Robust machine learning segmentation for large-scale analysis of heterogeneous clinical brain MRI datasets.

Every year, millions of brain MRI scans are acquired in hospitals, which is a figure considerably larger than the size of any research dataset. Therefore, the ability to analyze such scans could transform neuroimaging research. Yet, their potential remains untapped since no automated algorithm is robust enough to cope with the high variability in clinical acquisitions (MR contrasts, resolutions, orientations, artifacts, and subject populations). Here, we present SynthSeg+, an AI segmentation suite that enables robust analysis of heterogeneous clinical datasets. In addition to whole-brain segmentation, SynthSeg+ also performs cortical parcellation, intracranial volume estimation, and automated detection of faulty segmentations (mainly caused by scans of very low quality). We demonstrate SynthSeg+ in seven experiments, including an aging study on 14,000 scans, where it accurately replicates atrophy patterns observed on data of much higher quality. SynthSeg+ is publicly released as a ready-to-use tool to unlock the potential of quantitative morphometry.

Country-level gender inequality is associated with structural differences in the brains of women and men.

Gender inequality across the world has been associated with a higher risk to mental health problems and lower academic achievement in women compared to men. We also know that the brain is shaped by nurturing and adverse socio-environmental experiences. Therefore, unequal exposure to harsher conditions for women compared to men in gender-unequal countries might be reflected in differences in their brain structure, and this could be the neural mechanism partly explaining women's worse outcomes in gender-unequal countries. We examined this through a random-effects meta-analysis on cortical thickness and surface area differences between adult healthy men and women, including a meta-regression in which country-level gender inequality acted as an explanatory variable for the observed differences. A total of 139 samples from 29 different countries, totaling 7,876 MRI scans, were included. Thickness of the right hemisphere, and particularly the right caudal anterior cingulate, right medial orbitofrontal, and left lateral occipital cortex, presented no differences or even thicker regional cortices in women compared to men in gender-equal countries, reversing to thinner cortices in countries with greater gender inequality. These results point to the potentially hazardous effect of gender inequality on women's brains and provide initial evidence for neuroscience-informed policies for gender equality.

Differential cortical gray matter changes in early- and late-onset patients with amyotrophic lateral sclerosis.

Age at onset may be an important feature associated with distinct subtypes of amyotrophic lateral sclerosis (ALS). Little is known about the neuropathological mechanism of early-onset ALS (EO-ALS) and late-onset ALS (LO-ALS). Ninety ALS patients were divided into EO-ALS and LO-ALS group, and 128 healthy controls were matched into young controls(YCs) and old controls (OCs). A voxel-based morphometry approach was employed to investigate differences in gray matter volume (GMV). Significant age at onset-by-diagnosis interactions were found in the left parietal operculum, left precentral gyrus, bilateral postcentral gyrus, right occipital gyrus, and right orbitofrontal cortex. Post hoc analysis revealed a significant decrease in GMV in all affected regions of EO-ALS patients compared with YCs, with increased GMV in 5 of the 6 brain regions, except for the right orbitofrontal cortex, in LO-ALS patients compared with OCs. LO-ALS patients had a significantly increased GMV than EO-ALS patients after removing the aging effect. Correspondingly, GMV of the left postcentral gyrus correlated with disease severity in the 2 ALS groups. Our findings suggested that the pathological mechanisms in ALS patients with different ages at onset might differ. These findings provide unique insight into the clinical and biological heterogeneity of the 2 ALS subtypes.

A joint brain extraction and image quality assessment framework for fetal brain MRI slices.

Brain extraction and image quality assessment are two fundamental steps in fetal brain magnetic resonance imaging (MRI) 3D reconstruction and quantification. However, the randomness of fetal position and orientation, the variability of fetal brain morphology, maternal organs around the fetus, and the scarcity of data samples, all add excessive noise and impose a great challenge to automated brain extraction and quality assessment of fetal MRI slices. Conventionally, brain extraction and quality assessment are typically performed independently. However, both of them focus on the brain image representation, so they can be jointly optimized to ensure the network learns more effective features and avoid overfitting. To this end, we propose a novel two-stage dual-task deep learning framework with a brain localization stage and a dual-task stage for joint brain extraction and quality assessment of fetal MRI slices. Specifically, the dual-task module compactly contains a feature extraction module, a quality assessment head and a segmentation head with feature fusion for simultaneous brain extraction and quality assessment. Besides, a transformer architecture is introduced into the feature extraction module and the segmentation head. We utilize a multi-step training strategy to guarantee a stable and successful training of all modules. Finally, we validate our method by a 5-fold cross-validation and ablation study on a dataset with fetal brain MRI slices in different qualities, and perform a cross-dataset validation in addition. Experiments show that the proposed framework achieves very promising performance.

Latent diffusion model-based MRI superresolution enhances mild cognitive impairment prognostication and Alzheimer's disease classification.

INTRODUCTION: Timely diagnosis and prognostication of Alzheimer's disease (AD) and mild cognitive impairment (MCI) are pivotal for effective intervention. Artificial intelligence (AI) in neuroradiology may aid in such appropriate diagnosis and prognostication. This study aimed to evaluate the potential of novel diffusion model-based AI for enhancing AD and MCI diagnosis through superresolution (SR) of brain magnetic resonance (MR) images. METHODS: 1.5T brain MR scans of patients with AD or MCI and healthy controls (NC) from Alzheimer's Disease Neuroimaging Initiative 1 (ADNI1) were superresolved to 3T using a novel diffusion model-based generative AI (d3T*) and a convolutional neural network-based model (c3T*). Comparisons of image quality to actual 1.5T and 3T MRI were conducted based on signal-to-noise ratio (SNR), naturalness image quality evaluator (NIQE), and Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE). Voxel-based volumetric analysis was then conducted to study whether 3T* images offered more accurate volumetry than 1.5T images. Binary and multiclass classifications of AD, MCI, and NC were conducted to evaluate whether 3T* images offered superior AD classification performance compared to actual 1.5T MRI. Moreover, CNN-based classifiers were used to predict conversion of MCI to AD, to evaluate the prognostication performance of 3T* images. The classification performances were evaluated using accuracy, sensitivity, specificity, F1 score, Matthews correlation coefficient (MCC), and area under the receiver-operating curves (AUROC). RESULTS: Analysis of variance (ANOVA) detected significant differences in image quality among the 1.5T, c3T*, d3T*, and 3T groups across all metrics. Both c3T* and d3T* showed superior image quality compared to 1.5T MRI in NIQE and BRISQUE with statistical significance. While the hippocampal volumes measured in 3T* and 3T images were not significantly different, the hippocampal volume measured in 1.5T images showed significant difference. 3T*-based AD classifications showed superior performance across all performance metrics compared to 1.5T-based AD classification. Classification performance between d3T* and actual 3T was not significantly different. 3T* images offered superior accuracy in predicting the conversion of MCI to AD than 1.5T images did. CONCLUSIONS: The diffusion model-based MRI SR enhances the resolution of brain MR images, significantly improving diagnostic and prognostic accuracy for AD and MCI. Superresolved 3T* images closely matched actual 3T MRIs in quality and volumetric accuracy, and notably improved the prediction performance of conversion from MCI to AD.

Harmonizing three-dimensional MRI using pseudo-warping field guided GAN.

In pursuit of cultivating automated models for magnetic resonance imaging (MRI) to aid in diagnostics, an escalating demand for extensive, multisite, and heterogeneous brain imaging datasets has emerged. This potentially introduces biased outcomes when directly applied for subsequent analysis. Researchers have endeavored to address this issue by pursuing the harmonization of MRIs. However, most existing image-based harmonization methods for MRI are tailored for 2D slices, which may introduce inter-slice variations when they are combined into a 3D volume. In this study, we aim to resolve inconsistencies between slices by introducing a pseudo-warping field. This field is created randomly and utilized to transform a slice into an artificially warped subsequent slice. The objective of this pseudo-warping field is to ensure that generators can consistently harmonize adjacent slices to another domain, without being affected by the varying content present in different slices. Furthermore, we construct unsupervised spatial and recycle loss to enhance the spatial accuracy and slice-wise consistency across the 3D images. The results demonstrate that our model effectively mitigates inter-slice variations and successfully preserves the anatomical details of the images during the harmonization process. Compared to generative harmonization models that employ 3D operators, our model exhibits greater computational efficiency and flexibility.

In vivo T2 measurements of the fetal brain using single-shot fast spin echo sequences.

PURPOSE: We propose a quantitative framework for motion-corrected T2 fetal brain measurements in vivo and validate the single-shot fast spin echo (SS-FSE) sequence to perform these measurements. METHODS: Stacks of two-dimensional SS-FSE slices are acquired with different echo times (TE) and motion-corrected with slice-to-volume reconstruction (SVR). The quantitative T2 maps are obtained by a fit to a dictionary of simulated signals. The sequence is selected using simulated experiments on a numerical phantom and validated on a physical phantom scanned on a 1.5T system. In vivo quantitative T2 maps are obtained for five fetuses with gestational ages (GA) 21-35 weeks on the same 1.5T system. RESULTS: The simulated experiments suggested that a TE of 400 ms combined with the clinically utilized TEs of 80 and 180 ms were most suitable for T2 measurements in the fetal brain. The validation on the physical phantom confirmed that the SS-FSE T2 measurements match the gold standard multi-echo spin echo measurements. We measured average T2s of around 200 and 280 ms in the fetal brain grey and white matter, respectively. This was slightly higher than fetal T2* and the neonatal T2 obtained from previous studies. CONCLUSION: The motion-corrected SS-FSE acquisitions with varying TEs offer a promising practical framework for quantitative T2 measurements of the moving fetus.

Structural Changes in Brain White Matter Tracts Associated With Overactive Bladder Revealed by Diffusion Tensor Magnetic Resonance Imaging: Findings From a Symptoms of Lower Urinary Tract Dysfunction Research Network Cross-Sectional Case-Control Study.

PURPOSE: Our objective was to investigate structural changes in brain white matter tracts using diffusion tensor imaging (DTI) in patients with overactive bladder (OAB). MATERIALS AND METHODS: Treatment-seeking OAB patients and matched controls enrolled in the cross-sectional case-control LURN (Symptoms of Lower Urinary Tract Dysfunction Research Network) Neuroimaging Study received a brain DTI scan. Microstructural integrity of brain white matter was assessed using fractional anisotropy (FA) and mean diffusivity. OAB and urgency urinary incontinence (UUI) symptoms were assessed using the OAB Questionnaire Short-Form and International Consultation on Incontinence Questionnaire-Urinary Incontinence. The Lower Urinary Tract Symptoms Tool UUI questions and responses were correlated with FA values. RESULTS: Among 221 participants with evaluable DTI data, 146 had OAB (66 urinary urgency-only without UUI, 80 with UUI); 75 were controls. Compared with controls, participants with OAB showed decreased FA and increased mean diffusivity, representing greater microstructural abnormalities of brain white matter tracts among OAB participants. These abnormalities occurred in the corpus callosum, bilateral anterior thalamic radiation and superior longitudinal fasciculus tracts, and bilateral insula and parahippocampal region. Among participants with OAB, higher OAB Questionnaire Short-Form scores were associated with decreased FA in the left inferior fronto-occipital fasciculus, P < .0001. DTI differences between OAB and controls were driven by the urinary urgency-only (OAB-dry) but not the UUI (OAB-wet) subgroup. CONCLUSIONS: Abnormalities in microstructural integrity in specific brain white matter tracts were more frequent in OAB patients. More severe OAB symptoms were correlated with greater degree of microstructural abnormalities in brain white matter tracts in patients with OAB. TRIAL REGISTRATION: ClinicalTrials.gov: NCT02485808.

The impact of PET/CT and brain MRI for metastasis detection among patients with clinical T1-category lung cancer: Findings from a large-scale cohort study.

PURPOSE: [18F]-FDG PET/CT and brain MRI are common approaches to detect metastasis in patients of lung cancer. Current guidelines for the use of PET/CT and MRI in clinical T1-category lung cancer lack risk-based stratification and require optimization. This study stratified patients based on metastatic risk in terms of the lesions' size and morphological characteristics. METHODS: The detection rate of metastasis was measured in different sizes and morphological characteristics (solid and sub-solid) of tumors. To confirm the cut-off value for discriminating metastasis and overall survival (OS) prediction, the receiver operating characteristic (ROC) analysis was performed based on PET/CT metabolic parameters (SUVmax/SUVmean/SULpeak/MTV/TLG), followed by Kaplan-Meier analysis for survival in post-operation patients with and without PET/CT plus MRI. RESULTS: 2,298 patients were included. No metastasis was observed in patients with solid nodules < 8.0 mm and sub-solid nodules < 10.0 mm. The cut-off of PET/CT metabolic parameters on discriminating metastasis were 1.09 (SUVmax), 0.26 (SUVmean), 0.31 (SULpeak), 0.55 (MTV), and 0.81 (TLG), respectively. Patients undergoing PET/CT plus MRI exhibited longer OS compared to those who did not receive it in solid nodules ≥ 8.0 mm & sub-solid nodules ≥ 10.0 mm (HR, 0.44; p < 0.001); in solid nodules ≥ 8.0 mm (HR, 0.12; p<0.001) and in sub-solid nodules ≥ 10.0 mm (HR; 0.61; p=0.075), respectively. Compared to patients with metabolic parameters lower than cut-off values, patients with higher metabolic parameters displayed shorter OS: SUVmax (HR, 12.94; p < 0.001), SUVmean (HR, 11.33; p <0.001), SULpeak (HR, 9.65; p < 0.001), MTV (HR, 9.16; p = 0.031), and TLG (HR, 12.06; p < 0.001). CONCLUSION: The necessity of PET/CT and MRI should be cautiously evaluated in patients with solid nodules < 8.0 mm and sub-solid nodules < 10.0 mm, however, these examinations remained essential and beneficial for patients with solid nodules ≥ 8.0 mm and sub-solid nodules ≥ 10.0 mm.

T2*-Relaxometry MRI to Assess Third Trimester Placental and Fetal Brain Oxygenation and Placental Characteristics in Healthy Fetuses and Fetuses With Congenital Heart Disease.

BACKGROUND: Congenital heart disease (CHD) has been linked to impaired placental and fetal brain development. Assessing the placenta and fetal brain in parallel may help further our understanding of the relationship between development of these organs. HYPOTHESIS: 1) Placental and fetal brain oxygenation are correlated, 2) oxygenation in these organs is reduced in CHD compared to healthy controls, and 3) placental structure is altered in CHD. STUDY TYPE: Retrospective case-control. POPULATION: Fifty-one human fetuses with CHD (32 male; median [IQR] gestational age [GA] = 32.0 [30.9-32.9] weeks) and 30 from uncomplicated pregnancies with normal birth outcomes (18 male; median [IQR] GA = 34.5 [31.9-36.7] weeks). FIELD STRENGTH/SEQUENCE: 1.5 T single-shot multi-echo-gradient-echo echo-planar imaging. ASSESSMENT: Masking was performed using an automated nnUnet model. Mean brain and placental T2* and quantitative measures of placental texture, volume, and morphology were calculated. STATISTICAL TESTS: Spearman's correlation coefficient for determining the association between brain and placental T2*, and between brain and placental characteristics with GA. P-values for comparing brain T2*, placenta T2*, and placental characteristics between groups derived from ANOVA. Significance level P < 0.05. RESULTS: There was a significant positive association between placental and fetal brain T2* (⍴ = 0.46). Placental and fetal brain T2* showed a significant negative correlation with GA (placental T2* ⍴ = -0.65; fetal brain T2* ⍴ = -0.32). Both placental and fetal brain T2* values were significantly reduced in CHD, after adjusting for GA (placental T2*: control = 97 [±24] msec, CHD = 83 [±23] msec; brain T2*: control = 218 [±26] msec, CHD = 202 [±25] msec). Placental texture and morphology were also significantly altered in CHD (Texture: control = 0.84 [0.83-0.87], CHD = 0.80 [0.78-0.84]; Morphology: control = 9.9 [±2.2], CHD = 10.8 [±2.0]). For all fetuses, there was a significant positive association between placental T2* and placental texture (⍴ = 0.46). CONCLUSION: Placental and fetal brain T2* values are associated in healthy fetuses and those with CHD. Placental and fetal brain oxygenation are reduced in CHD. Placental appearance is significantly altered in CHD and shows associations with placental oxygenation, suggesting altered placental development and function may be related. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 3.

Nonalcoholic fatty liver disease, liver fibrosis, and structural brain imaging: The Cross-Cohort Collaboration.

BACKGROUND AND PURPOSE: Prior studies reported conflicting findings regarding the association of nonalcoholic fatty liver disease (NAFLD) and liver fibrosis with measures of brain health. We examined whether NAFLD and liver fibrosis are associated with structural brain imaging measures in middle- and old-age adults. METHODS: In this cross-sectional study among dementia- and stroke-free individuals, data were pooled from the Offspring and Third Generation cohorts of the Framingham Heart Study (FHS), the Rotterdam Study (RS), and the Study of Health in Pomerania. NAFLD was assessed through abdominal imaging. Transient hepatic elastography (FibroScan) was used to assess liver fibrosis in FHS and RS. Linear regression models were used to explore the relation of NAFLD and liver fibrosis with brain volumes, including total brain, gray matter, hippocampus, and white matter hyperintensities, adjusting for potential confounders. Results were combined using fixed effects meta-analysis. RESULTS: In total, 5660 and 3022 individuals were included for NAFLD and liver fibrosis analyses, respectively. NAFLD was associated with smaller volumes of total brain (ß = -3.5, 95% confidence interval [CI] = -5.4 to -1.7), total gray matter (ß = -1.9, 95% CI = -3.4 to -0.3), and total cortical gray matter (ß = -1.9, 95% CI = -3.7 to -0.01). In addition, liver fibrosis (defined as liver stiffness measure ≥8.2 kPa) was related to smaller total brain volumes (ß = -7.3, 95% CI = -11.1 to -3.5). Heterogeneity between studies was low. CONCLUSIONS: NAFLD and liver fibrosis may be directly related to brain aging. Larger and prospective studies are warranted to validate these findings and identify liver-related preventive strategies for neurodegeneration.

The clinical and neuroimaging differences between vascular parkinsonism and Parkinson's disease: a case-control study.

BACKGROUND: Parkinson's disease (PD) and vascular parkinsonism (VaP) have highly overlapping phenotypes, and different prognosis. This study comprehensively investigated the clinical, brain MRI and transcranial sonography differences between VaP and PD. METHODS: Forty-eight patients with PD, 27 patients with VaP, and 29 healthy controls were compared. All patients were assessed using the MDS-UPDRS, Berg Balance Scale (BBS), Ten-Meter Walking Test (10-MWT), Time Up and Go Test, and Non-Motor Symptoms Scale. Beck Depression Inventory, PD questionnaire- 39, international urine incontinence scale, cognitive assessment scales, MRI brain and transcranial colour-coded doppler. The study was registered on clinical-Trial.gov (NCT04308135) on 03/12/2020. RESULTS: VaP patients showed significantly older age of onset, shorter disease duration, lower drug doses and levodopa responsiveness, higher On and Off axial scores, On and Off BBS, higher On scores for PIGD, rigidity, bradykinesia and total motor MDS-UPDRS, lower On and Off tremor, lower-half predominance, lower asymmetrical presentation and symmetric index than PD patients. VaP patients had worse non-motor symptoms Scale (NMSS) than controls except for perceptual problems/hallucinations but better symptoms than PD patients except for urinary dysfunction. Quality of life (QoL) was impaired in VaP patients and was correlated with motor function and NMSs. The VaP group had significantly higher white matter lesions and brain atrophy, with lower hyperechogenicity of the substantia nigra and more impaired cerebral vascular resistance and vasoreactivity than the PD group. CONCLUSIONS: VaP has a characteristic motor and non-motor profile, with impaired QoL, white matter, and transcranial sonography abnormalities that differentiate it from PD. Further studies are warranted to explore the role of vascular lesions in the pathogenesis of VaP. TRIAL REGISTRATION: The registered identifier NCT04308135 on clinical-Trial.gov. Registered on 03/12/2020.

T1 hypointense brain lesions in NMOSD and its relevance with disability: a single institution cross-sectional study.

BACKGROUND: T1 hypointense lesions are considered a surrogate marker of tissue destruction. Although there is a shortage of evidence about T1 hypointense brain lesions, black holes, in patients with Neuromyelitis Optica Spectrum Disorder (NMOSD), the clinical significance of these lesions is not well determined. OBJECTIVES: The impact of T1 hypointense brain lesions on the clinical status and the disability level of patients with NMOSD was sought in this study. METHODS: A total of 83 patients with the final diagnosis of NMOSD were recruited. Aquaporin-4 measures were collected. The expanded disability status scale (EDSS) and MRI studies were also extracted. T1 hypointense and T2/FLAIR hyperintense lesions were investigated. The correlation of MRI findings, AQP-4, and EDSS was assessed. RESULTS: T1 hypointense brain lesions were detected in 22 patients. Mean ± SD EDSS was 3.7 ± 1.5 and significantly higher in patients with brain T1 hypointense lesions than those without them (p-value = 0.01). Noticeably, patients with more than four T1 hypointense lesions had EDSS scores ≥ 4. The presence of T2/FLAIR hyperintense brain lesions correlated with EDSS (3.6 ± 1.6 vs 2.3 ± 1.7; p-value = 0.01). EDSS was similar between those with and without positive AQP-4 (2.7 ± 1.6 vs. 3.2 ± 1.7; p-value = 0.17). Also, positive AQP-4 was not more prevalent in patients with T1 hypointense brain lesions than those without them (50.9 vs 45.4%; p-value = 0.8). CONCLUSION: We demonstrated that the presence of the brain T1-hypointense lesions corresponds to a higher disability level in NMOSD.

Stuttering associated with a pathogenic variant in the chaperone protein cyclophilin 40.

Stuttering is a common speech disorder that interrupts speech fluency and tends to cluster in families. Typically, stuttering is characterized by speech sounds, words or syllables which may be repeated or prolonged and speech that may be further interrupted by hesitations or 'blocks'. Rare variants in a small number of genes encoding lysosomal pathway proteins have been linked to stuttering. We studied a large four-generation family in which persistent stuttering was inherited in an autosomal dominant manner with disruption of the cortico-basal-ganglia-thalamo-cortical network found on imaging. Exome sequencing of three affected family members revealed the PPID c.808C>T (p.Pro270Ser) variant that segregated with stuttering in the family. We generated a Ppid p.Pro270Ser knock-in mouse model and performed ex vivo imaging to assess for brain changes. Diffusion-weighted MRI in the mouse revealed significant microstructural changes in the left corticospinal tract, as previously implicated in stuttering. Quantitative susceptibility mapping also detected changes in cortico-striatal-thalamo-cortical loop tissue composition, consistent with findings in affected family members. This is the first report to implicate a chaperone protein in the pathogenesis of stuttering. The humanized Ppid murine model recapitulates network findings observed in affected family members.

Understanding the relationship between type-2 diabetes, MRI markers of neurodegeneration and small vessel disease, and dementia risk: a mediation analysis.

To explore to which extent neurodegeneration and cerebral small vessel disease (SVD) could mediate the association between type-2 diabetes and higher dementia risk. The analytical sample consisted in 2228 participants, out of the Three-City study, aged 65 and older, free of dementia at baseline who underwent brain MRI. Diabetes was defined by medication intake or fasting or non-fasting elevated glucose levels. Dementia status was assessed every 2 to 3 years, during up to 12 years of follow-up. Brain parenchymal fraction (BPF) and white matter hyperintensities volume (WMHV) were selected as markers of neurodegeneration and cerebral SVD respectively. We performed a mediation analysis of the effect of baseline BPF and WMHV (mediators) on the association between diabetes and dementia risk using linear and Cox models adjusted for age, sex, education level, hypertension, hypercholesterolemia, BMI, smoking and alcohol drinking status, APOE-ε4 status, and study site. At baseline, 8.8% of the participants had diabetes. Diabetes (yes vs. no) was associated with higher WMHV (ßdiab = 0.193, 95% CI 0.040; 0.346) and lower BPF (ßdiab = -0.342, 95% CI -0.474; -0.210), as well as with an increased risk of dementia over 12 years of follow-up (HRdiab = 1.65, 95% CI 1.04; 2.60). The association between diabetes status and dementia risk was statistically mediated by higher WMHV (HRdiab=1.05, 95% CI 1.01; 1.11, mediated part = 10.8%) and lower BPF (HRdiab = 1.12, 95% CI 1.05; 1.20, mediated part = 22.9%). This study showed that both neurodegeneration and cerebral SVD statistically explained almost 30% of the association between diabetes and dementia.

Evaluation of 3D T1-weighted spoiled gradient echo MR image quality using artificial intelligence image reconstruction techniques in the pediatric brain.

PURPOSE: To assess image quality and diagnostic confidence of 3D T1-weighted spoiled gradient echo (SPGR) MRI using artificial intelligence (AI) reconstruction. MATERIALS AND METHODS: This prospective, IRB-approved study enrolled 50 pediatric patients (mean age = 11.8 ± 3.1 years) undergoing clinical brain MRI. In addition to standard of care (SOC) compressed SENSE (CS = 2.5), 3D T1-weighted SPGR images were obtained with higher CS acceleration factors (5 and 8) to evaluate the ability of AI reconstruction to improve image quality and reduce scan time. Images were reviewed independently on dedicated research PACS workstations by two neuroradiologists. Quantitative analysis of signal intensities to calculate apparent grey and white matter signal to noise (aSNR) and grey-white matter apparent contrast to noise ratios (aCNR) was performed. RESULTS: AI improved overall image quality compared to standard CS reconstruction in 35% (35/100) of evaluations in CS = 2.5 (average scan time = 221 ± 6.9 s), 100% (46/46) of CS = 5 (average scan time = 113.3 ± 4.6 s) and 94% (47/50) of CS = 8 (average scan time = 74.1 ± 0.01 s). Quantitative analysis revealed significantly higher grey matter aSNR, white matter aSNR and grey-white matter aCNR with AI reconstruction compared to standard reconstruction for CS 5 and 8 (all p-values < 0.001), however not for CS 2.5. CONCLUSIONS: AI reconstruction improved overall image quality and gray-white matter qualitative and quantitative aSNR and aCNR in highly accelerated (CS = 5 and 8) 3D T1W SPGR images in the majority of pediatric patients.

Vanishing white matter disease: imaging, clinical and molecular correlation in Brazilian families.

PURPOSE: To characterize Vanishing White Matter Disease (VWM) cases from a Brazilian University Tertiary hospital, focusing on brain magnetic resonance image (MRI) aspects, clinical and molecular data. METHODS: Medical records and brain MRI of 13 genetically confirmed VWM patients were reviewed. Epidemiological data such as age at symptom onset, gender and main symptoms were analyzed, along with genetic mutations and MRI characteristics, such as the distribution of white matter lesions and atrophy. RESULTS: The majority of patients were female, with the age of symptom onset ranging from 1 year and 6 months to 40 years. All mutations were identified in the EIF2B5 gene, the most prevalent being c.338G > A (p.Arg113His), and a novel mutation related to the disease was discovered, c.1051G > A (p.Gly351Ser). Trauma or infection were significant triggers. The most frequent symptoms were ataxia and limb spasticity. All MRI scans displayed deep white matter involvement, cystic degeneration, with U-fibers relatively spared and a predilection for the frontoparietal region. Lesions in the corpus callosum and posterior fossa were present in all patients. Follow-up exams revealed the evolution of white matter lesions and cerebral atrophy, which correlated with clinical deterioration. CONCLUSIONS: VWM affects various age groups, with a significant clinical and genetic variability. A novel mutation associated with the disease is highlighted. MRI reveals a typical pattern of white matter involvement, characterized by diffuse lesions in the periventricular and deep regions, with subsequent extension to the subcortical areas, accompanied by cystic degeneration, and plays a crucial role in diagnosis and follow-up.

An individual-level weighted artificial neural network method to improve the systematic bias in BrainAGE analysis.

BrainAGE is a commonly used machine learning technique to measure the accelerated/delayed development pattern of human brain structure/function with neuropsychiatric disorders. However, recent studies have shown a systematic bias ("regression toward mean" effect) in the BrainAGE method, which indicates that the prediction error is not uniformly distributed across Chronological Ages: for the older individuals, the Brain Ages would be under-estimated but would be over-estimated for the younger individuals. In the present study, we propose an individual-level weighted artificial neural network method and apply it to simulation datasets (containing 5000 simulated subjects) and a real dataset (containing 135 subjects). Results show that compared with traditional machine learning methods, the individual-level weighted strategy can significantly reduce the "regression toward mean" effect, while the prediction performance can achieve the comparable level with traditional machine learning methods. Further analysis indicates that the sigmoid active function for artificial neural network shows better performance than the relu active function. The present study provides a novel strategy to reduce the "regression toward mean" effect of BrainAGE analysis, which is helpful to improve accuracy in exploring the atypical brain structure/function development pattern of neuropsychiatric disorders.

The association between cannabis use and neuroimaging measures in older adults: findings from the UK biobank.

BACKGROUND: Cannabis use has increased in recent years. However, the long-term implications of cannabis use on brain health remain unknown. We explored the associations of cannabis use with volumetric brain magnetic resonance imaging (MRI) measures in dementia-free older adults. METHODS: This cross-sectional and longitudinal study included dementia-free participants of the UK Biobank aged ≥60 years. Linear regression models were used to evaluate the association of cannabis use and patterns of use with volumetric brain MRI measures. The association between cannabis use and change in brain MRI measures over time was also tested. All models were adjusted for potential confounders. RESULTS: The sample included 19,932 participants (mean age 68 ± 5 years, 48% men), 3,800 (19%) reported lifetime use of cannabis. Cannabis use was associated with smaller total, white, grey and peripheral cortical grey matter volumes (B = -6,690 ± 1,157; P < 0.001, B = -4,396 ± 766; P < 0.001, B = -2,140 ± 690; P = 0.002 and B = -2,451 ± 606; P < 0.001, respectively). Among cannabis users, longer duration of use was associated with smaller total brain, grey and cortical grey matter volumes (B = -7,878 ± 2,396; P = 0.001, B = -5,411 ± 1,430; P < 0.001, B = -5,396 ± 1,254; P < 0.001, respectively), and with increased white matter hyperintensity volume (B = 0.09 ± 0.03; P = 0.008). Additionally, current vs. former users (B = -10,432 ± 4,395; P = 0.020) and frequent versus non-frequent users (B = -2,274 ± 1,125; P = 0.043) had smaller grey and cortical grey matter volumes, respectively. No significant associations were observed between cannabis use and change in brain MRI measures. DISCUSSION: Our findings suggest that cannabis use, particularly longer duration and frequent use, may be related to smaller grey and white matter volumes in older ages, but not to late-life changes in these measures over time.

Stroke in a young adult: looking beyond the diffusion-weighted imaging sequence.

A 37-year-old migrainous woman was referred several hours after a sudden dysarthria lasting less than 1 min. Brain MRI undertaken a few hours later showed no acute cerebral infarction on the diffusion-weighted imaging (DWI) sequence. However, an intraluminal thrombus within a distal branch of the middle cerebral artery (MCA) was seen on the T2*-weighted gradient echo images as well as hyperintense vessels related to slow flow beyond the occlusion's site. Control MRI 24 h later confirmed the presence of a small infarct affecting the distal portion of the MCA territory but the thrombus was no longer visible. A comprehensive etiologic workup revealed the presence of a patent foramen ovale, as the underlying cause of the patient's stroke. Though the sensitivity of DWI sequence for acute ischemia is high, false negative exist particularly in cases of transient ischemic attack, brainstem or lacunar strokes. A careful brain MRI examination or the use of perfusion weighted imaging sequences are useful to avoid misdiagnosis particularly in young patients.