Machine Learning-based Individual Assessment of Cortical Atrophy Pattern in Alzheimer's Disease Spectrum: Development of the Classifier and Longitudinal Evaluation

Volume and Permeability of White Matter Hyperintensity on Cognition: A DCE Imaging Study of an Older Cohort With and Without Cognitive Impairment

BACKGROUND

The impact of blood-brain barrier (BBB) leakage on white matter hyperintensity (WMH) subtypes (location) and its association with clinical factors and cognition remains unclear.

PURPOSE

To investigate the relationship between WMH volume, permeability, clinical factors, and cognition in older individuals across the cognitive spectrum.

STUDY TYPE

Prospective, cross-sectional.

SUBJECTS

A total of 193 older adults with/without cognitive impairment; 128 females; mean age 70.1 years (standard deviation 6.8).

FIELD STRENGTH/SEQUENCE

3 T, GE Dynamic contrast-enhanced, three-dimensional (3D) Magnetization-prepared rapid gradient-echo (MPRAGE T1WI), 3D fluid-attenuated inversion recovery (FLAIR).

ASSESSMENT

Periventricular WMH (PWMH), deep WMH (DWMH), and normal-appearing white matter (NAWM) were segmented using FMRIB automatic segmentation tool algorithms on 3D FLAIR. Hippocampal volume and cortex volume were segmented on 3D T1WI. BBB permeability (Ktrans) and blood plasma volume (Vp) were determined using the Patlak model. Vascular risk factors and cognition were assessed.

STATISTICAL TESTS

Univariate and multivariate analyses were performed to identify factors associated with WMH permeability. Logistic regression analysis assessed the association between WMH imaging features and cognition, adjusting for age, sex, apolipoprotein E4 status, education, and brain volumes. A P-value <0.05 was considered significant.

RESULTS

PWMH exhibited higher Ktrans (0.598 ± 0.509 × 10-3 minute-1) compared to DWMH (0.496 ± 0.478 × 10-3 minute-1) and NAWM (0.476 ± 0.398 × 10-3 minute-1). Smaller PWMH volume and cardiovascular disease (CVD) history were significantly associated with higher Ktrans in PWMH. In DWMH, higher Ktrans were associated with CVD history and cortical volume. In NAWM, it was linked to CVD history and dyslipidemia. Larger PWMH volume (odds ratio [OR] 1.106, confidence interval [CI]: 1.021-1.197) and smaller hippocampal volume (OR 0.069; CI: 0.019-0.253) were independently linked to worse global cognition after covariate adjustment.

DATA CONCLUSION

Elevated BBB leakage in PWMH was associated with lower PWMH volume and prior CVD history. Notably, PWMH volume, rather than permeability, was correlated with cognitive decline, suggesting that BBB leakage in WMH may be a consequence of CVD rather than indicate disease progression.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY: Stage 3.

Deep learning-based quantification of brain atrophy using 2D T1-weighted MRI for Alzheimer's disease classification

Background

Determining brain atrophy is crucial for the diagnosis of neurodegenerative diseases. Despite detailed brain atrophy assessments using three-dimensional (3D) T1-weighted magnetic resonance imaging, their practical utility is limited by cost and time. This study introduces deep learning algorithms for quantifying brain atrophy using a more accessible two-dimensional (2D) T1, aiming to achieve cost-effective differentiation of dementia of the Alzheimer's type (DAT) from cognitively unimpaired (CU), while maintaining or exceeding the performance obtained with T1-3D individuals and to accurately predict AD-specific atrophy similarity and atrophic changes [W-scores and Brain Age Index (BAI)].

Methods

Involving 924 participants (478 CU and 446 DAT), our deep learning models were trained on cerebrospinal fluid (CSF) volumes from 2D T1 images and compared with 3D T1 images. The performance of the models in differentiating DAT from CU was assessed using receiver operating characteristic analysis. Pearson's correlation analyses were used to evaluate the relations between 3D T1 and 2D T1 measurements of cortical thickness and CSF volumes, AD-specific atrophy similarity, W-scores, and BAIs.

Results

Our deep learning models demonstrated strong correlations between 2D and 3D T1-derived CSF volumes, with correlation coefficients r ranging from 0.805 to 0.971. The algorithms based on 2D T1 accurately distinguished DAT from CU with high accuracy (area under the curve values of 0.873), which were comparable to those of algorithms based on 3D T1. Algorithms based on 2D T1 image-derived CSF volumes showed high correlations in AD-specific atrophy similarity (r = 0.915), W-scores for brain atrophy (0.732 ≤ r ≤ 0.976), and BAIs (r = 0.821) compared with those based on 3D T1 images.

Conclusion

Deep learning-based analysis of 2D T1 images is a feasible and accurate alternative for assessing brain atrophy, offering diagnostic precision comparable to that of 3D T1 imaging. This approach offers the advantage of the availability of T1-2D imaging, as well as reduced time and cost, while maintaining diagnostic precision comparable to T1-3D.

A Two-Year Observational Study to Evaluate Conversion Rates from High- and Low-Risk Patients with Amnestic Mild Cognitive Impairment to Probable Alzheimer's Disease in a Real-World Setting

Background

Predicting conversion to probable Alzheimer&s disease (AD) from amnestic mild cognitive impairment (aMCI) is difficult but important. A nomogram was developed previously for determining the risk of 3-year probable AD conversion in aMCI.

Objective

To compare the probable AD conversion rates with cognitive and neurodegenerative changes for 2 years from high- and low risk aMCI groups classified using the nomogram.

Methods

This prospective, multicenter, observational study was conducted in Korea. A total of patients were classified as high- or low-risk aMCI according to the nomogram and followed-up for 2 years to compare the annual conversion rate to probable AD and brain structure changes between the two groups.

Results

In total, 176 (high-risk, 85; low-risk, 91) and 160 (high-risk, 77; low-risk, 83) patients completed the 1-year and 2-year follow-up, respectively. The probable AD conversion rate was significantly higher in the high-risk (Year 1, 28.9%; Year 2, 46.1%) versus low-risk group (Year 1, 0.0%; Year 2, 4.9%, both p < 0.0001). Mean changes from baseline in Seoul Neuropsychological Screening Battery-Dementia Version, Clinical Dementia Rating-Sum of Box, and Korean version of the Instrumental Activities of Daily Living scores and cortical atrophy index at Years 1 and 2 were significantly greater in the high-risk group (p < 0.0001).

Conclusions

The high-risk aMCI group, as determined by the nomogram, had a higher conversion rate to probable AD and faster cognitive decline and neurodegeneration change than the low-risk group. These real-world results have clinical implications that help clinicians in accurately predicting patient outcomes and facilitating early decision-making.Trial Registration: ClinicalTrials.gov (NCT03448445).

Deep learning model for individualized trajectory prediction of clinical outcomes in mild cognitive impairment

Objectives

Accurately predicting when patients with mild cognitive impairment (MCI) will progress to dementia is a formidable challenge. This work aims to develop a predictive deep learning model to accurately predict future cognitive decline and magnetic resonance imaging (MRI) marker changes over time at the individual level for patients with MCI.

Methods

We recruited 657 amnestic patients with MCI from the Samsung Medical Center who underwent cognitive tests, brain MRI scans, and amyloid-β (Aβ) positron emission tomography (PET) scans. We devised a novel deep learning architecture by leveraging an attention mechanism in a recurrent neural network. We trained a predictive model by inputting age, gender, education, apolipoprotein E genotype, neuropsychological test scores, and brain MRI and amyloid PET features. Cognitive outcomes and MRI features of an MCI subject were predicted using the proposed network.

Results

The proposed predictive model demonstrated good prediction performance (AUC = 0.814 ± 0.035) in five-fold cross-validation, along with reliable prediction in cognitive decline and MRI markers over time. Faster cognitive decline and brain atrophy in larger regions were forecasted in patients with Aβ (+) than with Aβ (-).

Conclusion

The proposed method provides effective and accurate means for predicting the progression of individuals within a specific period. This model could assist clinicians in identifying subjects at a higher risk of rapid cognitive decline by predicting future cognitive decline and MRI marker changes over time for patients with MCI. Future studies should validate and refine the proposed predictive model further to improve clinical decision-making.

Deep learning to predict rapid progression of Alzheimer's disease from pooled clinical trials: A retrospective study

The rate of progression of Alzheimer's disease (AD) differs dramatically between patients. Identifying the most is critical because when their numbers differ between treated and control groups, it distorts the outcome, making it impossible to tell whether the treatment was beneficial. Much recent effort, then, has gone into identifying RPs. We pooled de-identified placebo-arm data of three randomized controlled trials (RCTs), EXPEDITION, EXPEDITION 2, and EXPEDITION 3, provided by Eli Lilly and Company. After processing, the data included 1603 mild-to-moderate AD patients with 80 weeks of longitudinal observations on neurocognitive health, brain volumes, and amyloid-beta (Aβ) levels. RPs were defined by changes in four neurocognitive/functional health measures. We built deep learning models using recurrent neural networks with attention mechanisms to predict RPs by week 80 based on varying observation periods from baseline (e.g., 12, 28 weeks). Feature importance scores for RP prediction were computed and temporal feature trajectories were compared between RPs and non-RPs. Our evaluation and analysis focused on models trained with 28 weeks of observation. The models achieved robust internal validation area under the receiver operating characteristic (AUROCs) ranging from 0.80 (95% CI 0.79-0.82) to 0.82 (0.81-0.83), and the area under the precision-recall curve (AUPRCs) from 0.34 (0.32-0.36) to 0.46 (0.44-0.49). External validation AUROCs ranged from 0.75 (0.70-0.81) to 0.83 (0.82-0.84) and AUPRCs from 0.27 (0.25-0.29) to 0.45 (0.43-0.48). Aβ plasma levels, regional brain volumetry, and neurocognitive health emerged as important factors for the model prediction. In addition, the trajectories were stratified between predicted RPs and non-RPs based on factors such as ventricular volumes and neurocognitive domains. Our findings will greatly aid clinical trialists in designing tests for new medications, representing a key step toward identifying effective new AD therapies.

An overview of clinical machine learning applications in neurology

Machine learning techniques for clinical applications are evolving, and the potential impact this will have on clinical neurology is important to recognize. By providing a broad overview on this growing paradigm of clinical tools, this article aims to help healthcare professionals in neurology prepare to navigate both the opportunities and challenges brought on through continued advancements in machine learning. This narrative review first elaborates on how machine learning models are organized and implemented. Machine learning tools are then classified by clinical application, with examples of uses within neurology described in more detail. Finally, this article addresses limitations and considerations regarding clinical machine learning applications in neurology.

Polycyclic aromatic hydrocarbons and changes in brain cortical thickness and an Alzheimer's disease-specific marker for cortical atrophy in adults: A longitudinal neuroimaging study of the EPINEF cohort

Although several epidemiological studies have suggested that exposure to polycyclic aromatic hydrocarbons (PAHs) may induce brain atrophy, no longitudinal study has investigated the effect of PAH exposure on brain structural changes. This study examined the longitudinal associations between urinary PAH metabolites and brain cortical thickness. We obtained urinary concentrations of PAH metabolites and brain magnetic resonance images from 327 adults (≥50 years of age) without dementia at baseline and 3-year follow-up. We obtained whole-brain and regional cortical thicknesses, as well as an Alzheimer's disease (AD)-specific marker for cortical atrophy (a higher score indicated a greater similarity to patients with AD) at baseline and follow-up. We built a linear mixed-effect model including each of urinary PAH metabolites as the time-varying exposure variable of interest. We found that increases in urinary concentrations of 1-hydroxypyrene (β = -0.004; 95% CI, -0.008 to -0.001) and 2-hydroxyfluorene (β = -0.011; 95% CI, -0.015 to -0.006) were significantly associated with a reduced whole-brain cortical thickness. A urinary concentration of 2-hydroxyfluorene was significantly associated with an increased AD-specific cortical atrophy score (β = 2.031; 95% CI, 0.512 to 3.550). The specific brain regions showing the association of urinary concentrations of 1-hydroxypyrene, 2-naphthol, 1-hydroxyphenanthrene, or 2-hydroxyfluorene with cortical thinning were the frontal, parietal, temporal, and cingulate lobes. These findings suggested that exposure to PAHs may reduce brain cortical thickness and increase the similarity to AD-specific cortical atrophy patterns in adults.

From genetic correlations of Alzheimer's disease to classification with artificial neural network models

Sporadic Alzheimer's disease (AD) is a complex neurological disorder characterized by many risk loci with potential associations with different traits and diseases. AD, characterized by a progressive loss of neuronal functions, manifests with different symptoms such as decline in memory, movement, coordination, and speech. The mechanisms underlying the onset of AD are not always fully understood, but involve a multiplicity of factors. Early diagnosis of AD plays a central role as it can offer the possibility of early treatment, which can slow disease progression. Currently, the methods of diagnosis are cognitive testing, neuroimaging, or cerebrospinal fluid analysis that can be time-consuming, expensive, invasive, and not always accurate. In the present study, we performed a genetic correlation analysis using genome-wide association statistics from a large study of AD and UK Biobank, to examine the association of AD with other human traits and disorders. In addition, since hippocampus, a part of cerebral cortex could play a central role in several traits that are associated with AD; we analyzed the gene expression profiles of hippocampus of AD patients applying 4 different artificial neural network models. We found 65 traits correlated with AD grouped into 9 clusters: medical conditions, fluid intelligence, education, anthropometric measures, employment status, activity, diet, lifestyle, and sexuality. The comparison of different 4 neural network models along with feature selection methods on 5 Alzheimer's gene expression datasets showed that the simple basic neural network model obtains a better performance (66% of accuracy) than other more complex methods with dropout and weight regularization of the network.

Development and validation of a deep learning-based automatic segmentation model for assessing intracranial volume: comparison with NeuroQuant, FreeSurfer, and SynthSeg

Background and purpose

To develop and validate a deep learning-based automatic segmentation model for assessing intracranial volume (ICV) and to compare the accuracy determined by NeuroQuant (NQ), FreeSurfer (FS), and SynthSeg.

Materials and methods

This retrospective study included 60 subjects [30 Alzheimer's disease (AD), 21 mild cognitive impairment (MCI), 9 cognitively normal (CN)] from a single tertiary hospital for the training and validation group (50:10). The test group included 40 subjects (20 AD, 10 MCI, 10 CN) from the ADNI dataset. We propose a robust ICV segmentation model based on the foundational 2D UNet architecture trained with four types of input images (both single and multimodality using scaled or unscaled T1-weighted and T2-FLAIR MR images). To compare with our model, NQ, FS, and SynthSeg were also utilized in the test group. We evaluated the model performance by measuring the Dice similarity coefficient (DSC) and average volume difference.

Results

The single-modality model trained with scaled T1-weighted images showed excellent performance with a DSC of 0.989 ± 0.002 and an average volume difference of 0.46% ± 0.38%. Our multimodality model trained with both unscaled T1-weighted and T2-FLAIR images showed similar performance with a DSC of 0.988 ± 0.002 and an average volume difference of 0.47% ± 0.35%. The overall average volume difference with our model showed relatively higher accuracy than NQ (2.15% ± 1.72%), FS (3.69% ± 2.93%), and SynthSeg (1.88% ± 1.18%). Furthermore, our model outperformed the three others in each subgroup of patients with AD, MCI, and CN subjects.

Conclusion

Our deep learning-based automatic ICV segmentation model showed excellent performance for the automatic evaluation of ICV.

Comparison of Normative Percentiles of Brain Volume Obtained from NeuroQuant® vs. DeepBrain® in the Korean Population: Correlation with Cranial Shape

Purpose

This study aimed to compare the volume and normative percentiles of brain volumetry in the Korean population using quantitative brain volumetric MRI analysis tools NeuroQuant® (NQ) and DeepBrain® (DB), and to evaluate whether the differences in the normative percentiles of brain volumetry between the two tools is related to cranial shape.

Materials and Methods

In this retrospective study, we analyzed the brain volume reports obtained from NQ and DB in 163 participants without gross structural brain abnormalities. We measured three-dimensional diameters to evaluate the cranial shape on T1-weighted images. Statistical analyses were performed using intra-class correlation coefficients and linear correlations.

Results

The mean normative percentiles of the thalamus (90.8 vs. 63.3 percentile), putamen (90.0 vs. 60.0 percentile), and parietal lobe (80.1 vs. 74.1 percentile) were larger in the NQ group than in the DB group, whereas that of the occipital lobe (18.4 vs. 68.5 percentile) was smaller in the NQ group than in the DB group. We found a significant correlation between the mean normative percentiles obtained from the NQ and cranial shape: the mean normative percentile of the occipital lobe increased with the anteroposterior diameter and decreased with the craniocaudal diameter.

Conclusion

The mean normative percentiles obtained from NQ and DB differed significantly for many brain regions, and these differences may be related to cranial shape.

Research on nonstroke dementia screening and cognitive function prediction model for older people based on brain atrophy characteristics

BACKGROUND

Brain atrophy is an important feature in dementia and is meaningful to explore a brain atrophy model to predict dementia. Using machine learning algorithm to establish a dementia model and cognitive function model based on brain atrophy characteristics is unstoppable.

METHOD

We acquired 157 dementia and 156 normal old people.s clinical information and MRI data, which contains 44 brain atrophy features, including visual scale assessment of brain atrophy and multiple linear measurement indexes and brain atrophy index. Five machine learning models were used to establish prediction models for dementia, general cognition, and subcognitive domains.

RESULTS

The extreme Gradient Boosting (XGBoost) model had the best effect in predicting dementia, with a sensitivity of 0.645, a specificity of 0.839, and the area under curve (AUC) of 0.784. In this model, the important brain atrophy features for predicting dementia were temporal horn ratio, cella media index, suprasellar cistern ratio, and the thickness of the corpus callosum genu.

CONCLUSION

For nonstroke elderly people, the machine learning model based on clinical head MRI brain atrophy features had good predictive value for dementia, general cognitive impairment, immediate memory impairment, word fluency disorder, executive dysfunction, and visualspatial disorder.

Agreement and Reliability between Clinically Available Software Programs in Measuring Volumes and Normative Percentiles of Segmented Brain Regions

OBJECTIVE

To investigate the agreement and reliability of estimating the volumes and normative percentiles (N%) of segmented brain regions among NeuroQuant (NQ), DeepBrain (DB), and FreeSurfer (FS) software programs, focusing on the comparison between NQ and DB.

MATERIALS AND METHODS

Three-dimensional T1-weighted images of 145 participants (48 healthy participants, 50 patients with mild cognitive impairment, and 47 patients with Alzheimer's disease) from a single medical center (SMC) dataset and 130 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset were included in this retrospective study. All images were analyzed with DB, NQ, and FS software to obtain volume estimates and N% of various segmented brain regions. We used Bland-Altman analysis, repeated measures ANOVA, reproducibility coefficient, effect size, and intraclass correlation coefficient (ICC) to evaluate inter-method agreement and reliability.

RESULTS

Among the three software programs, the Bland-Altman plot showed a substantial bias, the ICC showed a broad range of reliability (0.004-0.97), and repeated-measures ANOVA revealed significant mean volume differences in all brain regions. Similarly, the volume differences of the three software programs had large effect sizes in most regions (0.73-5.51). The effect size was largest in the pallidum in both datasets and smallest in the thalamus and cerebral white matter in the SMC and ADNI datasets, respectively. N% of NQ and DB showed an unacceptably broad Bland-Altman limit of agreement in all brain regions and a very wide range of ICC values (-0.142-0.844) in most brain regions.

CONCLUSION

NQ and DB showed significant differences in the measured volume and N%, with limited agreement and reliability for most brain regions. Therefore, users should be aware of the lack of interchangeability between these software programs when they are applied in clinical practice.

Unified framework for brain connectivity-based biomarkers in neurodegenerative disorders

Background

Brain connectivity is useful for deciphering complex brain dynamics controlling interregional communication. Identifying specific brain phenomena based on brain connectivity and quantifying their levels can help explain or diagnose neurodegenerative disorders.

Objective

This study aimed to establish a unified framework to identify brain connectivity-based biomarkers associated with disease progression and summarize them into a single numerical value, with consideration for connectivity-specific structural attributes.

Methods

This study established a framework that unifies the processes of identifying a brain connectivity-based biomarker and mapping its abnormality level into a single numerical value, called a biomarker abnormality summarized from the identified connectivity (BASIC) score. A connectivity-based biomarker was extracted in the form of a connected component associated with disease progression. BASIC scores were constructed to maximize Kendall's rank correlation with the disease, considering the spatial autocorrelation between adjacent edges. Using functional connectivity networks, we validated the BASIC scores in various scenarios.

Results

Our proposed framework was successfully applied to construct connectivity-based biomarker scores associated with disease progression, characterized by two, three, and five stages of Alzheimer's disease, and reflected the continuity of brain alterations as the diseases advanced. The BASIC scores were not only sensitive to disease progression, but also specific to the trajectory of a particular disease. Moreover, this framework can be utilized when disease stages are measured on continuous scales, resulting in a notable prediction performance when applied to the prediction of the disease.

Conclusion

Our unified framework provides a method to identify brain connectivity-based biomarkers and continuity-reflecting BASIC scores that are sensitive and specific to disease progression.

Cognitive Sequelae and Hippocampal Dysfunction in Chronic Kidney Disease following 5/6 Nephrectomy

Neurological disorders are prevalent in patients with chronic kidney disease (CKD). Vascular factors and uremic toxins are involved with cognitive impairment in CKD. In addition, vascular dementia-induced alterations in the structure and function of the hippocampus can lead to deficits in hippocampal synaptic plasticity and cognitive function. However, regardless of this clinical evidence, the pathophysiology of cognitive impairment in patients with CKD is not fully understood. We used male Sprague Dawley rats and performed 5/6 nephrectomy to observe the changes in behavior, field excitatory postsynaptic potential, and immunostaining of the hippocampus following CKD progression. We measured the hippocampus volume on magnetic resonance imaging scans in the controls (n = 34) and end-stage renal disease (ESRD) hemodialysis patients (n = 42). In four cognition-related behavior assays, including novel object recognition, Y-maze, Barnes maze, and classical contextual fear conditioning, we identified deficits in spatial working memory, learning and memory, and contextual memory, as well as the ability to distinguish familiar and new objects, in the rats with CKD. Immunohistochemical staining of Na⁺/H⁺ exchanger1 was increased in the hippocampus of the CKD rat models. We performed double immunofluorescent staining for aquaporin-4 and glial fibrillary acidic protein and then verified the high coexpression in the hippocampus of the CKD rat model. Furthermore, results from recoding of the field excitatory postsynaptic potential (fEPSP) in the hippocampus showed the reduced amplitude and slope of fEPSP in the CKD rats. ESRD patients with cognitive impairment showed a significant decrease in the hippocampus volume compared with ESRD patients without cognitive impairment or the controls. Our findings suggest that uremia resulting from decreased kidney function may cause the destruction of the blood–brain barrier and hippocampus-related cognitive impairment in CKD.

Harmonization of Multicenter Cortical Thickness Data by Linear Mixed Effect Model

Objective

Analyzing neuroimages being useful method in the field of neuroscience and neurology and solving the incompatibilities across protocols and vendors have become a major problem. We referred to this incompatibility as "center effects," and in this study, we attempted to correct such center effects of cortical feature obtained from multicenter magnetic resonance images (MRIs).

Methods

For MRI of a total of 4,321 multicenter subjects, the harmonized w-score was calculated by correcting biological covariates such as age, sex, years of education, and intercranial volume (ICV) as fixed effects and center information as a random effect. Afterward, we performed classification tasks using principal component analysis (PCA) and linear discriminant analysis (LDA) to check whether the center effect was successfully corrected from the harmonized w-score.

Results

First, an experiment was conducted to predict the dataset origin of a random subject sampled from two different datasets, and it was confirmed that the prediction accuracy of linear mixed effect (LME) model-based w-score was significantly closer to the baseline than that of raw cortical thickness. As a second experiment, we classified the data of the normal and patient groups of each dataset, and LME model-based w-score, which is biological-feature-corrected values, showed higher classification accuracy than the raw cortical thickness data. Afterward, to verify the compatibility of the dataset used for LME model training and the dataset that is not, intraobject comparison and w-score RMSE calculation process were performed.

Conclusion

Through comparison between the LME model-based w-score and existing methods and several classification tasks, we showed that the LME model-based w-score sufficiently corrects the center effects while preserving the disease effects from the dataset. We also showed that the preserved disease effects have a match with well-known disease atrophy patterns such as Alzheimer's disease or Parkinson's disease. Finally, through intrasubject comparison, we found that the difference between centers decreases in the LME model-based w-score compared with the raw cortical thickness and thus showed that our model well-harmonizes the data that are not used for the model training.

Choroid Plexus Volume and Permeability at Brain MRI within the Alzheimer Disease Clinical Spectrum

Background Mounting evidence suggests that the choroid plexus (CP) plays an important role in the pathophysiology of Alzheimer disease (AD), but its imaging profile in cognitive impairment remains unclear. Purpose To evaluate CP volume, permeability, and susceptibility by using MRI in patients at various stages of cognitive impairment. Materials and Methods This retrospective study evaluated patients with cognitive symptoms who underwent 3.0-T MRI of the brain, including dynamic contrast-enhanced (DCE) imaging and quantitative susceptibility mapping (QSM), between January 2013 and May 2020. CP volume was automatically segmented using three-dimensional T1-weighted sequences; the volume transfer constant (ie, K^trans) and fractional plasma volume (ie, V_p) were determined using DCE MRI, and susceptibility was assessed using QSM. The effects of CP volume, expressed as the ratio to intracranial volume, on cognition were evaluated using multivariable linear regression adjusted for age, sex, education, apolipoprotein E ε4 allele status, and volumetric measures. Results A total of 532 patients with cognitive symptoms (mean age, 72 years ± 9 [SD]; 388 women) were included: 78 with subjective cognitive impairment (SCI), 158 with early mild cognitive impairment (MCI), 149 with late MCI, and 147 with AD. Among these, 132 patients underwent DCE MRI and QSM. CP volume was greater in patients at more severe stages (ratio of intracranial volume × 10³: 0.9 ± 0.3 for SCI, 1.0 ± 0.3 for early MCI, 1.1 ± 0.3 for late MCI, and 1.3 ± 0.4 for AD; P < .001). Lower K^trans (r = -0.19; P = .03) and V_p (r = -0.20; P = .02) were negatively associated with CP volume; susceptibility was not (r = 0.15; P = .10). CP volume was negatively associated with memory (B = -0.67; standard error of the mean [SEM], 0.21; P = .01), executive function (B = -0.90; SEM, 0.31; P = .01), and global cognition (B = -0.82; SEM, 0.32; P = .01). Conclusion Among patients with cognitive symptoms, larger choroid plexus volume was associated with severity of cognitive impairment in the Alzheimer disease spectrum. Published under a CC BY 4.0 license. Online supplemental material is available for this article. See also the editorial by Chiang in this issue.

The effect of polycyclic aromatic hydrocarbons on changes in the brain structure of firefighters: An analysis using data from the Firefighters Research on Enhancement of Safety & Health study

Polycyclic aromatic hydrocarbons (PAHs) are formed during incomplete combustion of organic matter, and firefighters are highly exposed to these toxic compounds at fire sites. Exposure to PAHs can cause cognitive decline and neurodegeneration; however, to date, few studies have examined the potential effects of PAH exposure on structural changes in the brain. We aimed to investigate the association between the four types of PAH metabolites and the corresponding changes in neuroimaging markers based on smoking status and hypertension in male firefighters. For this, we utilized the 2-year follow-up data of 301 Korean male firefighters aged over 40 years. The concentrations of four PAH metabolites in urine were measured. Subcortical volume and cortical thickness were estimated using 3 T magnetic resonance imaging of the brain. A generalized linear model was used to investigate the effects of PAHs on changes in the subcortical volume and cortical thickness. We found an association between 1-hydroxyphenathrene (1-OHPHE) and 2-hydroxyfluorene (2-OHF) and changes in several brain regions in all the study participants. Individuals who had never smoked showed significantly thinner frontal (p < 0.001), parietal (p < 0.001), temporal (p < 0.001), and cingulate lobes (p < 0.001) with 1% increase each in the urinary concentration of 1-OHPHE. Hypertension interacted with the concentration of 1-OHPHE to reduce the volume of gray matter and cause cortical thinning in the frontal, parietal, and temporal lobes. Exposure to PAHs may reduce cortical thickness and subcortical volume, which are definitive markers of neurodegeneration. Notably, hypertension can accelerate the degenerative effects of PAHs.

Potential Applications of Artificial Intelligence in Clinical Trials for Alzheimer’s Disease

Clinical trials for Alzheimer’s disease (AD) face multiple challenges, such as the high screen failure rate and the even allocation of heterogeneous participants. Artificial intelligence (AI), which has become a potent tool of modern science with the expansion in the volume, variety, and velocity of biological data, offers promising potential to address these issues in AD clinical trials. In this review, we introduce the current status of AD clinical trials and the topic of machine learning. Then, a comprehensive review is focused on the potential applications of AI in the steps of AD clinical trials, including the prediction of protein and MRI AD biomarkers in the prescreening process during eligibility assessment and the likelihood stratification of AD subjects into rapid and slow progressors in randomization. Finally, this review provides challenges, developments, and the future outlook on the integration of AI into AD clinical trials.

Magnetic susceptibility in the deep gray matter may be modulated by apolipoprotein E4 and age with regional predilections: a quantitative susceptibility mapping study

PURPOSE

To examine the relationship between apolipoprotein E gene (APOE) mutation status and iron accumulation in the deep gray matter of subjects with cognitive symptoms using quantitative susceptibility mapping (QSM).

METHODS

A total of 105 patients with cognitive symptoms were enrolled. QSM data were generated from 3D gradient-echo data using an STI Suite algorithm. A region of interest-based analysis with QSM was performed in the deep gray matter. Differences between APOE4 carriers and non-carriers were assessed by analysis of covariance. Multiple regression analysis was performed to identify the factors associated with magnetic susceptibility.

RESULTS

Clinical characters such as age, education, MMSE, vascular risk burden, and systolic blood pressure differ between APOE4 carrier and non-carrier groups. The APOE4 carrier group had higher magnetic susceptibility values than the non-carrier group, with significant differences in the caudate (p = 0.004), putamen (p < 0.0001), and globus pallidus (p < 0.0001) which imply higher iron accumulation. In a multiple regression analysis, APOE4 status was found to be a predictor of magnetic susceptibility value in the globus pallidus (p = 0.03); age for magnetic susceptibility value in the caudate nucleus (p = 0.0064); and age and hippocampal atrophy for magnetic susceptibility value in the putamen (p < 0.05).

CONCLUSION

Our study demonstrates that magnetic susceptibility in globus pallidus is related to APOE4 status while those of caudate and putamen are related to other factors including age. It suggests that brain iron accumulation in the deep gray matter is modulated by APOE4 and age with differential regional predilection.

Monitoring Alzheimer's Disease Progression in Mild Cognitive Impairment Stage Using Machine Learning-Based FDG-PET Classification Methods

BACKGROUND

We previously introduced a machine learning-based Alzheimer's Disease Designation (MAD) framework for identifying AD-related metabolic patterns among neurodegenerative subjects.

OBJECTIVE

We sought to assess the efficiency of our MAD framework for tracing the longitudinal brain metabolic changes in the prodromal stage of AD.

METHODS

MAD produces subject scores using five different machine-learning algorithms, which include a general linear model (GLM), two different approaches of scaled subprofile modeling, and two different approaches of a support vector machine. We used our pre-trained MAD framework, which was trained based on metabolic brain features of 94 patients with AD and 111 age-matched cognitively healthy (CH) individuals. The MAD framework was applied on longitudinal independent test sets including 54 CHs, 51 stable mild cognitive impairment (sMCI), and 39 prodromal AD (pAD) patients at the time of the clinical diagnosis of AD, and two years prior.

RESULTS

The GLM showed excellent performance with area under curve (AUC) of 0.96 in distinguishing sMCI from pAD patients at two years prior to the time of the clinical diagnosis of AD while other methods showed moderate performance (AUC: 0.7-0.8). Significant annual increment of MAD scores were identified using all five algorithms in pAD especially when it got closer to the time of diagnosis (p < 0.001), but not in sMCI. The increased MAD scores were also significantly associated with cognitive decline measured by Mini-Mental State Examination in pAD (q < 0.01).

CONCLUSION

These results suggest that MAD may be a relevant tool for monitoring disease progression in the prodromal stage of AD.

Clinical Characteristic in Primary Progressive Aphasia in Relation to Alzheimer's Disease Biomarkers

BACKGROUND

Primary progressive aphasia (PPA) is associated with amyloid-β (Aβ) pathology. However, clinical feature of PPA based on Aβ positivity remains unclear.

OBJECTIVE

We aimed to assess the prevalence of Aβ positivity in patients with PPA and compare the clinical characteristics of patients with Aβ-positive (A+) and Aβ-negative (A-) PPA. Further, we applied Aβ and tau classification system (AT system) in patients with PPA for whom additional information of in vivo tau biomarker was available.

METHODS

We recruited 110 patients with PPA (41 semantic [svPPA], 27 non-fluent [nfvPPA], 32 logopenic [lvPPA], and 10 unclassified [ucPPA]) who underwent Aβ-PET imaging at multi centers. The extent of language impairment and cortical atrophy were compared between the A+ and A-PPA subgroups using general linear models.

RESULTS

The prevalence of Aβ positivity was highest in patients with lvPPA (81.3%), followed by ucPPA (60.0%), nfvPPA (18.5%), and svPPA (9.8%). The A+ PPA subgroup manifested cortical atrophy mainly in the left superior temporal/inferior parietal regions and had lower repetition scores compared to the A-PPA subgroup. Further, we observed that more than 90% (13/14) of the patients with A+ PPA had tau deposition.

CONCLUSION

Our findings will help clinicians understand the patterns of language impairment and cortical atrophy in patients with PPA based on Aβ deposition. Considering that most of the A+ PPA patents are tau positive, understanding the influence of Alzheimer's disease biomarkers on PPA might provide an opportunity for these patients to participate in clinical trials aimed for treating atypical Alzheimer's disease.

Cohort Profile: The Environmental-Pollution-Induced Neurological EFfects (EPINEF) study, a multicenter cohort study of Korean adults

The general population is exposed to numerous environmental pollutants, and it remains unclear which pollutants affect the brain, accelerating brain aging and increasing the risk of dementia. The Environmental-Pollution-Induced Neurological Effects study is a multi-city prospective cohort study aiming to comprehensively investigate the effect of different environmental pollutants on brain structures, neuropsychological function, and the development of dementia in adults. The baseline data of 3,775 healthy elderly people were collected from August 2014 to March 2018. The eligibility criteria were age ≥50 years and no self-reported history of dementia, movement disorders, or stroke. The assessment included demographics and anthropometrics, laboratory test results, and individual levels of exposure to air pollution. A neuroimaging sub-cohort was also recruited with 1,022 participants during the same period, and brain magnetic resonance imaging and neuropsychological tests were conducted. The first follow-up environmental pollutant measurements will start in 2022 and the follow-up for the sub-cohort will be conducted every 3-4 years. We have found that subtle structural changes in the brain may be induced by exposure to airborne pollutants such as particulate matter 10 μm or less in diameter (PM₁₀), particulate matter 2.5 μm or less in diameter (PM_2.5) and Mn₁₀, manganese in PM₁₀; Mn_2.5, manganese in PM_2.5. PM₁₀, PM_2.5, and nitrogen dioxide in healthy adults. This study provides a basis for research involving large-scale, long-term neuroimaging assessments in community-based populations.

Hippocampal blood-brain barrier permeability is related to the APOE4 mutation status of elderly individuals without dementia

Blood-brain barrier (BBB) disruption, modulated by APOE4 mutation, is implicated in the pathogenesis of cognitive decline. We determined whether BBB permeability differed according to cognitive functioning and APOE4 status in elderly subjects without dementia. In this prospective study, 33 subjects with mild cognitive impairment (MCI) and 33 age-matched controls (normal cognition [NC]) underwent 3 T brain magnetic resonance imaging. The Patlak model was used to calculate tissue permeability (K_trans). A region-of interest analysis of K_trans was performed to compare relevant brain regions. Effects of K_trans on cognitive functioning were evaluated with linear regression analysis adjusted for confounding factors. NC and MCI groups did not differ in terms of vascular risk factors or hippocampal K_trans, except for hippocampal volume. Hippocampal K_trans was significantly higher in APOE4 carriers than in non-carriers (p = 0.007). Factors which predicted cognitive functioning included hippocampal volume (beta=-0.445, standard error [SE]=0.137, p = 0.003) and hippocampal BBB permeability (beta = 0.142, SE = 0.050, p = 0.008) after correcting for age, education, and APOE4 status. This suggests that hippocampal BBB permeability is associated with APOE4 mutation, and may predict cognitive functioning. BBB permeability imaging represents a distinct imaging biomarker for APOE4 mutations in NC and MCI subjects and for determining the degree of APOE4-related pathology.

Evaluation of Reproducibility of Brain Volumetry between Commercial Software, Inbrain and Established Research Purpose Method, FreeSurfer

Background and Purpose

We aimed to determine the intermethod reproducibility between the commercial software Inbrain (MIDAS IT) and the established research-purpose method FreeSurfer, as well as the effect of MRI resolution and the pathological condition of subjects on their intermethod reproducibility.

Methods

This study included 45 healthy volunteers and 85 patients with mild cognitive impairment (MCI). In 43 of the 85 patients with MCI, three-dimensional, T1-weighted MRI data were obtained at an in-plane resolution of 1.2 mm. The data of the remaining 42 patients with MCI and the healthy volunteers were obtained at an in-plane resolution of 1.0 mm. The within-subject coefficient of variation (CoV), intraclass correlation coefficient (ICC), and effect size were calculated, and means were compared using paired t-tests. The parameters obtained at 1.0-mm and 1.2-mm resolutions in patients with MCI were compared to evaluate the effect of the in-plane resolution on the intermethod reproducibility. The parameters obtained at a 1.0-mm in-plane resolution in patients with MCI and healthy volunteers were used to analyze the effect of subject condition on intermethod reproducibility.

Results

Overall the two methods showed excellent reproducibility across all regions of the brain (CoV=0.5–3.9, ICC=0.93 to >0.99). In the subgroup of healthy volunteers, the intermethod reliability was only good in some regions (frontal, temporal, cingulate, and insular). The intermethod reproducibility was better in the 1.0-mm group than the 1.2-mm group in all regions other than the nucleus accumbens.

Conclusions

Inbrain and FreeSurfer showed good-to-excellent intermethod reproducibility for volumetric measurements. Nevertheless, some noticeable differences were found based on subject condition, image resolution, and brain region.

Expert Opinions and Recommendations for the Clinical Use of Quantitative Analysis Software for MRI-Based Brain Volumetry

치매를 비롯한 퇴행성 신경 질환의 초기 진단에 자기공명영상을 이용한 뇌 위축 평가와 정량적 용적 분석이 중요하다. 뇌 위축의 시각적 평가는 주관적으로 평가자에 따라 다른 결과를 보여주기 때문에, 객관적인 결과를 제공하면서 임상 적용도 가능한 소프트웨어의 수요와 개발이 늘어나고 있다. 이러한 임상용 소프트웨어의 실제 임상 적용은 영상 검사의 표준화가 선행되어야 하고, 개발된 소프트웨어의 검증이 반드시 필요하다. 따라서 대한신경두경부영상의학회는 뇌용적 분석 임상용 소프트웨어의 임상적 활용에 대한 의견을 제시하기 위해 전문위원회를 구성하고 현재까지 발표된 연구를 정리하였다. 그리고, 정량화 분석을 위한 영상 검사의 표준화 및 소프트웨어의 임상 적용에 대한 전문가 의견을 제시하기 위하여 공동 작업을 수행하였다. 본 종설에서는 뇌 자기공명영상의 정량화 분석의 필요성 및 배경, 정량화 분석을 위한 임상용 소프트웨어의 소개 및 기존의 표준품(reference standard)과의 진단능 비교, 영상 획득의 표준화, 분석 및 평가의 표준화, 소프트웨어의 임상 적용에 대한 전문가 의견, 제한점 및 대처 방법 등 대한신경두경부영상의학회의 전문가 권고안을 소개하는 것이 목적이다.

Development of Random Forest Algorithm Based Prediction Model of Alzheimer's Disease Using Neurodegeneration Pattern

OBJECTIVE

Alzheimer's disease (AD) is the most common type of dementia and the prevalence rapidly increased as the elderly population increased worldwide. In the contemporary model of AD, it is regarded as a disease continuum involving preclinical stage to severe dementia. For accurate diagnosis and disease monitoring, objective index reflecting structural change of brain is needed to correctly assess a patient's severity of neurodegeneration independent from the patient's clinical symptoms. The main aim of this paper is to develop a random forest (RF) algorithm-based prediction model of AD using structural magnetic resonance imaging (MRI).

METHODS

We evaluated diagnostic accuracy and performance of our RF based prediction model using newly developed brain segmentation method compared with the Freesurfer's which is a commonly used segmentation software.

RESULTS

Our RF model showed high diagnostic accuracy for differentiating healthy controls from AD and mild cognitive impairment (MCI) using structural MRI, patient characteristics, and cognitive function (HC vs. AD 93.5%, AUC 0.99; HC vs. MCI 80.8%, AUC 0.88). Moreover, segmentation processing time of our algorithm (<5 minutes) was much shorter than of Freesurfer's (6-8 hours).

CONCLUSION

Our RF model might be an effective automatic brain segmentation tool which can be easily applied in real clinical practice.

Concordance in detecting amyloid positivity between 18F-florbetaben and 18F-flutemetamol amyloid PET using quantitative and qualitative assessments

We aimed to quantitatively and qualitatively assess whether there is a discrepancy in detecting amyloid beta (Aβ) positivity between 18F-florbetaben (FBB) and 18F-flutemetamol (FMM) positron emission tomography (PET). We obtained paired FBB and FMM PET images from 107 participants. Three experts visually quantified the Aβ deposition as positive or negative. Quantitative assessment was performed using global cortical standardized uptake value ratio (SUVR) with the whole cerebellum as the reference region. Inter-rater agreement was excellent for FBB and FMM. The concordance rates between FBB and FMM were 94.4% (101/107) for visual assessment and 98.1% (105/107) for SUVR cut-off categorization. Both FBB and FMM showed high agreement rates between visual assessment and SUVR positive or negative categorization (93.5% in FBB and 91.2% in FMM). When the two ligands were compared based on SUVR cut-off categorization as standard of truth, although not statistically significant, the false-positive rate was higher in FMM (9.1%) than in FBB (1.8%) (p = 0.13). Our findings suggested that both FBB and FMM had excellent agreement when used to quantitatively and qualitatively evaluate Aβ deposits, thus, combining amyloid PET data associated with the use of different ligands from multi-centers is feasible.

Clinically Available Software for Automatic Brain Volumetry: Comparisons of Volume Measurements and Validation of Intermethod Reliability

OBJECTIVE

To compare two clinically available MR volumetry software, NeuroQuant® (NQ) and Inbrain® (IB), and examine the inter-method reliabilities and differences between them.

MATERIALS AND METHODS

This study included 172 subjects (age range, 55-88 years; mean age, 71.2 years), comprising 45 normal healthy subjects, 85 patients with mild cognitive impairment, and 42 patients with Alzheimer's disease. Magnetic resonance imaging scans were analyzed with IB and NQ. Mean differences were compared with the paired t test. Inter-method reliability was evaluated with Pearson's correlation coefficients and intraclass correlation coefficients (ICCs). Effect sizes were also obtained to document the standardized mean differences.

RESULTS

The paired t test showed significant volume differences in most regions except for the amygdala between the two methods. Nevertheless, inter-method measurements between IB and NQ showed good to excellent reliability (0.72 < r < 0.96, 0.83 < ICC < 0.98) except for the pallidum, which showed poor reliability (left: r = 0.03, ICC = 0.06; right: r = -0.05, ICC = -0.09). For the measurements of effect size, volume differences were large in most regions (0.05 < r < 6.15). The effect size was the largest in the pallidum and smallest in the cerebellum.

CONCLUSION

Comparisons between IB and NQ showed significantly different volume measurements with large effect sizes. However, they showed good to excellent inter-method reliability in volumetric measurements for all brain regions, with the exception of the pallidum. Clinicians using these commercial software should take into consideration that different volume measurements could be obtained depending on the software used.

Cortical thinning pattern according to differential nigrosome involvement in patients with Parkinson's disease

The pathological hallmark of Parkinson's disease (PD) is the progressive degeneration of dopaminergic neurons in the substantia nigra pars compacta, where the dopaminergic neurons form five clusters called nigrosomes 1-5 (N1-N5). N1 is the largest and considered to be the most affected by PD, followed by N2, N4, N3, and N5. Recently, an MRI study suggested a sequential progression of loss from N1 to N4. As the extent of cortical thinning widens as PD progresses, we aimed to define cortical thinning patterns according to the differential involvement of N1 and N4 in PD patients. Cortical thickness was analyzed in 83 PD patients (29 with N1 loss on at least one side of the brain, but no N4 loss; and 54 with N4 loss on at least one side) and 35 healthy subjects with age, sex, disease duration, and intracranial volume as covariates. On patient-wise analysis, for areas with more cortical thinning than the controls, PD patients with N4 loss had wider cortical thinning involving more dorsolateral prefrontal cortex and temporal areas than PD patients with only N1 loss, but cortical thinning did not significantly differ between these two patient groups. However, cortical thinning was more apparent in hemisphere-level analysis with statistically significant clusters being found more in hemispheres with N4 loss than hemispheres with N1 loss in PD patients compared to normal hemispheres of the controls. Cortical thinning occurred in a similar propagation pattern to that seen with PD progression, supporting past hypotheses on the sequential progression of nigrosome loss from N1 to N4.

Imaging biomarkers in neurodegeneration: current and future practices

There is an increasing role for biological markers (biomarkers) in the understanding and diagnosis of neurodegenerative disorders. The application of imaging biomarkers specifically for the in vivo investigation of neurodegenerative disorders has increased substantially over the past decades and continues to provide further benefits both to the diagnosis and understanding of these diseases. This review forms part of a series of articles which stem from the University College London/University of Gothenburg course "Biomarkers in neurodegenerative diseases". In this review, we focus on neuroimaging, specifically positron emission tomography (PET) and magnetic resonance imaging (MRI), giving an overview of the current established practices clinically and in research as well as new techniques being developed. We will also discuss the use of machine learning (ML) techniques within these fields to provide additional insights to early diagnosis and multimodal analysis.

Cohort Profile: Firefighter Research on the Enhancement of Safety and Health (FRESH), a Prospective Cohort Study on Korean Firefighters

Firefighters have a high risk of developing cardiovascular and mental disorders due to their physical and chemical environments. However, in Korea, few studies have been conducted on environmental risk of firefighters. The Firefighter Research on the Enhancement of Safety and Health (FRESH) study aimed to discover the risk factors for cardiovascular disease and mental disorders among firefighters. Former and current firefighters were recruited from three university hospitals. A total of 1022 participants completed baseline health examinations from 2016 to 2017. All participants were scheduled for follow-ups every 2 years. Baseline health survey, laboratory testing of blood and urine samples, blood heavy metal concentration, urine polycyclic aromatic hydrocarbons (PAHs) metabolites, stress-related hormone test, natural killer cell activity, as well as physical and mental health examinations that focused on cardiovascular and mental disorders, were conducted. In addition, 3 Tesla (3T) brain magnetic resonance imaging (MRI) and neuropsychological tests were also performed to investigate structural and functional changes in the brains of 352 firefighters aged >40 years or new hires with less than 1 year of service.

Machine learning based hierarchical classification of frontotemporal dementia and Alzheimer's disease

Background

In a clinical setting, an individual subject classification model rather than a group analysis would be more informative. Specifically, the subtlety of cortical atrophy in some frontotemporal dementia (FTD) patients and overlapping patterns of atrophy among three FTD clinical syndromes including behavioral variant FTD (bvFTD), non-fluent/agrammatic variant primary progressive aphasia (nfvPPA), and semantic variant PPA (svPPA) give rise to the need for classification models at the individual level. In this study, we aimed to classify each individual subject into one of the diagnostic categories in a hierarchical manner by employing a machine learning-based classification method.

Methods

We recruited 143 patients with FTD, 50 patients with Alzheimer's disease (AD) dementia, and 146 cognitively normal subjects. All subjects underwent a three-dimensional volumetric brain magnetic resonance imaging (MRI) scan, and cortical thickness was measured using FreeSurfer. We applied the Laplace Beltrami operator to reduce noise in the cortical thickness data and to reduce the dimension of the feature vector. Classifiers were constructed by applying both principal component analysis and linear discriminant analysis to the cortical thickness data. For the hierarchical classification, we trained four classifiers using different pairs of groups: Step 1 - CN vs. FTD + AD, Step 2 - FTD vs. AD, Step 3 - bvFTD vs. PPA, Step 4 - svPPA vs. nfvPPA. To evaluate the classification performance for each step, we used a10-fold cross-validation approach, performed 1000 times for reliability.

Results

The classification accuracy of the entire hierarchical classification tree was 75.8%, which was higher than that of the non-hierarchical classifier (73.0%). The classification accuracies of steps 1–4 were 86.1%, 90.8%, 86.9%, and 92.1%, respectively. Changes in the right frontotemporal area were critical for discriminating behavioral variant FTD from PPA. The left frontal lobe discriminated nfvPPA from svPPA, while the bilateral anterior temporal regions were critical for identifying svPPA.

Conclusions

In the present study, our automated classifier successfully classified FTD clinical subtypes with good to excellent accuracy. Our classifier may help clinicians diagnose FTD subtypes with subtle cortical atrophy and facilitate appropriate specific interventions.

•

We developed a machine learning-based automated classifier for differential diagnosis of FTD clinical syndromes and AD.

•

Our classifier achieved good to excellent accuracy for each classification step.

•

Discriminative regions are similar to previously known cortical atrophic patterns in each clinical syndrome.