed parameters, according to the "curse of dimensionality", which determines an explosion of resource requirements. Neural networks (NNs) have been proposed as a feasible alternative, but this approach is still in its infancy. In this work, we design a deep learning approach to MRF using a fully connected network (FCN). In the first part we investigate, by means of simulations, how the NN performance scales with the number of parameters to be retrieved in comparison with the standard dictionary approach. Four MRF sequences were considered: IR-FISP, bSSFP, IR-FISP-B1 , and IR-bSSFP-B1 , the latter two designed to be more specific for B1+ parameter encoding. Estimation accuracy, memory usage, and computational time required to perform the estimation task were considered to compare the scalability capabilities of the dictionary-based and the NN approaches. In the second part we study optimal training procedures by including different data augmentation and preprocessing strategies during training to achieve better accuracy and robustness to noise and undersampling artifacts. The study is conducted using the IR-FISP MRF sequence exploiting both simulations and in vivo acquisitions. Results demonstrate that the NN approach outperforms the dictionary-based approach in terms of scalability capabilities. Results also allow us to heuristically determine the optimal training strategy to make an FCN able to predict T1 , T2 , and M0 maps that are in good agreement with those obtained with the original dictionary approach. k-SVD denoising is proposed and found to be critical as a preprocessing step to handle undersampled data.

Molecular biomarkers correlate with brain grey and white matter changes in patients with mitochondrial m.3243A > G mutation.

INTRODUCTION: The mitochondrial DNA (mtDNA) m.3243A > G mutation in the MT-TL1 gene results in a multi-systemic disease, that is commonly associated with neurodegenerative changes in the brain. METHODS: Seventeen patients harboring the m3243A > G mutation were enrolled (age 43.1 ± 11.4 years, 10 M/7F). A panel of plasma biomarkers including lactate acid, alanine, L-arginine, fibroblast growth factor 21 (FGF-21), growth/differentiation factor 15 (GDF-15) and circulating cell free -mtDNA (ccf-mtDNA), as well as blood, urine and muscle mtDNA heteroplasmy were evaluated. Patients also underwent a brain standardized MR protocol that included volumetric T1-weighted images and diffusion-weighted MRI. Twenty sex- and age-matched healthy controls were included. Voxel-wise analysis was performed on T1-weighted and diffusion imaging, respectively with VBM (voxel-based morphometry) and TBSS (Tract-based Spatial Statistics). Ventricular lactate was also evaluated by 1H-MR spectroscopy. RESULTS: A widespread cortical gray matter (GM) loss was observed, more severe (p < 0.001) in the bilateral calcarine, insular, frontal and parietal cortex, along with infratentorial cerebellar cortex. High urine mtDNA mutation load, high levels of plasma lactate and alanine, low levels of plasma arginine, high levels of serum FGF-21 and ventricular lactate accumulation significantly (p < 0.05) correlated with the reduced brain GM density. Widespread microstructural alterations were highlighted in the white matter, significantly (p < 0.05) correlated with plasma alanine and arginine levels, with mtDNA mutation load in urine, with high level of serum GDF-15 and with high content of plasma ccf-mtDNA. CONCLUSIONS: Our results suggest that the synergy of two pathogenic mechanisms, mtDNA-related mitochondrial respiratory deficiency and defective nitric oxide metabolism, contributes to the brain neurodegeneration in m.3243A > G patients.

Anticholinergic Medication Burden in Parkinson's Disease Outpatients.

BACKGROUND: Individuals with Parkinson's disease (PD) may be especially vulnerable to future cognitive decline from anticholinergic medications. OBJECTIVE: To characterize anticholinergic medication burden, determine the co-occurrence of anticholinergic and cholinesterase inhibitors, and to assess the correlations among anticholinergic burden scales in PD outpatients. METHODS: We studied 670 PD outpatients enrolled in a clinic registry between 2012 and 2020. Anticholinergic burden was measured with the Anticholinergic Cognitive Burden Scale (ACB), Anticholinergic Drug Scale (ADS), Anticholinergic Risk Scale (ARS), and Drug Burden Index-Anticholinergic component (DBI-Ach). Correlations between scales were assessed with weighted kappa coefficients. RESULTS: Between 31.5 to 46.3% of PD patients were taking medications with anticholinergic properties. Among the scales applied, the ACB produced the highest prevalence of medications with anticholinergic properties (46.3%). Considering only medications with definite anticholinergic activity (scores of 2 or 3 on ACB, ADS, or ARS), the most common anticholinergic drug classes were antiparkinsonian (8.2%), antipsychotic (6.4%), and urological (3.3%) medications. Cholinesterase inhibitors and medications with anticholinergic properties were co-prescribed to 5.4% of the total cohort. The most highly correlated scales were ACB and ADS (κ= 0.71), ACB and ARS (κ= 0.67), and ADS and ARS (κ= 0.55). CONCLUSION: A high proportion of PD patients (20%) were either taking antiparkinsonian, urological, or antipsychotic anticholinergic medications or were co-prescribed anticholinergic medications and cholinesterase inhibitors. By virtue of its detection of a high prevalence of anticholinergic medication usage and its high correlation with other scales, our data support use of the ACB scale to assess anticholinergic burden in PD patients.

A deep learning approach for magnetic resonance fingerprinting: Scaling capabilities and good training practices investigated by simulations.

MR fingerprinting (MRF) is an innovative approach to quantitative MRI. A typical disadvantage of dictionary-based MRF is the explosive growth of the dictionary as a function of the number of reconstructed parameters, an instance of the curse of dimensionality, which determines an explosion of resource requirements. In this work, we describe a deep learning approach for MRF parameter map reconstruction using a fully connected architecture. Employing simulations, we have investigated how the performance of the Neural Networks (NN) approach scales with the number of parameters to be retrieved, compared to the standard dictionary approach. We have also studied optimal training procedures by comparing different strategies for noise addition and parameter space sampling, to achieve better accuracy and robustness to noise. Four MRF sequences were considered: IR-FISP, bSSFP, IR-FISP-B1, and IR-bSSFP-B1. A comparison between NN and the dictionary approaches in reconstructing parameter maps as a function of the number of parameters to be retrieved was performed using a numerical brain phantom. Results demonstrated that training with random sampling and different levels of noise variance yielded the best performance. NN performance was at least as good as the dictionary-based approach in reconstructing parameter maps using Gaussian noise as a source of artifacts: the difference in performance increased with the number of estimated parameters because the dictionary method suffers from the coarse resolution of the parameter space sampling. The NN proved to be more efficient in memory usage and computational burden, and has great potential for solving large-scale MRF problems.