Explore the vast range of titles available.

•A publicly available deep learning tool to segment the hypothalamus and its subunits.•Our tool outperforms inter-rater accuracy and approaches intra-rater precision level.•It can robustly generalise to unseen heterogeneous datasets.•It yields a rejection rate of less than 1% in a QC analysis performed on 675 scans.•It detects subtle subunit-specific hypothalamic atrophy in Alzheimer’s Disease. Despite the crucial role of the hypothalamus in the regulation of the human body, neuroimaging studies of this structure and its nuclei are scarce. Such scarcity partially stems from the lack of automated segmentation tools, since manual delineation suffers from scalability and reproducibility issues. Due to the small size of the hypothalamus and the lack of image contrast in its vicinity, automated segmentation is difficult and has been long neglected by widespread neuroimaging packages like FreeSurfer or FSL. Nonetheless, recent advances in deep machine learning are enabling us to tackle difficult segmentation problems with high accuracy. In this paper we present a fully automated tool based on a deep convolutional neural network, for the segmentation of the whole hypothalamus and its subregions from T1-weighted MRI scans. We use aggressive data augmentation in order to make the model robust to T1-weighted MR scans from a wide array of different sources, without any need for preprocessing. We rigorously assess the performance of the presented tool through extensive analyses, including: inter- and intra-rater variability experiments between human observers; comparison of our tool with manual segmentation; comparison with an automated method based on multi-atlas segmentation; assessment of robustness by quality control analysis of a larger, heterogeneous dataset (ADNI); and indirect evaluation with a volumetric study performed on ADNI. The presented model outperforms multi-atlas segmentation scores as well as inter-rater accuracy level, and approaches intra-rater precision. Our method does not require any preprocessing and runs in less than a second on a GPU, and approximately 10 seconds on a CPU. The source code as well as the trained model are publicly available at https://github.com/BBillot/hypothalamus_seg, and will also be distributed with FreeSurfer.

C9orf72 hexanucleotide repeat expansions are the most common cause of familial frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (ALS) worldwide. The clinical presentation is often indistinguishable from classic FTD or ALS, although neuropsychiatric symptoms are more prevalent and, for ALS, behavioural and cognitive symptoms occur more frequently. Pathogenic repeat length is in the hundreds or thousands, but the minimum length that increases risk of disease, and how or whether the repeat size affects phenotype, are unclear. Like in many patients with FTD and ALS, neuronal inclusions that contain TARDBP are seen, but are not universal, and the characteristic pathological finding is of dipeptide repeat (DPR) proteins, formed by unconventional repeat-associated non-ATG translation. Possible mechanisms of neurodegeneration include loss of C9orf72 protein and function, RNA toxicity, and toxicity from the DPR proteins, but which of these is the major pathogenic mechanism is not yet certain.

The neurodegenerative disorder Alzheimer's disease is characterised by the formation of β-amyloid plaques and neurofibrillary tangles in the brain parenchyma, which cause synapse and neuronal loss. This leads to clinical symptoms, such as progressive memory deficits. Clinically, these pathological changes can be detected in the cerebrospinal fluid and with brain imaging, although reliable blood tests for plaque and tangle pathologies remain to be developed. Plaques and tangles often co-exist with other brain pathologies, including aggregates of transactive response DNA-binding protein 43 and Lewy bodies, but the extent to which these contribute to the severity of Alzheimer's disease is currently unknown. In this ‘At a glance’ article and poster, we summarise the molecular biomarkers that are being developed to detect Alzheimer's disease and its related pathologies. We also highlight the biomarkers that are currently in clinical use and include a critical appraisal of the challenges associated with applying these biomarkers for diagnostic and prognostic purposes of Alzheimer's disease and related neurodegenerative disorders, also in their prodromal clinical phases.

Background Microglial dysfunction is implicated in frontotemporal lobar degeneration (FTLD). Although studies have reported excessive microglial activation or senescence (dystrophy) in Alzheimer’s disease (AD), few have explored this in FTLD. We examined regional patterns of microglial burden, activation and dystrophy in sporadic and genetic FTLD, sporadic AD and controls. Methods Immunohistochemistry was performed in frontal and temporal grey and white matter from 50 pathologically confirmed FTLD cases (31 sporadic, 19 genetic: 20 FTLD-tau, 26 FTLD-TDP, four FTLD-FUS), five AD cases and five controls, using markers to detect phagocytic (CD68-positive) and antigen-presenting (CR3/43-positive) microglia, and microglia in general (Iba1-positive). Microglial burden and activation (morphology) were assessed quantitatively for each microglial phenotype. Iba1-positive microglia were assessed semi-quantitatively for dystrophy severity and qualitatively for rod-shaped and hypertrophic morphology. Microglia were compared in each region between FTLD, AD and controls, and between different pathological subtypes of FTLD, including its main subtypes (FTLD-tau, FTLD-TDP, FTLD-FUS), and subtypes of FTLD-tau, FTLD-TDP and genetic FTLD. Microglia were also compared between grey and white matter within each lobe for each group. Results There was a higher burden of phagocytic and antigen-presenting microglia in FTLD and AD cases than controls, but activation was often not increased. Burden was generally higher in white matter than grey matter, but activation was greater in grey matter. However, microglia varied regionally according to FTLD subtype and disease mechanism. Dystrophy was more severe in FTLD and AD than controls, and more severe in white than grey matter, but this also varied regionally and was particularly extensive in FTLD due to progranulin ( GRN ) mutations. Presence of rod-shaped and hypertrophic microglia also varied by FTLD subtype. Conclusions This study demonstrates regionally variable microglial involvement in FTLD and links this to underlying disease mechanisms. This supports investigation of microglial dysfunction in disease models and consideration of anti-senescence therapies in clinical trials.

Neural network breakdown is a key issue in neurodegenerative disease, but the mechanisms are poorly understood. Here we investigated patterns of brain atrophy produced by defined molecular lesions in the two common forms of genetically mediated frontotemporal lobar degeneration (FTLD). Nine patients with progranulin (GRN) mutations and eleven patients with microtubule-associated protein tau (MAPT) mutations had T1 MR brain imaging. Brain volumetry and grey and white matter voxel-based morphometry (VBM) were used to assess patterns of cross-sectional atrophy in the two groups. In a subset of patients with longitudinal MRI rates of whole-brain atrophy were derived using the brain-boundary-shift integral and a VBM-like analysis of voxel-wise longitudinal volume change was performed. The GRN mutation group showed asymmetrical atrophy whereas the MAPT group showed symmetrical atrophy. Brain volumes were smaller in the GRN group with a faster rate of whole-brain atrophy. VBM delineated a common anterior cingulate–prefrontal–insular pattern of atrophy in both disease groups. Additional disease-specific profiles of grey and white matter loss were identified on both cross-sectional and longitudinal imaging: GRN mutations were associated with asymmetrical inferior frontal, temporal and inferior parietal lobe grey matter atrophy and involvement of long intrahemispheric association white matter tracts, whereas MAPT mutations were associated with symmetrical anteromedial temporal lobe and orbitofrontal grey matter atrophy and fornix involvement. The findings suggest that the effects of GRN and MAPT mutations are expressed in partly overlapping but distinct anatomical networks that link specific molecular dysfunction with clinical phenotype.

•We add diffusion MRI to Bayesian thalamic nuclei segmentation with structural MRI.•Adding fiber tracts to probabilistic atlases enables orientation modelling.•Thalamus segmentation from joint structural and diffusion MRI improves accuracy.•Atlas and companion segmentation code are freely distributed with FreeSurfer. The human thalamus is a highly connected brain structure, which is key for the control of numerous functions and is involved in several neurological disorders. Recently, neuroimaging studies have increasingly focused on the volume and connectivity of the specific nuclei comprising this structure, rather than looking at the thalamus as a whole. However, accurate identification of cytoarchitectonically designed histological nuclei on standard in vivo structural MRI is hampered by the lack of image contrast that can be used to distinguish nuclei from each other and from surrounding white matter tracts. While diffusion MRI may offer such contrast, it has lower resolution and lacks some boundaries visible in structural imaging. In this work, we present a Bayesian segmentation algorithm for the thalamus. This algorithm combines prior information from a probabilistic atlas with likelihood models for both structural and diffusion MRI, allowing segmentation of 25 thalamic labels per hemisphere informed by both modalities. We present an improved probabilistic atlas, incorporating thalamic nuclei identified from histology and 45 white matter tracts surrounding the thalamus identified in ultra-high gradient strength diffusion imaging. We present a family of likelihood models for diffusion tensor imaging, ensuring compatibility with the vast majority of neuroimaging datasets that include diffusion MRI data. The use of these diffusion likelihood models greatly improves identification of nuclear groups versus segmentation based solely on structural MRI. Dice comparison of 5 manually identifiable groups of nuclei to ground truth segmentations show improvements of up to 10 percentage points. Additionally, our chosen model shows a high degree of reliability, with median test-retest Dice scores above 0.85 for four out of five nuclei groups, whilst also offering improved detection of differential thalamic involvement in Alzheimer’s disease (AUROC 81.98%). The probabilistic atlas and segmentation tool will be made publicly available as part of the neuroimaging package FreeSurfer (https://freesurfer.net/fswiki/ThalamicNucleiDTI).

Frontotemporal dementia (FTD) is a highly heritable group of neurodegenerative disorders, with around 30% of patients having a strong family history. The majority of that heritability is accounted for by autosomal dominant mutations in the chromosome 9 open reading frame 72 (C9orf72), progranulin (GRN), and microtubule-associated protein tau (MAPT) genes, with mutations more rarely seen in a number of other genes. This review will discuss the recent updates in the field of genetic FTD. Age at symptom onset in genetic FTD is variable with recently identified genetic modifiers including TMEM106B (in GRN carriers particularly) and a polymorphism at a locus containing two overlapping genes LOC101929163 and C6orf10 (in C9orf72 carriers). Behavioural variant FTD (bvFTD) is the most common diagnosis in each of the genetic groups, although in C9orf72 carriers amyotrophic lateral sclerosis either alone, or with bvFTD, is also common. An atypical neuropsychiatric presentation is also seen in C9orf72 carriers and family members of carriers are at greater risk of psychiatric disorders including schizophrenia and autistic spectrum disorders. Large natural history studies of presymptomatic genetic FTD are now underway both in Europe/Canada (GENFI—the Genetic FTD Initiative) and in the US (ARTFL/LEFFTDS study), collaborating together under the banner of the FTD Prevention Initiative (FPI). These studies are taking forward the validation of cognitive, imaging and fluid biomarkers that aim to robustly measure disease onset, staging and progression in genetic FTD. Grey matter changes on MRI and hypometabolism on FDG-PET are seen at least 10 years before symptom onset with white matter abnormalities seen earlier, but the pattern and exact timing of changes differ between different genetic groups. In contrast, tau PET has yet to show promise in genetic FTD. Three key fluid biomarkers have been identified so far that are likely to be helpful in clinical trials—CSF or blood neurofilament light chain levels (in all groups), CSF or blood progranulin levels (in GRN carriers) and CSF poly(GP) dipeptide repeat protein levels (in C9orf72 carriers). Increased knowledge about genetic FTD has led to more clinical presymptomatic genetic testing but this has not yet been mirrored in the development of either an accepted FTD-specific testing protocol or provision of appropriate psychological support mechanisms for those living through the at-risk phase. This will become even more relevant as disease-modifying therapy trials start in each of the genetic groups over the next few years.

Cortical thickness estimation performed in-vivo via magnetic resonance imaging is an important technique for the diagnosis and understanding of the progression of neurodegenerative diseases. Currently, two different computational paradigms exist, with methods generally classified as either surface or voxel-based. This paper provides a much needed comparison of the surface-based method FreeSurfer and two voxel-based methods using clinical data. We test the effects of computing regional statistics using two different atlases and demonstrate that this makes a significant difference to the cortical thickness results. We assess reproducibility, and show that FreeSurfer has a regional standard deviation of thickness difference on same day scans that is significantly lower than either a Laplacian or Registration based method and discuss the trade off between reproducibility and segmentation accuracy caused by bending energy constraints. We demonstrate that voxel-based methods can detect similar patterns of group-wise differences as well as FreeSurfer in typical applications such as producing group-wise maps of statistically significant thickness change, but that regional statistics can vary between methods. We use a Support Vector Machine to classify patients against controls and did not find statistically significantly different results with voxel based methods compared to FreeSurfer. Finally we assessed longitudinal performance and concluded that currently FreeSurfer provides the most plausible measure of change over time, with further work required for voxel based methods. ►We compare FreeSurfer with voxel based cortical thickness estimation methods. ►Choice of atlas makes a significant difference on regional statistics. ►FreeSurfer has lower variance on same day scans than voxel based methods. ►In cross-sectional comparison, voxel based methods compared well with FreeSurfer. ►In longitudinal comparison, FreeSurfer most clearly separates groups.

Background The primary progressive aphasias (PPA) represent a group of usually sporadic neurodegenerative disorders with three main variants: the nonfluent or agrammatic variant (nfvPPA), the semantic variant (svPPA), and the logopenic variant (lvPPA). They are usually associated with a specific underlying pathology: nfvPPA with a primary tauopathy, svPPA with a TDP-43 proteinopathy, and lvPPA with underlying Alzheimer’s disease (AD). Little is known about their cause or pathophysiology, but prior studies in both AD and svPPA have suggested a role for neuroinflammation. In this study, we set out to investigate the role of chemokines across the PPA spectrum, with a primary focus on central changes in cerebrospinal fluid (CSF) Methods Thirty-six participants with sporadic PPA (11 svPPA, 13 nfvPPA, and 12 lvPPA) as well as 19 healthy controls were recruited to the study and donated CSF and plasma samples. All patients with lvPPA had a tau/Aβ42 biomarker profile consistent with AD, whilst this was normal in the other PPA groups and controls. We assessed twenty chemokines in CSF and plasma using Proximity Extension Assay technology: CCL2 (MCP-1), CCL3 (MIP-1a), CCL4 (MIP-1β), CCL7 (MCP-3), CCL8 (MCP-2), CCL11 (eotaxin), CCL13 (MCP-4), CCL19, CCL20, CCL23, CCL25, CCL28, CX3CL1 (fractalkine), CXCL1, CXCL5, CXCL6, CXCL8 (IL-8), CXCL9, CXCL10, and CXCL11. Results In CSF, CCL19 and CXCL6 were decreased in both svPPA and nfvPPA compared with controls whilst CXCL5 was decreased in the nfvPPA group with a borderline significant decrease in the svPPA group. In contrast, CCL2, CCL3 and CX3CL1 were increased in lvPPA compared with controls and nfvPPA (and greater than svPPA for CX3CL1). CXCL1 was also increased in lvPPA compared with nfvPPA but not the other groups. CX3CL1 was significantly correlated with CSF total tau concentrations in the controls and each of the PPA groups. Fewer significant differences were seen between groups in plasma, although in general, results were in the opposite direction to CSF, i.e. decreased in lvPPA compared with controls (CCL3 and CCL19), and increased in svPPA (CCL8) and nfvPPA (CCL13). Conclusion Differential alteration of chemokines across the PPA variants is seen in both CSF and plasma. Importantly, these results suggest a role for neuroinflammation in these poorly understood sporadic disorders, and therefore also a potential future therapeutic target.

Background Cerebrospinal fluid (CSF) biomarkers are increasingly being used to support a diagnosis of Alzheimer’s disease (AD). Their clinical utility for differentiating AD from non-AD neurodegenerative dementias, such as dementia with Lewy bodies (DLB) or frontotemporal dementia (FTD), is less well established. We aimed to determine the diagnostic utility of an extended panel of CSF biomarkers to differentiate AD from a range of other neurodegenerative dementias. Methods We used immunoassays to measure conventional CSF markers of amyloid and tau pathology (amyloid beta (Aβ)1–42, total tau (T-tau), and phosphorylated tau (P-tau)) as well as amyloid processing (AβX-38, AβX-40, AβX-42, soluble amyloid precursor protein (sAPP)α, and sAPPβ), large fibre axonal degeneration (neurofilament light chain (NFL)), and neuroinflammation (YKL-40) in 245 patients with a variety of dementias and 30 controls. Patients fulfilled consensus criteria for AD ( n  = 156), DLB ( n  = 20), behavioural variant frontotemporal dementia (bvFTD; n  = 45), progressive non-fluent aphasia (PNFA; n  = 17), and semantic dementia (SD; n  = 7); approximately 10% were pathology/genetically confirmed ( n  = 26). Global tests based on generalised least squares regression were used to determine differences between groups. Non-parametric receiver operating characteristic (ROC) curves and area under the curve (AUC) analyses were used to quantify how well each biomarker discriminated AD from each of the other diagnostic groups (or combinations of groups). CSF cut-points for the major biomarkers found to have diagnostic utility were validated using an independent cohort which included causes of AD ( n  = 104), DLB ( n  = 5), bvFTD ( n  = 12), PNFA ( n  = 3), SD ( n  = 9), and controls ( n  = 10). Results There were significant global differences in Aβ1–42, T-tau, T-tau/Aβ1–42 ratio, P-tau-181, NFL, AβX-42, AβX-42/X-40 ratio, APPα, and APPβ between groups. At a fixed sensitivity of 85%, AβX-42/X-40 could differentiate AD from controls, bvFTD, and SD with specificities of 93%, 85%, and 100%, respectively; for T-tau/Aβ1–42 these specificities were 83%, 70%, and 86%. AβX-42/X-40 had similar or higher specificity than Aβ1–42. No biomarker or ratio could differentiate AD from DLB or PNFA with specificity > 50%. Similar sensitivities and specificities were found in the independent validation cohort for differentiating AD and other dementias and in a pathology/genetically confirmed sub-cohort. Conclusions CSF AβX-42/X-40 and T-tau/Aβ1–42 ratios have utility in distinguishing AD from controls, bvFTD, and SD. None of the biomarkers tested had good specificity at distinguishing AD from DLB or PNFA.