Brain Atrophy Subtypes and the ATN Classification Scheme in Alzheimer's Disease.

INTRODUCTION: We investigated the association between atrophy subtypes of Alzheimer's disease (AD), the ATN classification scheme, and key demographic and clinical factors in 2 cohorts with different source characteristics (a highly selective research-oriented cohort, the Alzheimer's Disease Neuroimaging Initiative [ADNI]; and a naturalistic heterogeneous clinically oriented cohort, Karolinska Imaging Dementia Study [KIDS]). METHODS: A total of 382 AD patients were included. Factorial analysis of mixed data was used to investigate associations between AD subtypes based on brain atrophy patterns, ATN profiles based on cerebrospinal fluid biomarkers, and age, sex, Mini Mental State Examination (MMSE), cerebrovascular disease (burden of white matter signal abnormalities, WMSAs), and APOE genotype. RESULTS: Older patients with high WMSA burden, belonging to the typical AD subtype and showing A+T+N+ or A+T+N- profiles clustered together and were mainly from ADNI. Younger patients with low WMSA burden, limbic-predominant or minimal atrophy AD subtypes, and A+T-N- or A+T-N+ profiles clustered together and were mainly from KIDS. APOE ε4 carriers more frequently showed the A+T-N- and A+T+N- profiles. CONCLUSIONS: Our findings align with the recent framework for biological subtypes of AD: the combination of risk factors, protective factors, and brain pathologies determines belonging of AD patients to distinct subtypes.

A signature pattern of cortical atrophy in dementia with Lewy bodies: A study on 333 patients from the European DLB consortium.

INTRODUCTION: We explored regional brain atrophy patterns and their clinical correlates in dementia with Lewy bodies (DLB). METHODS: In this multicentre study, we included a total of 333 patients with DLB, 352 patients with Alzheimer's disease (AD), and 233 normal controls and used medial temporal lobe atrophy, posterior atrophy, and frontal atrophy (GCA-F) visual rating scales. Patients were classified according to four atrophy patterns. RESULTS: Patients with DLB had higher scores on all the three atrophy scales than normal controls but had less medial temporal lobe atrophy than those with AD (all P values < .001). A signature hippocampal-sparing pattern of regional atrophy was observed in DLB. The magnetic resonance imaging measures showed 65% ability to discriminate between DLB and AD and marginally contributed to the discrimination over and above the core clinical features. DISCUSSION: The most common pattern of atrophy of DLB was hippocampal-sparing. Future studies should explore whether comorbid AD pathology underlies the atrophy patterns seen in DLB.

Association between vascular comorbidity and progression of Alzheimer's disease: a two-year observational study in Norwegian memory clinics.

BACKGROUND: Vascular risk factors increase the risk of Alzheimer's disease (AD), but there is limited evidence on whether comorbid vascular conditions and risk factors have an impact on disease progression. The aim of this study was to examine the association between vascular disease and vascular risk factors and progression of AD. METHODS: In a longitudinal observational study in three Norwegian memory clinics, 282 AD patients (mean age 73.3 years, 54% female) were followed for mean 24 (16-37) months. Vascular risk factors and vascular diseases were registered at baseline, and the vascular burden was estimated by the Framingham Stroke Risk Profile (FSRP). Cerebral medical resonance images (MRIs) were assessed for white matter hyperintensities (WMH), lacunar and cortical infarcts. The associations between vascular comorbidity and progression of dementia as measured by annual change in Clinical Dementia Rating Sum of Boxes (CDR-SB) scores were analysed by multiple regression analyses, adjusted for age and sex. RESULTS: Hypertension occurred in 83%, hypercholesterolemia in 53%, diabetes in 9%, 41% were overweight, and 10% were smokers. One third had a history of vascular disease; 16% had heart disease and 15% had experienced a cerebrovascular event. MRI showed lacunar infarcts in 16%, WMH with Fazekas score 2 in 26%, and Fazekas score 3 in 33%. Neither the vascular risk factors and diseases, the FSRP score, nor cerebrovascular disease was associated with disease progression in AD. CONCLUSIONS: Although vascular risk factors and vascular diseases were prevalent, no impact on the progression of AD after 2 years was shown.

Comparison of automated volumetry of the hippocampus using NeuroQuant® and visual assessment of the medial temporal lobe in Alzheimer's disease.

Background Different clinically feasible methods for evaluation of medial temporal lobe atrophy exists and are useful in diagnostic work-up of Alzheimer's disease (AD). Purpose To compare the diagnostic properties of two clinically available magnetic resonance imaging (MRI)-based methods-an automated volumetric software, NeuroQuant® (NQ) (evaluation of hippocampus volume) and the Scheltens scale (visual evaluation of medial temporal lobe atrophy [MTA])-in patients with AD dementia, and subjective and mild cognitive impairment (non-dementia). Material and Methods MRIs from 56 patients (31 AD, 25 non-dementia) were assessed with both methods. Correlations between the methods were calculated and receiver operating curve (ROC) analyses that yield area under the curve (AUC) statistics were conducted. Results High correlations were found between the two MRI assessments for the total hippocampal volume measured with NQ and mean MTA score (-0.753, P < 0.001), for the right (-0.767, P < 0.001), and for the left (-0.675, P < 0.001) sides. The NQ total measure yielded somewhat higher AUC (0.88, "good") compared to the MTA mean measure (0.80, "good") in the comparison of patients with AD and non-dementia, but the accuracy was in favor of the MTA scale. Conclusion The two methods correlated highly and both methods reached equally "good" power.

Visual Evaluation of Medial Temporal Lobe Atrophy as a Clinical Marker of Conversion from Mild Cognitive Impairment to Dementia and for Predicting Progression in Patients with Mild Cognitive Impairment and Mild Alzheimer's Disease.

BACKGROUND/AIMS: To evaluate whether visual assessment of medial temporal lobe atrophy (vaMTA) can predict 2-year conversion from mild cognitive impairment (MCI) to dementia and progression of MCI and Alzheimer's disease dementia as measured by the Clinical Dementia Rating Scale Sum of Boxes score (CDR-SB). METHODS: vaMTA was performed in 94 patients with MCI according to the Winblad criteria and in 124 patients with AD according to ICD-10 and NINCDS-ADRDA criteria. Demographic data, the Consortium to Establish a Registry for Alzheimer's Disease 10-word delayed recall, APOE ɛ4 status, Cornell Scale for Depression in Dementia, and comorbid hypertension were used as covariates. RESULTS: vaMTA was associated with MCI conversion in an unadjusted model but not in an adjusted model (p = 0.075), where delayed recall and APOE ɛ4 status were significant predictors. With CDR-SB change as the outcome, an interaction between vaMTA and diagnosis was found, but in the adjusted model only delayed recall and age were significant predictors. For vaMTA below 2, the association between vaMTA and CDR-SB change differed between diagnostic groups. Similar results were found based on a trajectory analysis. CONCLUSION: In adjusted models, memory function, APOE ɛ4 status and age were significant predictors of disease progression, not vaMTA. The association between vaMTA and CDR-SB change was different in patients with MCI and Alzheimer's disease dementia.

Quantitative validation of a visual rating scale for frontal atrophy: associations with clinical status, APOE e4, CSF biomarkers and cognition.

OBJECTIVES: To validate a visual rating scale of frontal atrophy with quantitative imaging and study its association with clinical status, APOE ε4, CSF biomarkers, and cognition. METHODS: The AddNeuroMed and ADNI cohorts were combined giving a total of 329 healthy controls, 421 mild cognitive impairment patients, and 286 Alzheimer's disease (AD) patients. Thirty-four patients with frontotemporal dementia (FTD) were also included. Frontal atrophy was assessed with the frontal sub-scale of the global cortical atrophy scale (GCA-F) on T1-weighted images. Automated imaging markers of cortical volume, thickness, and surface area were evaluated. Manual tracing was also performed. RESULTS: The GCA-F scale reliably reflects frontal atrophy, with orbitofrontal, dorsolateral, and motor cortices being the regions contributing most to the GCA-F ratings. GCA-F primarily reflects reductions in cortical volume and thickness, although it was able to detect reductions in surface area too. The scale showed significant associations with clinical status and cognition. CONCLUSION: The GCA-F scale may have implications for clinical practice as supportive diagnostic tool for disorders demonstrating predominant frontal atrophy such as FTD and the executive presentation of AD. We believe that GCA-F is feasible for use in clinical routine for the radiological assessment of dementia and other disorders. KEY POINTS: • The GCA-F visual rating scale reliably reflects frontal brain atrophy. • Orbitofrontal, dorsolateral, and motor cortices are the most contributing regions. • GCA-F shows significant associations with clinical status and cognition. • GCA-F may be supportive diagnostic tool for disorders demonstrating predominant frontal atrophy. • GCA-F may be feasible for use in radiological routine.

Electroencephalography Is a Good Complement to Currently Established Dementia Biomarkers.

BACKGROUND/AIMS: Dementia biomarkers that are accessible and easily applicable in nonspecialized clinical settings are urgently needed. Quantitative electroencephalography (qEEG) is a good candidate, and the statistical pattern recognition (SPR) method has recently provided promising results. We tested the diagnostic value of qEEG-SPR in comparison to cognition, structural imaging, and cerebrospinal fluid (CSF) biomarkers. METHODS: A total of 511 individuals were recruited from the multicenter NORD EEG study [141 healthy controls, 64 subjective cognitive decline, 124 mild cognitive impairment, 135 Alzheimer's disease (AD), 15 dementia with Lewy bodies/Parkinson's disease with dementia (DLB/PDD), 32 other dementias]. The EEG data were recorded in a standardized way. Structural imaging data were visually rated using scales of atrophy in the medial temporal, frontal, and posterior cortex. RESULTS: qEEG-SPR outperformed structural imaging, cognition, and CSF biomarkers in DLB/PDD diagnosis, outperformed structural imaging in AD diagnosis, and improved the differential diagnosis of AD. In addition, qEEG-SPR allowed differentiation of two clinically different AD subtypes. CONCLUSION: Adding qEEG to the diagnostic workup substantially increases the detection of AD pathology even in pre-dementia stages and improves differential diagnosis. EEG could serve as a good complement to currently established dementia biomarkers since it is cheap, noninvasive, and extensively applied outside academic centers.

Automated CT-based segmentation and quantification of total intracranial volume.

OBJECTIVES: To develop an algorithm to segment and obtain an estimate of total intracranial volume (tICV) from computed tomography (CT) images. MATERIALS AND METHODS: Thirty-six CT examinations from 18 patients were included. Ten patients were examined twice the same day and eight patients twice six months apart (these patients also underwent MRI). The algorithm combines morphological operations, intensity thresholding and mixture modelling. The method was validated against manual delineation and its robustness assessed from repeated imaging examinations. Using automated MRI software, the comparability with MRI was investigated. Volumes were compared based on average relative volume differences and their magnitudes; agreement was shown by a Bland-Altman analysis graph. RESULTS: We observed good agreement between our algorithm and manual delineation of a trained radiologist: the Pearson's correlation coefficient was r = 0.94, tICVml[manual] = 1.05 × tICVml[automated] - 33.78 (R(2) = 0.88). Bland-Altman analysis showed a bias of 31 mL and a standard deviation of 30 mL over a range of 1265 to 1526 mL. CONCLUSIONS: tICV measurements derived from CT using our proposed algorithm have shown to be reliable and consistent compared to manual delineation. However, it appears difficult to directly compare tICV measures between CT and MRI. KEY POINTS: • Automated estimation of tICV is in good agreement with manual tracing. • Consistent tICV estimations from repeated measurements demonstrate the robustness of the algorithm. • Automatically segmented volumes seem less variable than those from manual tracing. • Unbiased and automated tlCV estimation is possible from CT.

Medial temporal lobe atrophy and depressive symptoms in elderly patients with and without Alzheimer disease.

OBJECTIVE: To determine whether depressive symptoms are associated with medial temporal lobe atrophy in older people with and without Alzheimer disease (AD). METHOD: A total of 368 memory clinic patients with AD, mild cognitive impairment, and subjective cognitive impairment (SCI) were included. Depressive symptoms were defined as a score of 8 or higher on Cornell Scale for Depression in Dementia or use of antidepressant medications. Magnetic resonance imaging and computer tomography scans were rated for medial temporal lobe atrophy (MTA), using the Scheltens scale. For a subsample (n = 57 patients), hippocampal volume was manually traced. RESULTS: Based on visual assessment, AD patients with depressive symptoms had less atrophy of the right medial temporal lobe (odds ratio [OR] for having MTA: 0.39; 95% confidence interval [CI] 0.16-0.99) and decreased scores on Scheltens scale for the left medial temporal lobe (OR: 0.43, 95% CI 0.19-0.96) in comparison to AD patients without depressive symptoms. In the subgroup where manual tracing was used to measure hippocampal volume, people with SCI experiencing depressive symptoms had smaller right (mean difference: 0.28 cm(3); P = .005) and left (mean difference 0.32 cm(3); P = .002) hippocampal volumes compared to people with SCI who did not have depressive symptoms. CONCLUSION: Hippocampal atrophy was more pronounced among patients having SCI with depressive symptoms, while the medial temporal lobe was less atrophic in patients having AD with depressive symptoms than those without depressive symptoms. These findings suggest that different mechanisms underlie depression in older people with and without AD and may explain some of the inconsistent observations in previous studies.

Overtime reliability of medial temporal lobe atrophy rating in a clinical setting.

BACKGROUND: Medial temporal lobe atrophy (MTA) is one of the first magnetic resonance imaging (MRI) signs in patients with Alzheimer's disease (AD) and used as a measure of disease progression. Visual assessment of MTA is easy to perform but the reliability of MTA rating over time has not been studied. PURPOSE: To investigate what happens to the MTA rating scores if two radiologists rate the same MRI scans six times over a period of 1 year. MATERIAL AND METHODS: One hundred outpatients were included in this study. All patients underwent MRI with a protocol and sequences used for geriatric patients, according to local clinical standards. One neuroradiologist and one general radiologist independently of each other performed retrospective visual assessments of MTA six times, using the same scans, over a period of 1 year. RESULTS: Intra-rater kappa varied between κ 0.65 and 0.84 for the neuroradiologist and κ 0.38 and 0.74 for the general radiologist. Intra-rater weighted kappa (wκ) values showed almost perfect agreement for both investigators (wκ 0.83-0.94). Inter-rater reliability showed fair to moderate agreement, with the kappa value varying from κ 0.29 to 0.48 and weighted kappa values ranging from wκ 0.72 to 0.84. There was a statistically significant difference in rating between the two investigators. CONCLUSION: Visual assessment of MTA repeated over time has a high grade of reproducibility when performed by an experienced investigator. The reproducibility drops when assessment is rarely performed. Inter-rater reliability is low when two investigators not working together are compared.

Comparison between visual assessment of MTA and hippocampal volumes in an elderly, non-demented population.

BACKGROUND: It is important to have a replicable easy method for monitoring atrophy progression in Alzheimer's disease. Volumetric methods for calculating hippocampal volume are time-consuming and commonly used in research. Visual assessments of medial temporal lobe atrophy (vaMTA) is a rapid method for clinical use. This method has not been tested in a large non-demented population in comparison with volumetry measurements. Since hippocampal volume decreases with time even in normal aging there is also a need to study the normal age differences of medial temporal lobe atrophy. PURPOSE: To compare visual assessment of medial temporal lobe atrophy (vaMTA) with hippocampal volume in a healthy, non-demented elderly population. To describe normal ageing using vaMTA. MATERIAL AND METHODS: Non-demented individuals aged 60, 66, 72, 78, 81, 84, and ≥87 years old were recruited from the Swedish National study on Ageing and Care in Kungsholmen (SNAC-K), Sweden. Standard magnetic resonance imaging (MRI) scans, vaMTA, and calculations of hippocampal volumes were performed in 544 subjects. RESULTS: Significant correlation (r(s) = -0.32, P < 0.001, sin; and r(s) = -0.26, P < 0.001, dx) was found between hippocampal volume measurements and vaMTA. In normal ageing, almost 95% of ≤66-year-olds had a medial temporal lobe atrophy (MTA) score ≤1, with possible scores ranging from 0 to 4. Subjects aged 72, 78, and 81 years scored ≤2, while the two oldest age groups had scores ≤3. CONCLUSION: There was a highly significant correlation between volumetric measurements of the hippocampus and MTA scoring. In normal ageing, there is increasing MTA score. For non-demented elderly individuals ≤70 years, an MTA score of 0-1 may be considered normal, compared with MTA ≤2 for 70-80-years and MTA 3 for >80-year-old individuals.

MRI Markers of Small Vessel Disease and the APOE Allele in Cognitive Impairment.

Objective: The apolipoprotein E (APOE) ε4 allele is the main genetic risk factor for dementia and Alzheimer's disease (AD), but the underlying mechanism for the increased risk is not well understood. Cerebral small vessel disease (SVD) is prevalent among patients with cognitive impairment and is thought to play an important role in the pathophysiology of dementia. We aimed to investigate the association between the APOE ε genotype and magnetic resonance imaging (MRI) markers of SVD in a memory clinic population. Material and Methods: This is a cross-sectional study with a total of 520 patients undergoing dementia investigation, including an MRI brain scan and APOE genotyping in all patients enrolled, and cerebrospinal fluid (CSF) analysis for routine AD biomarkers in 399 patients. MR images were assessed for markers of SVD: cerebral microbleeds (CMBs), cortical superficial siderosis, intracerebral hemorrhage, white matter hyperintensities, lacunar infarcts, and enlarged perivascular spaces. Results: Apolipoprotein E carriers with AD had a higher number of CMBs when looking at all brain regions and lobar brain regions (p < 0.001). A lower number of CMBs were seen in APOE ε2 (p < 0.05), ε3 and ε3/3 carriers (p < 0.001) when looking at all brain regions. A higher number of CMBs in deep and infratentorial regions were seen in APOE ε2 and ε3 (p < 0.05). In APOE ε4/4 carriers, CMBs, cortical superficial siderosis, white matter hyperintensities, and enlarged perivascular spaces were associated with lower levels of CSF amyloid ß (Aß) 42 in the whole cohort, and in individuals with AD and mild cognitive impairment (p < 0.05). Conclusion: Apolipoprotein E ε4 is associated with MRI markers of SVD related to amyloid pathology, specifically CMBs and Aß42 plaque formation in the brain, as reflected by decreased CSF Aß42 levels, whereas APOE ε3 and ε2 are associated with the markers of hypertensive arteriopathy, as reflected by the association with CMBs in deep and infratentorial brain regions.

Parsing heterogeneity within dementia with Lewy bodies using clustering of biological, clinical, and demographic data.

BACKGROUND: Dementia with Lewy bodies (DLB) includes various core clinical features that result in different phenotypes. In addition, Alzheimer's disease (AD) and cerebrovascular pathologies are common in DLB. All this increases the heterogeneity within DLB and hampers clinical diagnosis. We addressed this heterogeneity by investigating subgroups of patients with similar biological, clinical, and demographic features. METHODS: We studied 107 extensively phenotyped DLB patients from the European DLB consortium. Factorial analysis of mixed data (FAMD) was used to identify dimensions in the data, based on sex, age, years of education, disease duration, Mini-Mental State Examination (MMSE), cerebrospinal fluid (CSF) levels of AD biomarkers, core features of DLB, and regional brain atrophy. Subsequently, hierarchical clustering analysis was used to subgroup individuals based on the FAMD dimensions. RESULTS: We identified 3 dimensions using FAMD that explained 38% of the variance. Subsequent hierarchical clustering identified 4 clusters. Cluster 1 was characterized by amyloid-ß and cerebrovascular pathologies, medial temporal atrophy, and cognitive fluctuations. Cluster 2 had posterior atrophy and showed the lowest frequency of visual hallucinations and cognitive fluctuations and the worst cognitive performance. Cluster 3 had the highest frequency of tau pathology, showed posterior atrophy, and had a low frequency of parkinsonism. Cluster 4 had virtually normal AD biomarkers, the least regional brain atrophy and cerebrovascular pathology, and the highest MMSE scores. CONCLUSIONS: This study demonstrates that there are subgroups of DLB patients with different biological, clinical, and demographic characteristics. These findings may have implications in the diagnosis and prognosis of DLB, as well as in the treatment response in clinical trials.

The cholinergic system in subtypes of Alzheimer's disease: an in vivo longitudinal MRI study.

BACKGROUND: The heterogeneity within Alzheimer's disease (AD) seriously challenges the development of disease-modifying treatments. We investigated volume of the basal forebrain, hippocampus, and precuneus in atrophy subtypes of AD and explored the relevance of subtype stratification in a small clinical trial on encapsulated cell biodelivery (ECB) of nerve growth factor (NGF) to the basal forebrain. METHODS: Structural MRI data was collected for 90 amyloid-positive patients and 69 amyloid-negative healthy controls at baseline, 6-, 12-, and 24-month follow-up. The effect of the NGF treatment was investigated in 10 biopsy-verified AD patients with structural MRI data at baseline and at 6- or 12-month follow-up. Patients were classified as typical, limbic-predominant, hippocampal-sparing, or minimal atrophy AD, using a validated visual assessment method. Volumetric analyses were performed using a region-of-interest approach. RESULTS: All AD subtypes showed reduced basal forebrain volume as compared with the healthy controls. The limbic-predominant subtype showed the fastest basal forebrain atrophy rate, whereas the minimal atrophy subtype did not show any significant volume decline over time. Atrophy rates of the hippocampus and precuneus also differed across subtypes. Our preliminary data from the small NGF cohort suggest that the NGF treatment seemed to slow the rate of atrophy in the precuneus and hippocampus in some hippocampal-sparing AD patients and in one typical AD patient. CONCLUSIONS: The cholinergic system is differentially affected in distinct atrophy subtypes of AD. Larger studies in the future should confirm that this differential involvement of the cholinergic system may contribute to subtype-specific response to cholinergic treatment. Our preliminary findings suggest that future clinical trials should target specific subtypes of AD, or at least report treatment effects stratified by subtype. TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT01163825. Registered 14 July 2010.

The combined effect of amyloid-ß and tau biomarkers on brain atrophy in dementia with Lewy bodies.

BACKGROUND: Alzheimer's disease (AD)-related pathology is frequently found in patients with dementia with Lewy bodies (DLB). However, it is unknown how amyloid-ß and tau-related pathologies influence neurodegeneration in DLB. Understanding the mechanisms underlying brain atrophy in DLB can improve our knowledge about disease progression, differential diagnosis, drug development and testing of anti-amyloid and anti-tau therapies in DLB. OBJECTIVES: We aimed at investigating the combined effect of CSF amyloid-ß42, phosphorylated tau and total tau on regional brain atrophy in DLB in the European DLB (E-DLB) cohort. METHODS: 86 probable DLB patients from the E-DLB cohort with CSF and MRI data were included. Random forest was used to analyze the association of CSF biomarkers (predictors) with visual rating scales for medial temporal lobe atrophy (MTA), posterior atrophy (PA) and global cortical atrophy scale-frontal subscale (GCA-F) (outcomes), including age, sex, education and disease duration as extra predictors. RESULTS: DLB patients with abnormal MTA scores had abnormal CSF Aß42, shorter disease duration and older age. DLB patients with abnormal PA scores had abnormal levels of CSF Aß42 and p-tau, older age, lower education and shorter disease duration. Abnormal GCA-F scores were associated with lower education, male sex, and older age, but not with any AD-related CSF biomarker. CONCLUSIONS: This study shows preliminary data on the potential combined effect of amyloid-ß and tau-related pathologies on the integrity of posterior brain cortices in DLB patients, whereas only amyloid-ß seems to be related to MTA. Future availability of α-synuclein biomarkers will help us to understand the effect of α-synuclein and AD-related pathologies on brain integrity in DLB.

The reliability of a deep learning model in clinical out-of-distribution MRI data: A multicohort study.

Deep learning (DL) methods have in recent years yielded impressive results in medical imaging, with the potential to function as clinical aid to radiologists. However, DL models in medical imaging are often trained on public research cohorts with images acquired with a single scanner or with strict protocol harmonization, which is not representative of a clinical setting. The aim of this study was to investigate how well a DL model performs in unseen clinical datasets-collected with different scanners, protocols and disease populations-and whether more heterogeneous training data improves generalization. In total, 3117 MRI scans of brains from multiple dementia research cohorts and memory clinics, that had been visually rated by a neuroradiologist according to Scheltens' scale of medial temporal atrophy (MTA), were included in this study. By training multiple versions of a convolutional neural network on different subsets of this data to predict MTA ratings, we assessed the impact of including images from a wider distribution during training had on performance in external memory clinic data. Our results showed that our model generalized well to datasets acquired with similar protocols as the training data, but substantially worse in clinical cohorts with visibly different tissue contrasts in the images. This implies that future DL studies investigating performance in out-of-distribution (OOD) MRI data need to assess multiple external cohorts for reliable results. Further, by including data from a wider range of scanners and protocols the performance improved in OOD data, which suggests that more heterogeneous training data makes the model generalize better. To conclude, this is the most comprehensive study to date investigating the domain shift in deep learning on MRI data, and we advocate rigorous evaluation of DL models on clinical data prior to being certified for deployment.

AVRA: Automatic visual ratings of atrophy from MRI images using recurrent convolutional neural networks.

Quantifying the degree of atrophy is done clinically by neuroradiologists following established visual rating scales. For these assessments to be reliable the rater requires substantial training and experience, and even then the rating agreement between two radiologists is not perfect. We have developed a model we call AVRA (Automatic Visual Ratings of Atrophy) based on machine learning methods and trained on 2350 visual ratings made by an experienced neuroradiologist. It provides fast and automatic ratings for Scheltens' scale of medial temporal atrophy (MTA), the frontal subscale of Pasquier's Global Cortical Atrophy (GCA-F) scale, and Koedam's scale of Posterior Atrophy (PA). We demonstrate substantial inter-rater agreement between AVRA's and a neuroradiologist ratings with Cohen's weighted kappa values of κw = 0.74/0.72 (MTA left/right), κw = 0.62 (GCA-F) and κw = 0.74 (PA). We conclude that automatic visual ratings of atrophy can potentially have great scientific value, and aim to present AVRA as a freely available toolbox.

Phenotypic variability and neuropsychological findings associated with C9orf72 repeat expansions in a Bulgarian dementia cohort.

BACKGROUND: The GGGGCC repeat expansion in the C9orf72 gene was recently identified as a major cause of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) in several European populations. The objective of this study was to determine the frequency of C9orf72 repeat expansions in a Bulgarian dementia cohort and to delineate the associated clinical features. METHODS AND FINDINGS: PCR-based assessments of the C9orf72 hexanucleotide repeat expansion in all study samples (including 82 FTD, 37 Alzheimer's disease (AD), and 16 other neurodegenerative/dementia disorder cases) were performed. We report the clinical, neuropsychological, and neuroimaging findings obtained for the C9orf72 repeat expansion carriers. Of the 135 cases screened, 3/82 (3.7%) of all FTD cases and 1/37 (2.7%) of all clinical AD cases had a C9orf72 repeat expansion. In this cohort, the C9orf72 pathological expansion was found in clinical diagnoses bridging the FTD, parkinsonism, ALS and AD spectrum. Interestingly, we showed early writing errors without aphasia in two subjects with C9orf72 expansions. CONCLUSIONS: This study represents the first genetic screening for C9orf72 repeat expansions in a Bulgarian dementia cohort. The C9orf72 repeat expansion does not appear to be a common cause of FTD and related disorders. This report confirms the notion that C9orf72 repeat expansions underlie a broad spectrum of neurodegenerative phenotypes. Relatively isolated agraphia in two cases with C9orf72 repeat expansions is a strong motivation to provide detailed and sophisticated oral and written language assessments that can be used to more precisely characterize early cognitive deficits in these heterogeneous conditions.

The contribution of small vessel disease to subtypes of Alzheimer's disease: a study on cerebrospinal fluid and imaging biomarkers.

We investigated whether subtypes of Alzheimer's disease (AD), that is, typical, limbic-predominant, hippocampal-sparing, and minimal atrophy AD, had a specific signature of small vessel disease and neurodegeneration. Four hundred twenty-three clinically diagnosed AD patients were included (161 typical, 121 limbic-predominant, 70 hippocampal-sparing, 71 minimal atrophy). One hundred fifty-six fulfilled a biomarkers-based AD diagnosis. White matter hyperintensities and cerebral microbleeds (CMB) had the highest prevalence in limbic-predominant AD, and the lowest prevalence in minimal atrophy AD. CMB existed evenly in lobar and deep brain areas in limbic-predominant, typical, and hippocampal-sparing AD. In minimal atrophy AD, CMB were mainly located in brain lobar areas. Perivascular spaces in the centrum semiovale were more prevalent in typical AD. Small vessel disease contributed to the prediction of Mini-Mental State Examination. Minimal atrophy AD showed highly pathological levels of cerebrospinal fluid Aß1-42, total tau, and phosphorylated tau, in the absence of overt brain atrophy. Cerebral amyloid angiopathy seems to have a stronger contribution to hippocampal-sparing and minimal atrophy AD, whereas hypertensive arteriopathy may have a stronger contribution to typical and limbic-predominant AD.