We built and validated a deep learning algorithm predicting the individual diagnosis of Alzheimer's disease (AD) and mild cognitive impairment who will convert to AD (c-MCI) based on a single ...cross-sectional brain structural MRI scan. Convolutional neural networks (CNNs) were applied on 3D T1-weighted images from ADNI and subjects recruited at our Institute (407 healthy controls HC, 418 AD, 280 c-MCI, 533 stable MCI s-MCI). CNN performance was tested in distinguishing AD, c-MCI and s-MCI. High levels of accuracy were achieved in all the classifications, with the highest rates achieved in the AD vs HC classification tests using both the ADNI dataset only (99%) and the combined ADNI + non-ADNI dataset (98%). CNNs discriminated c-MCI from s-MCI patients with an accuracy up to 75% and no difference between ADNI and non-ADNI images. CNNs provide a powerful tool for the automatic individual patient diagnosis along the AD continuum. Our method performed well without any prior feature engineering and regardless the variability of imaging protocols and scanners, demonstrating that it is exploitable by not-trained operators and likely to be generalizable to unseen patient data. CNNs may accelerate the adoption of structural MRI in routine practice to help assessment and management of patients.
•CNNs predict AD and MCI with high accuracy based on a single T1-weighted image•CNNs discriminate c-MCI from s-MCI patients with an accuracy up to 75%•CNNs are exploitable by not-trained operators•CNNs are likely to be generalizable to unseen patient data
Early-onset Alzheimer's disease (EOAD) is characterized by young age of onset (< 65 years), severe neurodegeneration, and rapid disease progression, thus differing significantly from typical ...late-onset Alzheimer's disease. Growing evidence suggests a primary role of neuroinflammation in AD pathogenesis. However, the role of microglia activation in EOAD remains a poorly explored field. Investigating microglial activation and its influence on the development of synaptic dysfunction and neuronal loss in EOAD may contribute to the understanding of its pathophysiology and to subject selection in clinical trials. In our study, we aimed to assess the amount of neuroinflammation and neurodegeneration and their relationship in EOAD patients, through positron emission tomography (PET) measures of microglia activation and brain metabolic changes.
We prospectively enrolled 12 EOAD patients, classified according to standard criteria, who underwent standard neurological and neuropsychological evaluation, CSF analysis, brain MRI, and both
F-FDG PET and
C-(R)-PK11195 PET. Healthy controls databases were used for statistical comparison.
F-FDG PET brain metabolism in single subjects and as a group was assessed by an optimized SPM voxel-wise single-subject method.
C-PK11195 PET binding potentials were obtained using reference regions selected with an optimized clustering procedure followed by a parametric analysis. We performed a topographic interaction analysis and correlation analysis in AD-signature metabolic dysfunctional regions and regions of microglia activation. A network connectivity analysis was performed using the interaction regions of hypometabolism and
C-PK11195 PET BP increases.
EOAD patients showed a significant and extended microglia activation, as
C-PK11195 PET binding potential increases, and hypometabolism in typical AD-signature brain regions, i.e., temporo-parietal cortex, with additional variable frontal and occipital hypometabolism in the EOAD variants. There was a spatial concordance in the interaction areas and significant correlations between the two biological changes. The network analysis showed a disruption of frontal connectivity induced by the metabolic/microglia effects.
The severe microglia activation characterizing EOAD and contributing to neurodegeneration may be a marker of rapid disease progression. The coupling between brain glucose hypometabolism and local immune response in AD-signature regions supports their biological interaction.
In this multicentre study in clinical settings, we assessed the accuracy of optimized procedures for FDG-PET brain metabolism and CSF classifications in predicting or excluding the conversion to ...Alzheimer's disease (AD) dementia and non-AD dementias.
We included 80 MCI subjects with neurological and neuropsychological assessments, FDG-PET scan and CSF measures at entry, all with clinical follow-up. FDG-PET data were analysed with a validated voxel-based SPM method. Resulting single-subject SPM maps were classified by five imaging experts according to the disease-specific patterns, as “typical-AD”, “atypical-AD” (i.e. posterior cortical atrophy, asymmetric logopenic AD variant, frontal-AD variant), “non-AD” (i.e. behavioural variant FTD, corticobasal degeneration, semantic variant FTD; dementia with Lewy bodies) or “negative” patterns. To perform the statistical analyses, the individual patterns were grouped either as “AD dementia vs. non-AD dementia (all diseases)” or as “FTD vs. non-FTD (all diseases)”. Aβ42, total and phosphorylated Tau CSF-levels were classified dichotomously, and using the Erlangen Score algorithm. Multivariate logistic models tested the prognostic accuracy of FDG-PET-SPM and CSF dichotomous classifications. Accuracy of Erlangen score and Erlangen Score aided by FDG-PET SPM classification was evaluated.
The multivariate logistic model identified FDG-PET “AD” SPM classification (Expβ = 19.35, 95% C.I. 4.8–77.8, p < 0.001) and CSF Aβ42 (Expβ = 6.5, 95% C.I. 1.64–25.43, p < 0.05) as the best predictors of conversion from MCI to AD dementia. The “FTD” SPM pattern significantly predicted conversion to FTD dementias at follow-up (Expβ = 14, 95% C.I. 3.1–63, p < 0.001). Overall, FDG-PET-SPM classification was the most accurate biomarker, able to correctly differentiate either the MCI subjects who converted to AD or FTD dementias, and those who remained stable or reverted to normal cognition (Expβ = 17.9, 95% C.I. 4.55–70.46, p < 0.001).
Our results support the relevant role of FDG-PET-SPM classification in predicting progression to different dementia conditions in prodromal MCI phase, and in the exclusion of progression, outperforming CSF biomarkers.
•Appropriate biomarkers measures improve early dementia diagnosis in MCI.•FDG-PET-SPM maps and CSF Aβ42 are the best predictors of AD dementia conversion.•FDG-PET-SPM maps accurately predict conversion to different dementia conditions.•A negative FDG-PET-SPM pattern characterizes stable or reverter MCI cases.
Purpose
The aim of this study was to evaluate the supportive role of molecular and structural biomarkers (CSF protein levels, FDG PET and MRI) in the early differential diagnosis of dementia in a ...large sample of patients with neurodegenerative dementia, and in determining the risk of disease progression in subjects with mild cognitive impairment (MCI).
Methods
We evaluated the supportive role of CSF Aβ
42
, t-Tau, p-Tau levels, conventional brain MRI and visual assessment of FDG PET SPM t-maps in the early diagnosis of dementia and the evaluation of MCI progression.
Results
Diagnosis based on molecular biomarkers showed the best fit with the final diagnosis at a long follow-up. FDG PET SPM t-maps had the highest diagnostic accuracy in Alzheimer’s disease and in the differential diagnosis of non-Alzheimer’s disease dementias. The p-tau/Aβ
42
ratio was the only CSF biomarker providing a significant classification rate for Alzheimer’s disease. An Alzheimer’s disease-positive metabolic pattern as shown by FDG PET SPM in MCI was the best predictor of conversion to Alzheimer’s disease.
Conclusion
In this clinical setting, FDG PET SPM t-maps and the p-tau/Aβ
42
ratio improved clinical diagnostic accuracy, supporting the importance of these biomarkers in the emerging diagnostic criteria for Alzheimer’s disease dementia. FDG PET using SPM t-maps had the highest predictive value by identifying hypometabolic patterns in different neurodegenerative dementias and normal brain metabolism in MCI, confirming its additional crucial exclusionary role.
(18)FFDG-PET imaging has been recognized as a crucial diagnostic marker in Mild Cognitive Impairment (MCI), supporting the presence or the exclusion of Alzheimer's Disease (AD) pathology. A clinical ...heterogeneity, however, underlies MCI definition. In this study, we aimed to evaluate the predictive role of single-subject voxel-based maps of (18)FFDG distribution generated through statistical parametric mapping (SPM) in the progression to different dementia subtypes in a sample of 45 MCI. Their scans were compared to a large normal reference dataset developed and validated for comparison at single-subject level. Additionally, Aβ42 and Tau CSF values were available in 34 MCI subjects. Clinical follow-up (mean 28.5 ± 7.8 months) assessed subsequent progression to AD or non-AD dementias. The SPM analysis showed: 1) normal brain metabolism in 14 MCI cases, none of them progressing to dementia; 2) the typical temporo-parietal pattern suggestive for prodromal AD in 15 cases, 11 of them progressing to AD; 3) brain hypometabolism suggestive of frontotemporal lobar degeneration (FTLD) subtypes in 7 and dementia with Lewy bodies (DLB) in 2 subjects (all fulfilled FTLD or DLB clinical criteria at follow-up); and 4) 7 MCI cases showed a selective unilateral or bilateral temporo-medial hypometabolism without the typical AD pattern, and they all remained stable. In our sample, objective voxel-based analysis of (18)FFDG-PET scans showed high predictive prognostic value, by identifying either normal brain metabolism or hypometabolic patterns suggestive of different underlying pathologies, as confirmed by progression at follow-up. These data support the potential usefulness of this SPM (18)FFDG PET analysis in the early dementia diagnosis and for improving subject selection in clinical trials based on MCI definition.
•Theta density increase is the earliest and most sensitive EEG marker of AD pathology.•Alpha2 density progressively decreases following the progression of AD pathology.•EEG graph analysis of ADD ...patients shows network derangement at theta and alpha2 band.•EEG/fMRI integration model empowered EEG diagnostic performances.
We evaluated the value of resting-state EEG source biomarkers to characterize mild cognitive impairment (MCI) subjects with an Alzheimer’s disease (AD)-like cerebrospinal fluid (CSF) profile and to track neurodegeneration throughout the AD continuum. We further applied a resting-state functional MRI (fMRI)-driven model of source reconstruction and tested its advantage in terms of AD diagnostic accuracy.
Thirty-nine consecutive patients with AD dementia (ADD), 86 amnestic MCI, and 33 healthy subjects enter the EEG study. All ADD subjects, 37 out of 86 MCI patients and a distinct group of 53 healthy controls further entered the fMRI study. MCI subjects were divided according to the CSF phosphorylated tau/β amyloid-42 ratio (MCIpos: ≥ 0.13, MCIneg: < 0.13). Using Exact low-resolution brain electromagnetic tomography (eLORETA), EEG lobar current densities were estimated at fixed frequencies and analyzed. To combine the two imaging techniques, networks mostly affected by AD pathology were identified using Independent Component Analysis applied to fMRI data of ADD subjects. Current density EEG analysis within ICA-based networks at selected frequency bands was performed. Afterwards, graph analysis was applied to EEG and fMRI data at ICA-based network level.
ADD patients showed a widespread slowing of spectral density. At a lobar level, MCIpos subjects showed a widespread higher theta density than MCIneg and healthy subjects; a lower beta2 density than healthy subjects was also found in parietal and occipital lobes. Evaluating EEG sources within the ICA-based networks, alpha2 band distinguished MCIpos from MCIneg, ADD and healthy subjects with good accuracy. Graph analysis on EEG data showed an alteration of connectome configuration at theta frequency in ADD and MCIpos patients and a progressive disruption of connectivity at alpha2 frequency throughout the AD continuum.
Theta frequency is the earliest and most sensitive EEG marker of AD pathology. Furthermore, EEG/fMRI integration highlighted the role of alpha2 band as potential neurodegeneration biomarker.
Focal repetitive transcranial magnetic stimulation (rTMS) has been applied to improve cognition in Alzheimer's disease (AD) with conflicting results. We applied rTMS in AD in a pilot ...placebo-controlled study using the H2-coil. H-coils are suitable for targeting wider neuronal structures compared with standard focal coils, in particular the H2-coil stimulates simultaneously the frontal-parietal-temporal lobes bilaterally. Thirty patients (mean age 70.9 year, SD 8.1; mean MMSE score 16.9, SD 5.5) were randomized to sham or real 10 Hz rTMS stimulation with the H2-coil. Each patient underwent 3 sessions/week for 4 weeks, followed by 4 weeks with maintenance treatment (1 session/week). Primary outcome was improvement of ADAS-cog at 4 and 8 weeks compared with baseline. A trend toward an improved ADAS-cog score over time was observed for patients undergoing real rTMS, with actively treated patients experiencing a mean decrease of -1.01 points at the ADAS-Cog scale score per time point (95% CIs -0.02 to -3.13,
< 0.04). This trend was no longer evident 2 months after the end of treatment. Real rTMS showed no significant effect on MMSE and BDI changes over time. These preliminary findings suggest that rTMS with H-coil is feasible and safe in patients with probable AD and might provide beneficial, even though transient, effects on cognition. This study prompts larger studies in the early stages of AD, combining rTMS and cognitive rehabilitation.
www.ClinicalTrials.gov, identifier: NCT04562506.
Mild cognitive impairment (MCI) is a heterogeneous syndrome resulting from Alzheimer's disease (AD) as well as to non-AD and non-neurodegenerative conditions. A subset of patients with amnestic MCI ...(aMCI) present with an unusually long-lasting course, a slow rate of clinical neuropsychological progression, and evidence of focal involvement of medial temporal lobe structures. In the present study, we explored positron emission tomography (PET) and cerebrospinal fluid (CSF) biomarkers in a sample of subjects with aMCI with such clinical features in order to provide in vivo evidence to improve disease characterisation in this subgroup.
Thirty consecutive subjects with aMCI who had long-lasting memory impairment (more than 4 years from symptom onset) and a very slow rate of cognitive progression were included. All subjects underwent fluorodeoxyglucose-positron emission tomography (FDG-PET) metabolic imaging. A measure of cerebral amyloid load, by PET and/or CSF, was obtained in 26 of 30 subjects. The mean clinical follow-up was 58.3 ± 10.1 months.
No patient progressed to dementia during the follow-up. The typical AD FDG-PET pattern of temporoparietal hypometabolism was not present in any of the subjects. In contrast, a selective medial temporal lobe hypometabolism was present in all subjects, with an extension to frontolimbic regions in some subjects. PET imaging showed absent or low amyloid load in the majority of samples. The values were well below those reported in prodromal AD, and they were slightly elevated in only two subjects, consistent with the CSF β-amyloid (1-42) protein values. Notably, no amyloid load was present in the hippocampal structures.
FDG-PET and amyloid-PET together with CSF findings questioned AD pathology as a unique neuropathological substrate in this aMCI subgroup with long-lasting disease course. The possibility of alternative pathological conditions, such as argyrophilic grain disease, primary age-related tauopathy or age-related TDP-43 proteinopathy, known to spread throughout the medial temporal lobe and limbic system structures should be considered in these patients with MCI.
Imaging and histopathological studies have demonstrated that structural changes of the retina affect subjects with Alzheimer's disease (AD) or mild cognitive impairment (MCI). The aim of this study ...was to quantitatively investigate the retinal vessels in these disorders, using dynamic vessel analyzer (DVA) and optical coherence tomography angiography (OCTA) analysis. Twelve subjects with AD, 12 subjects with MCI, and 32 gender- and age-matched controls were prospectively enrolled. Mean ± SD age was 72.9 ± 7.2 years in the AD group, 76.3 ± 6.9 years in the MCI group, and 71.6 ± 5.9 years in the control group (p = 0.104). In the DVA dynamic analysis, the arterial dilation was decreased in the AD group (0.77 ± 2.06%), in the comparison with the control group (3.53 ± 1.25%, p = 0.002). The reaction amplitude was decreased both in AD (0.21 ± 1.80%, <0.0001) and MCI (2.29 ± 1.81%, p = 0.048) subjects, compared with controls (3.86 ± 1.94%). OCTA variables did not differ among groups. In the Pearson correlation analysis, amyloid β level in the cerebrospinal fluid was directly correlated with the arterial dilation (R = 0.441, p = 0.040) and reaction amplitude (R = 0.580, p = 0.005). This study demonstrate that Alzheimer's and MCI subjects are characterized by a significant impairment of the retinal neurovascular coupling. This impairment is inversely correlated with the level of amyloid β in the cerebrospinal fluid.
Dementia with Lewy Bodies (DLB) is characterized by a prominent deficit in visuospatial abilities. Visuospatial impairment is also detectable in the course of Alzheimer's dementia (AD). However, ...visuospatial impairment presents some differences in these two conditions, suggesting pathological involvement of distinct brain circuits. Recent studies applied a new method to score the Mini Mental State Examination (MMSE) pentagon copy subtest, namely the Qualitative Scoring Pentagon Test (QSPT), which is a sensitive measure of visuospatial abilities. Using 18Ffluorodeoxy-glucose positron emission tomography (FDG-PET), we assessed the relationship between in vivo brain metabolic dysfunction and visuospatial deficits, in terms of QSPT total value, in DLB and AD.
Sixty Patients were diagnosed as DLB (n = 35) and AD (n = 25) dementia according with the standard research diagnostic criteria. Each patient underwent a FDG-PET scan as support for the final diagnosis. Patients underwent an extended neuropsychological evaluation, including MMSE, language, memory, executive functions and visuospatial abilities tests. The MMSE QSPT scoring was calculated following the methods by Caffarra et al. (2013). Offline voxel-wise correlation analysis between QSPT total scores and FDG-PET brain metabolism was then performed, correcting for MMSE, sex and disease duration.
Both groups presented reduced visuospatial performances, as assessed by QSPT scores. DLB compared to AD showed a statistically significant difference in QSPT rotation parameter (p = 0.022). In DLB, worse performance at QSPT total score, i.e. more severe visuospatial impairment, correlated with brain occipital hypometabolism (i.e. lateral occipital cortex, calcarine cortex, fusiform and lingual gyri). In AD, worse performance at QSPT total score correlated with brain hypometabolism in the right parietal cortex (i.e. superior and inferior parietal cortex and angular gyrus).
These findings reveal that visuospatial deficits may derive from distinct brain alterations in AD and DLB. We propose that the inabilities to perform correctly the QSPT task is related to altered visuoperceptual process in DLB, and visuospatial process in AD. This is consistent with our results showing hypometabolism in brain system related to visuoperceptual processing, namely the occipital cortex in DLB, and visuospatial processing, namely parietal cortex in AD.
•Qualitative Scoring Pentagon Test is useful for visuospatial deficits assessment.•Visuospatial deficits commonly occur in both Lewy body and Alzheimer's dementia.•Visuospatial deficits correlate with occipital hypometabolism in Lewy body dementia.•Visuospatial deficits correlate with parietal hypometabolism in Alzheimer's dementia.•Different neurocircuits underlie visuospatial deficits in the two conditions.