Comparing two artificial intelligence software packages for normative brain volumetry in memory clinic imaging

Purpose To compare two artificial intelligence software packages performing normative brain volumetry and explore whether they could differently impact dementia diagnostics in a clinical context. Methods Sixty patients (20 Alzheimer’s disease, 20 frontotemporal dementia, 20 mild cognitive impairment) and 20 controls were included retrospectively. One MRI per subject was processed by software packages from two proprietary manufacturers, producing two quantitative reports per subject. Two neuroradiologists assigned forced-choice diagnoses using only the normative volumetry data in these reports. They classified the volumetric profile as “normal,” or “abnormal”, and if “abnormal,” they specified the most likely dementia subtype. Differences between the packages’ clinical impact were assessed by comparing (1) agreement between diagnoses based on software output; (2) diagnostic accuracy, sensitivity, and specificity; and (3) diagnostic confidence. Quantitative outputs were also compared to provide context to any diagnostic differences. Results Diagnostic agreement between packages was moderate, for distinguishing normal and abnormal volumetry (K = .41–.43) and for specific diagnoses (K = .36–.38). However, each package yielded high inter-observer agreement when distinguishing normal and abnormal profiles (K = .73–.82). Accuracy, sensitivity, and specificity were not different between packages. Diagnostic confidence was different between packages for one rater. Whole brain intracranial volume output differed between software packages (10.73%, p < .001), and normative regional data interpreted for diagnosis correlated weakly to moderately (rs = .12–.80). Conclusion Different artificial intelligence software packages for quantitative normative assessment of brain MRI can produce distinct effects at the level of clinical interpretation. Clinics should not assume that different packages are interchangeable, thus recommending internal evaluation of packages before adoption.

Diagnostic Blood Biomarkers in Alzheimer’s Disease

Potential biomarkers for Alzheimer’s disease (AD) include amyloid β1–42 (Aβ1–42), t-Tau, p-Tau181, neurofilament light chain (NFL), and neuroimaging biomarkers. Their combined use is useful for diagnosing and monitoring the progress of AD. Therefore, further development of a combination of these biomarkers is essential. We investigated whether plasma NFL/Aβ1–42 can serve as a plasma-based primary screening biomarker reflecting brain neurodegeneration and amyloid pathology in AD for monitoring disease progression and early diagnosis. We measured the NFL and Aβ1–42 concentrations in the CSF and plasma samples and performed correlation analysis to evaluate the utility of these biomarkers in the early diagnosis and monitoring of AD spectrum disease progression. Pearson’s correlation analysis was used to analyse the associations between the fluid biomarkers and neuroimaging data. The study included 136 participants, classified into five groups: 28 cognitively normal individuals, 23 patients with preclinical AD, 22 amyloid-negative patients with amnestic mild cognitive impairment, 32 patients with prodromal AD, and 31 patients with AD dementia. With disease progression, the NFL concentrations increased and Aβ1–42 concentrations decreased. The plasma and CSF NFL/Aβ1–42 were strongly correlated (r = 0.558). Plasma NFL/Aβ1–42 was strongly correlated with hippocampal volume/intracranial volume (r = 0.409). In early AD, plasma NFL/Aβ1–42 was associated with higher diagnostic accuracy than the individual biomarkers. Moreover, in preclinical AD, plasma NFL/Aβ1–42 changed more rapidly than the CSF t-Tau or p-Tau181 concentrations. Our findings highlight the utility of plasma NFL/Aβ1–42 as a non-invasive plasma-based biomarker for early diagnosis and monitoring of AD spectrum disease progression.

Brain ageing in schizophrenia: evidence from 26 international cohorts via the ENIGMA Schizophrenia consortium

Schizophrenia (SZ) is associated with an increased risk of life-long cognitive impairments, age-related chronic disease, and premature mortality. We investigated evidence for advanced brain ageing in adult SZ patients, and whether this was associated with clinical characteristics in a prospective meta-analytic study conducted by the ENIGMA Schizophrenia Working Group. The study included data from 26 cohorts worldwide, with a total of 2803 SZ patients (mean age 34.2 years; range 18-72 years; 67% male) and 2598 healthy controls (mean age 33.8 years, range 18-73 years, 55% male). Brain-predicted age was individually estimated using a model trained on independent data based on 68 measures of cortical thickness and surface area, 7 subcortical volumes, lateral ventricular volumes and total intracranial volume, all derived from T1-weighted brain magnetic resonance imaging (MRI) scans. Deviations from a healthy brain ageing trajectory were assessed by the difference between brain-predicted age and chronological age (brain-predicted age difference [brain-PAD]). On average, SZ patients showed a higher brain-PAD of +3.64 years (95% CI: 3.01, 4.26; I2 = 55.28%) compared to controls, after adjusting for age and sex (Cohen's d = 0.50). Among SZ patients, brain-PAD was not associated with specific clinical characteristics (age of onset, duration of illness, symptom severity, or antipsychotic use and dose). This large-scale collaborative study suggests advanced structural brain ageing in SZ. Longitudinal studies of SZ and a range of mental and somatic health outcomes will help to further evaluate the clinical implications of increased brain-PAD and its ability to be influenced by interventions.

Neuropsychological and neuroanatomical features of patients with behavioral variant Alzheimer's disease (AD): a comparison to behavioral variant frontotemporal dementia and amnestic AD groups

An understudied non-amnestic variant of Alzheimer's disease (AD), behavioral variant AD (bvAD) is associated with progressive personality, behavior, or executive dysfunction and frontal atrophy. This study characterizes the neuropsychological and neuroanatomical features associated with bvAD by comparing it to behavioral variant frontotemporal dementia (bvFTD), amnestic AD (aAD), and subjects with normal cognition. Subjects included 16 bvAD, 67 bvFTD, and 18 aAD patients, and 26 healthy controls. Compared to bvFTD, bvAD showed more significant visuospatial impairments (Rey Figure copy and recall), more irritability (Neuropsychological Inventory), and equivalent verbal memory (Philadelphia Verbal Learning Test). Compared to aAD, bvAD indicated more executive dysfunction (F-letter fluency) and better visuospatial performance. Neuroimaging analysis found that bvAD showed cortical thinning relative to bvFTD posteriorly in left temporal-occipital regions; bvFTD had cortical thinning relative to bvAD in left inferior frontal cortex. bvAD had cortical thinning relative to aAD in prefrontal and anterior temporal regions. All patient groups had lower volumes than controls in both anterior and posterior hippocampus. However, bvAD patients had higher average volume than aAD patients in posterior hippocampus and higher volume than bvFTD patients in anterior hippocampus after adjustment for age and intracranial volume. Findings demonstrated that underlying pathology mediates disease presentation in bvAD and bvFTD.

Beyond “Sex Prediction”: Estimating and Interpreting Multivariate Sex Differences and Similarities in the Brain

Previous studies have shown that machine-learning (ML) algorithms can “predict” sex based on brain anatomical/ functional features. The high classification accuracy achieved by ML algorithms is often interpreted as revealing large differences between the brains of males and females and as confirming the existence of “male/female brains”. However, classification and estimation are quite different concepts, and using classification metrics as surrogate estimates of between-group differences results in major statistical and interpretative distortions. The present study illustrates these distortions and provides a novel and detailed assessment of multivariate sex differences in gray matter volume (GMVOL) that does not rely on classification metrics. Moreover, modeling and clustering techniques and analyses of similarities (ANOSIM) were used to identify the brain areas that contribute the most to these multivariate differences, and to empirically assess whether they assemble into two sex-typical profiles. Results revealed that multivariate sex differences in GMVOL: 1) are “large” if not adjusted for total intracranial volume (TIV) variation, but “small” when controlling for this variable; 2) differ in size between individuals and also depends on the ML algorithm used for their calculation 3) do not stem from two sex-typical profiles, and so describing them in terms of “male/female brains” is misleading.

Volume of Amygdala Subregions and Clinical Manifestations in Patients With First-Episode, Drug-Naïve Major Depression

We examined amygdala subregion volumes in patients with a first episode of major depression (MD) and in healthy subjects. Covariate-adjusted linear regression was performed to compare the MD and healthy groups, and adjustments for age, gender, and total estimated intracranial volume showed no differences in amygdala subregion volumes between the healthy and MD groups. Within the MD group, we examined the association between amygdala subregion volume and the 17-item Hamilton Rating Scale for Depression (HAMD) score and the HAMD subscale score, and found no association in the left amygdala. In the right amygdala, however, there was an inverse linear association between the HAMD total and the HAMD core and lateral nucleus and anterior-amygdaloid-regions. Furthermore, an inverse linear association was seen between the HAMD psychic and the lateral nucleus, anterior-amygdaloid-regions, transition, and whole amygdala. The findings of this study suggest that the severity of MD and some symptoms of MD are associated with right amygdala volume. There have been few reports on the relationship between MD and amygdala subregional volume, and further research is needed to accumulate more data for further validation.

Association of β-Amyloid and Basal Forebrain With Cortical Thickness and Cognition in Alzheimer and Lewy Body Disease Spectra

Objective:Cholinergic degeneration and β-amyloid contribute to brain atrophy and cognitive dysfunction in Alzheimer’s disease (AD) and Lewy body disease (LBD), but their relationship has not been comparatively evaluated.Methods:In this cross-sectional study, we recruited 28 normal controls (NC), 55 patients with AD mild cognitive impairment (MCI), 34 patients with AD dementia, 28 patients with LBD MCI, and 51 patients with LBD dementia. The subjects underwent cognitive evaluation, brain magnetic resonance imaging to measure the basal forebrain (BF) volume and global cortical thickness (CTh), and 18F-Florbetaben (FBB) positron emission tomography to measure the standardized uptake value ratio (SUVR). Using general linear models and path analyses, the association of FBB-SUVR and BF volume with the CTh and/or cognitive dysfunction were evaluated in AD spectrum (AD and NC) and LBD spectrum (LBD and NC), respectively. Covariates included age, sex, education, deep and periventricular white matter hyperintensities, intracranial volume, hypertension, diabetes mellitus, and hyperlipidemia.Results:BF volume mediated the association between FBB-SUVR and CTh both in AD and LBD spectra, while FBB-SUVR was associated with CTh independently of BF volume only in LBD spectrum. Significant correlation between voxel-wise FBB-SUVR and CTh was observed only in LBD group. FBB-SUVR was independently associated with widespread cognitive dysfunction both in AD and LBD spectra, especially in the memory domain [standardized beta (B) for AD spectrum = -0.60, B for LBD spectrum = -0.33]. In AD spectrum, BF volume was associated with memory dysfunction (B = 0.18), and CTh was associated with language (B = 0.21) and executive (B = 0.23) dysfunction. In LBD spectrum, however, BF volume and CTh were independently associated with widespread cognitive dysfunction.Conclusions:There is a common β-amyloid-related degenerative mechanism with or without the mediation of BF in AD and LBD spectra, while the association of BF atrophy with cognitive dysfunction is more profound and there is localized β-amyloid-cortical atrophy interaction in LBD spectrum.

Linking Vestibular Function and Subcortical Gray Matter Volume Changes in a Longitudinal Study of Aging Adults

Emerging evidence suggests a relationship between impairments of the vestibular (inner ear balance) system and alterations in the function and the structure of the central nervous system (CNS) in older adults. However, it is unclear whether age-related vestibular loss is associated with volume loss in brain regions known to receive vestibular input. To address this gap, we investigated the association between vestibular function and the volumes of four structures that process vestibular information (the hippocampus, entorhinal cortex, thalamus, and basal ganglia) in a longitudinal study of 97 healthy, older participants from the Baltimore Longitudinal Study of Aging. Vestibular testing included cervical vestibular-evoked myogenic potentials (cVEMP) to measure saccular function, ocular VEMP (oVEMP) to measure utricular function, and video head impulse tests to measure the horizontal semicircular canal vestibulo-ocular reflex (VOR). Participants in the sample had vestibular and brain MRI data for a total of one (18.6%), two (49.5%), and three (32.0%) visits. Linear mixed-effects regression was used to model regional volume over time as a function of vestibular physiological function, correcting for age, sex, intracranial volume, and intersubject random variation in the baseline levels and rates of change of volume over time. We found that poorer saccular function, characterized by lower cVEMP amplitude, is associated with reduced bilateral volumes of the basal ganglia and thalamus at each time point, demonstrated by a 0.0714 cm3 ± 0.0344 (unadjusted p = 0.038; 95% CI: 0.00397–0.139) lower bilateral-mean volume of the basal ganglia and a 0.0440 cm3 ± 0.0221 (unadjusted p = 0.046; 95% CI: 0.000727–0.0873) lower bilateral-mean volume of the thalamus for each 1-unit lower cVEMP amplitude. We also found a relationship between a lower mean VOR gain and lower left hippocampal volume (β = 0.121, unadjusted p = 0.018, 95% CI: 0.0212–0.222). There were no significant associations between volume and oVEMP. These findings provide insight into the specific brain structures that undergo atrophy in the context of age-related loss of peripheral vestibular function.

Inverse association between the anticholinergic burden and hippocampus volume in a population-based cohort across the entire adult age range

Many medications of different indications have a relevant anticholinergic activity. The anticholinergic burden of medication has been shown to have significant effects on the cognition and the risk for cognitive impairment and dementia particularly in older patients. So far, most of the studies used data from geriatric patients and the effect of the anticholinergic burden on brain structures is still unexplored. Our study aimed to analyze possible associations of hippocampus and cholinergic basal forebrain volumes as vulnerable brain structures for the development of dementia and the anticholinergic burden in a population-based cohort of non-demented participants spanning the adult age range from 21 to 80 years. We analyzed associations between medication-related anticholinergic burden and structural MRI volumes from participants (n = 3087, 52.2% female) of the population-based “Study of Health in Pomerania” (SHIP). Anticholinergic burden was obtained from the current medication plan using the Anticholinergic Burden Scale (ACB). All analyses were adjusted for age, sex, education, and total intracranial volume. We found statistically significant associations between the ACB and the left and right hippocampus volume but not for the basal forebrain cholinergic system. Complementary voxel-based analysis across all participants revealed FWE-corrected (p = < 0.05) clusters in the temporo-parietal regions reaching into frontal areas, showing reduced volumes with higher ACB scores. We identified an association between anticholinergic burden of medication on hippocampal volume suggesting a potential inverse effect of such medication. This association highlights the importance of a careful prescription of medication with anticholinergic activity at any adult age.

Intracranial compartment volumes and their distribution among clinically diagnosed patients with normal pressure hydrocephalus and brain atrophy in a Nigerian population

Normal pressure hydrocephalus is a form of non-obstructive hydrocephalus characterized by the triad of symptoms. These symptoms are typical or atypical. Because of its potential reversibility, researchers have paid attention to its diagnosis and in differentiating it with other forms of dementia such as brain atrophy. To determine the intracranial compartment volumes and their differences among patients with NPH and Brain Atrophy (BA). This was a cross-sectional study involving consenting patients diagnosed with NPH and BA who were referred for routine brain CT. Medical conditions known to influence intracranial volume were excluded. Age -matched normal control were clinically and radiologically confirmednormal and were also recruited. Intracranial volumes and CSF distribution determination was based on Cavalieri test point computation principle. Test point summation was on a locally developed software for this purpose. Overall, the values of BA recorded higher mean values for most of the intracranial compartment volumes. Intraventricular volume was consistently higher in NPH groups. Statistical difference exists among intraventricular and total intracranial CSF volume across patients with BA, NPH and control. A post hoc test revealed control-NPH comparison across these variables. Control-BA comparison was evident in total intracranial CSF volume only. Higher mean values of intracranialcompartment volumes were observed in patients with BA than among patients with NPH and their control. Anthropometric indices did not show any difference between patients with BA, NPH and their control. These indices can be used as basis in objectively differentiating NPH from BA.