Identifying subtypes of mild cognitive impairment from healthy aging based on multiple cortical features combined with volumetric measurements of the hippocampal subfields

Machine learning with multimodal neuroimaging data to classify stages of Alzheimer's disease: a systematic review and meta-analysis

In recent years, Alzheimer's disease (AD) has been a serious threat to human health. Researchers and clinicians alike encounter a significant obstacle when trying to accurately identify and classify AD stages. Several studies have shown that multimodal neuroimaging input can assist in providing valuable insights into the structural and functional changes in the brain related to AD. Machine learning (ML) algorithms can accurately categorize AD phases by identifying patterns and linkages in multimodal neuroimaging data using powerful computational methods. This study aims to assess the contribution of ML methods to the accurate classification of the stages of AD using multimodal neuroimaging data. A systematic search is carried out in IEEE Xplore, Science Direct/Elsevier, ACM DigitalLibrary, and PubMed databases with forward snowballing performed on Google Scholar. The quantitative analysis used 47 studies. The explainable analysis was performed on the classification algorithm and fusion methods used in the selected studies. The pooled sensitivity and specificity, including diagnostic efficiency, were evaluated by conducting a meta-analysis based on a bivariate model with the hierarchical summary receiver operating characteristics (ROC) curve of multimodal neuroimaging data and ML methods in the classification of AD stages. Wilcoxon signed-rank test is further used to statistically compare the accuracy scores of the existing models. With a 95% confidence interval of 78.87-87.71%, the combined sensitivity for separating participants with mild cognitive impairment (MCI) from healthy control (NC) participants was 83.77%; for separating participants with AD from NC, it was 94.60% (90.76%, 96.89%); for separating participants with progressive MCI (pMCI) from stable MCI (sMCI), it was 80.41% (74.73%, 85.06%). With a 95% confidence interval (78.87%, 87.71%), the Pooled sensitivity for distinguishing mild cognitive impairment (MCI) from healthy control (NC) participants was 83.77%, with a 95% confidence interval (90.76%, 96.89%), the Pooled sensitivity for distinguishing AD from NC was 94.60%, likewise (MCI) from healthy control (NC) participants was 83.77% progressive MCI (pMCI) from stable MCI (sMCI) was 80.41% (74.73%, 85.06%), and early MCI (EMCI) from NC was 86.63% (82.43%, 89.95%). Pooled specificity for differentiating MCI from NC was 79.16% (70.97%, 87.71%), AD from NC was 93.49% (91.60%, 94.90%), pMCI from sMCI was 81.44% (76.32%, 85.66%), and EMCI from NC was 85.68% (81.62%, 88.96%). The Wilcoxon signed rank test showed a low P-value across all the classification tasks. Multimodal neuroimaging data with ML is a promising future in classifying the stages of AD but more research is required to increase the validity of its application in clinical practice.

Atrophy of hippocampal subfields and amygdala nuclei in subjects with mild cognitive impairment progressing to Alzheimer's disease

The hippocampus and amygdala are the first brain regions to show early signs of Alzheimer's Disease (AD) pathology. AD is preceded by a prodromal stage known as Mild Cognitive Impairment (MCI), a crucial crossroad in the clinical progression of the disease. The topographical development of AD has been the subject of extended investigation. However, it is still largely unknown how the transition from MCI to AD affects specific hippocampal and amygdala subregions. The present study is set to answer that question. We analyzed data from 223 subjects: 75 healthy controls, 52 individuals with MCI, and 96 AD patients obtained from the ADNI. The MCI group was further divided into two subgroups depending on whether individuals in the 48 months following the diagnosis either remained stable (N = 21) or progressed to AD (N = 31). A MANCOVA test evaluated group differences in the volume of distinct amygdala and hippocampal subregions obtained from magnetic resonance images. Subsequently, a stepwise linear discriminant analysis (LDA) determined which combination of magnetic resonance imaging parameters was most effective in predicting the conversion from MCI to AD. The predictive performance was assessed through a Receiver Operating Characteristic analysis. AD patients displayed widespread subregional atrophy. MCI individuals who progressed to AD showed selective atrophy of the hippocampal subiculum and tail compared to stable MCI individuals, who were undistinguishable from healthy controls. Converter MCI showed atrophy of the amygdala's accessory basal, central, and cortical nuclei. The LDA identified the hippocampal subiculum and the amygdala's lateral and accessory basal nuclei as significant predictors of MCI conversion to AD. The analysis returned a sensitivity value of 0.78 and a specificity value of 0.62. These findings highlight the importance of targeted assessments of distinct amygdala and hippocampus subregions to help dissect the clinical and pathophysiological development of the MCI to AD transition.

Machine learning classifiers and associations of cognitive performance with hippocampal subfields in amnestic mild cognitive impairment

Background

Neuroimaging studies have demonstrated alterations in hippocampal volume and hippocampal subfields among individuals with amnestic mild cognitive impairment (aMCI). However, research on using hippocampal subfield volume modeling to differentiate aMCI from normal controls (NCs) is limited, and the relationship between hippocampal volume and overall cognitive scores remains unclear.

Methods

We enrolled 50 subjects with aMCI and 44 NCs for this study. Initially, a univariate general linear model was employed to analyze differences in the volumes of hippocampal subfields. Subsequently, two sets of dimensionality reduction methods and four machine learning techniques were applied to distinguish aMCI from NCs based on hippocampal subfield volumes. Finally, we assessed the correlation between the relative volumes of hippocampal subfields and cognitive test variables (Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment Scale (MoCA)).

Results

Significant volume differences were observed in several hippocampal subfields, notably in the left hippocampus. Specifically, the volumes of the hippocampal tail, subiculum, CA1, presubiculum, molecular layer, GC-ML-DG, CA3, CA4, and fimbria differed significantly between the two groups. The highest area under the curve (AUC) values for left and right hippocampal machine learning classifiers were 0.678 and 0.701, respectively. Moreover, the volumes of the left subiculum, left molecular layer, right subiculum, right CA1, right molecular layer, right GC-ML-DG, and right CA4 exhibited the strongest and most consistent correlations with MoCA scores.

Conclusion

Hippocampal subfield volume may serve as a predictive marker for aMCI. These findings underscore the sensitivity of hippocampal subfield volume to overall cognitive performance.

Enhanced Harris hawks optimization-based fuzzy k-nearest neighbor algorithm for diagnosis of Alzheimer's disease

In order to stop deterioration and give patients with Alzheimer's disease (AD) early therapy, it is crucial to correctly diagnose AD and its early stage, mild cognitive impairment (MCI). A framework for diagnosing AD is presented in this paper, which includes magnetic resonance imaging (MRI) image preprocessing, feature extraction, and the Fuzzy k-nearest neighbor algorithm (FKNN) model. In particular, the framework's novelty lies in the use of an improved Harris Hawks Optimization (HHO) algorithm named SSFSHHO, which integrates the Sobol sequence and Stochastic Fractal Search (SFS) mechanisms for optimizing the parameters of FKNN. The HHO method improves the quality of the initial population overall by incorporating the Sobol sequence, and the SFS mechanism increases the algorithm's capacity to get out of the local optimum solution. Comparisons with other classical meta-heuristic algorithms, state-of-the-art HHO variants in low and high dimensions, and enhanced meta-heuristic algorithms on 30 typical IEEE CEC2014 benchmark test problems show that the overall performance of SSFSHHO is significantly better than other comparative algorithms. Moreover, the created framework based on the SSFSHHO-FKNN model is employed to classify AD and MCI using MRI scans from the ADNI dataset, achieving high classification performance for 6 representative cases. Experimental findings indicate that the proposed algorithm performs better than a number of high-performance optimization algorithms and classical machine learning algorithms, thus offering a promising approach for AD classification. Additionally, the proposed strategy can successfully identify relevant features and enhance classification performance for AD diagnosis.

A graph theoretic approach to neurodegeneration: five data-driven neuropsychological subtypes in mild cognitive impairment

Mild cognitive Impairment (MCI) is notoriously heterogenous in terms of clinical presentation, neuroimaging correlates, and subsequent progression. Predicting who will progress to dementia, which type of dementia, and over what timeframe is challenging. Previous work has attempted to identify MCI subtypes using neuropsychological measures in an effort to address this challenge; however, there is no consensus on approach, which may account for some of the variability. Using a hierarchical community detection approach, we examined cognitive subtypes within an MCI sample (from the Alzheimer's Disease Neuroimaging Initiative [ADNI] study). We then examined whether these subtypes were related to biomarkers (e.g., cortical volumes, fluorodeoxyglucose (FDG)-positron emission tomography (PET) hypometabolism) or clinical progression. We identified five communities (i.e., cognitive subtypes) within the MCI sample: 1) predominantly memory impairment, 2) predominantly language impairment, 3) cognitively normal, 4) multidomain, with notable executive dysfunction, 5) multidomain, with notable processing speed impairment. Community membership was significantly associated with 1) cortical volume in the hippocampus, entorhinal cortex, and fusiform cortex; 2) FDG PET hypometabolism in the posterior cingulate, angular gyrus, and inferior/middle temporal gyrus; and 3) conversion to dementia at follow up. Overall, community detection as an approach appears a viable method for identifying unique cognitive subtypes in a neurodegenerative sample that were linked to several meaningful biomarkers and modestly with progression at one year follow up.

Age- and sex-related differences in cortical morphology and their relationships with cognitive performance in healthy middle-aged and older adults

Background

The impacts of age and sex on brain structures related to cognitive function may be important for understanding the role of aging in Alzheimer disease for both sexes. We intended to investigate the age and sex differences of cortical morphology in middle-aged and older adults and their relationships with the decline of cognitive function.

Methods

In this cross-sectional study, we examined the cortical morphology in 204 healthy middle-aged and older adult participants aged 45 to 89 years using structural magnetic resonance imaging (sMRI) data from the Dallas Lifespan Brain Study data set. Brain cortical thickness, surface complexity, and gyrification index were analyzed through a completely automated surface-based morphometric analysis using the CAT12 toolbox. Furthermore, we explored the correlation between cortical morphology differences and test scores for processing speed and working memory.

Results

There were no significant interactions of age and sex with cortical thickness, fractal dimension, or gyrification index. Rather, we found that both males and females showed age-related decreases in cortical thickness, fractal dimension, and gyrification index. There were significant sex differences in the fractal dimension in middle-aged participants and the gyrification index in older adult participants. In addition, there were significant positive correlations between the cortical thickness of the right superior frontal gyrus and Wechsler Adult Intelligence Scale (WAIS)-III Letter-Number Sequencing test scores in males (r=0.394; P<0.001; 95% CI for r values 0.216-0.577) and females (r=0.344; P<0.001; 95% CI for r values 0.197-0.491), respectively. Furthermore, a significant relationship between the gyrification index of the right supramarginal gyrus (SupraMG) and WAIS-III Digit Symbol test scores was observed in older adult participants (r=0.375; P<0.001; 95% CI for r values 0.203-0.522).

Conclusions

The results suggest that, compared with males, females have more extensive differences in cortical morphology. The gyrification index of the right SupraMG can be used as an imaging marker of sexual cognitive differences between males and females in older adults. This study helps to further understand sex differences in the aging of the brain and cognition.

Cognition and brain iron deposition in whole grey matter regions and hippocampal subfields

Regional brain iron accumulation is observed in many neurodegenerative diseases, including Alzheimer's disease and Parkinson's disease, and is associated with cognitive decline. We explored associations between age, cognition and iron content in grey matter regions and hippocampal subfields in 380 participants of the Aberdeen children of the 1950s cohort and their first-generation relatives (aged 26-72 years). Participants underwent cognitive assessment at the time of MRI scanning. Quantitative susceptibility mapping of these MRI data was used to assess iron content in grey matter regions and in hippocampal subfields. Principle component analysis was performed on cognitive test scores to create a general cognition score. Spline analysis was used with the Akaike information criterion to determine if order 1, 2 or 3 natural splines were optimal for assessing non-linear relationships between regional iron and age. Multivariate linear models were used to assess associations between regional iron and cognition. Higher iron correlated with older age in the left putamen across all ages and in the right putamen of only participants over 58. Whereas a decrease in iron with older age was observed in the right thalamus and left pallidum across all ages. Right amygdala iron levels were associated with poorer general cognition scores and poorer immediate recall scores. Iron was not associated with any measures of cognitive performance in other regions of interest. Our results suggest that, whilst iron in some regions was associated with cognitive performance, there is an overall lack of association between regional iron content and cognitive ability in cognitively healthy individuals.

IHGC-GAN: influence hypergraph convolutional generative adversarial network for risk prediction of late mild cognitive impairment based on imaging genetic data

Predicting disease progression in the initial stage to implement early intervention and treatment can effectively prevent the further deterioration of the condition. Traditional methods for medical data analysis usually fail to perform well because of their incapability for mining the correlation pattern of pathogenies. Therefore, many calculation methods have been excavated from the field of deep learning. In this study, we propose a novel method of influence hypergraph convolutional generative adversarial network (IHGC-GAN) for disease risk prediction. First, a hypergraph is constructed with genes and brain regions as nodes. Then, an influence transmission model is built to portray the associations between nodes and the transmission rule of disease information. Third, an IHGC-GAN method is constructed based on this model. This method innovatively combines the graph convolutional network (GCN) and GAN. The GCN is used as the generator in GAN to spread and update the lesion information of nodes in the brain region-gene hypergraph. Finally, the prediction accuracy of the method is improved by the mutual competition and repeated iteration between generator and discriminator. This method can not only capture the evolutionary pattern from early mild cognitive impairment (EMCI) to late MCI (LMCI) but also extract the pathogenic factors and predict the deterioration risk from EMCI to LMCI. The results on the two datasets indicate that the IHGC-GAN method has better prediction performance than the advanced methods in a variety of indicators.

Scan Once, Analyse Many: Using Large Open-Access Neuroimaging Datasets to Understand the Brain

We are now in a time of readily available brain imaging data. Not only are researchers now sharing data more than ever before, but additionally large-scale data collecting initiatives are underway with the vision that many future researchers will use the data for secondary analyses. Here I provide an overview of available datasets and some example use cases. Example use cases include examining individual differences, more robust findings, reproducibility-both in public input data and availability as a replication sample, and methods development. I further discuss a variety of considerations associated with using existing data and the opportunities associated with large datasets. Suggestions for further readings on general neuroimaging and topic-specific discussions are also provided.

Prediction of Subtle Cognitive Decline in Normal Aging: Added Value of Quantitative MRI and PET Imaging

Quantitative imaging processing tools have been proposed to improve clinic-radiological correlations but their added value at the initial stages of cognitive decline is still a matter of debate. We performed a longitudinal study in 90 community-dwelling elders with three neuropsychological assessments during a 4.5 year follow-up period, and visual assessment of medial temporal atrophy (MTA), white matter hyperintensities, cortical microbleeds (CMB) as well as amyloid positivity, and presence of abnormal FDG-PET patterns. Quantitative imaging data concerned ROI analysis of MRI volume, amyloid burden, and FDG-PET metabolism in several AD-signature areas. Multiple regression models, likelihood-ratio tests, and areas under the receiver operating characteristic curve (AUC) were used to compare quantitative imaging markers to visual inspection. The presence of more or equal to four CMB at inclusion and slight atrophy of the right MTL at follow-up were the only parameters to be independently related to the worst cognitive score explaining 6% of its variance. This percentage increased to 24.5% when the ROI-defined volume loss in the posterior cingulate cortex, baseline hippocampus volume, and MTL metabolism were also considered. When binary classification of cognition was made, the area under the ROC curve increased from 0.69 for the qualitative to 0.79 for the mixed imaging model. Our data reveal that the inclusion of quantitative imaging data significantly increases the prediction of cognitive changes in elderly controls compared to the single consideration of visual inspection.

Aberrant cerebral perfusion pattern in amnestic mild cognitive impairment and Parkinson's disease with mild cognitive impairment: a comparative arterial spin labeling study

Background

Mild cognitive impairment (MCI) has been defined as the prodromal stage of Alzheimer's disease and Parkinson's disease (PD) with dementia. We investigated the differences in regional perfusion properties among MCI subtypes and healthy control (HC) subjects by using arterial spin labeling (ASL).

Methods

Regional normalized CBF (z-CBF) and CBF-connectivity were analyzed from ASL data in 44 amnestic MCI (aMCI) patients, 42 PD-MCI patients, and 50 matched HC participants. The correlations between these significant regions and clinical performance were investigated separately using Spearman correlation analysis. Receiver operating characteristic analysis was generated to determine the differentiating ability of z-CBF values. z-CBF values in disease-related specific regions were extracted for group comparison.

Results

MCI subgroups showed overlapped impaired regions, aMCI group seemed more extensive than the PD-MCI group. PD-MCI patients had reduced z-CBF in the bilateral putamen, left precentral gyrus, left middle cingulate gyrus, and right middle frontal gyrus compared to aMCI group. Correlations to executive performance and motor severity were found in PD-MCI group, and correlations were to memory performance found in aMCI group. CBF-connectivity in left precentral gyrus, left middle cingulate gyrus, and right middle frontal gyrus were significantly altered. All of the significant clusters had good discriminatory ability.

Conclusions

Normalized CBF as measured by ASL revealed different patterns of perfusion between aMCI and PD-MCI, which were probably linked to distinct neural mechanisms. The present study indicates that z-CBF can provide specific perfusion information for further pathological and neuropsychological studies.

Analysis of Features of Alzheimer's Disease: Detection of Early Stage from Functional Brain Changes in Magnetic Resonance Images Using a Finetuned ResNet18 Network

One of the first signs of Alzheimer's disease (AD) is mild cognitive impairment (MCI), in which there are small variants of brain changes among the intermediate stages. Although there has been an increase in research into the diagnosis of AD in its early levels of developments lately, brain changes, and their complexity for functional magnetic resonance imaging (fMRI), makes early detection of AD difficult. This paper proposes a deep learning-based method that can predict MCI, early MCI (EMCI), late MCI (LMCI), and AD. The Alzheimer's Disease Neuroimaging Initiative (ADNI) fMRI dataset consisting of 138 subjects was used for evaluation. The finetuned ResNet18 network achieved a classification accuracy of 99.99%, 99.95%, and 99.95% on EMCI vs. AD, LMCI vs. AD, and MCI vs. EMCI classification scenarios, respectively. The proposed model performed better than other known models in terms of accuracy, sensitivity, and specificity.

Parametric estimation of reference signal intensity in the quantification of amyloid-beta deposition: an 18F-AV-45 study

BACKGROUND

Positron emission tomography (PET) with the radiotracer florbetapir (18F-AV-45) allows the pathophysiology of Alzheimer's disease (AD) to be tracked in vivo. The semi-quantification of amyloid-beta (Aβ) has been extensively evaluated with the standardized uptake value ratio (SUVR) but is susceptible to disturbance from the candidate reference region and the partial volume effect (PVE). In the present study, we applied the parametric estimation of reference signal intensity (PERSI) method to 18F-AV-45 PET images for intensity normalization.

METHODS

We enrolled 479 people with 18F-AV-45 images from the Alzheimer's Disease Neuroimaging Initiative database: 261 healthy controls (HCs), 102 patients with mild cognitive impairment (MCI), and 116 AD patients. We used white matter post-processed by PERSI (PERSI-WM) as the reference region and compared our proposed method with the traditional method for semi-quantification. SUVRs were calculated for eight regions of interest: the frontal lobe, the parietal lobe, the temporal lobe, the occipital lobe, the anterior cingulate cortex, the posterior cingulate cortex, the precuneus, and the global cortex. The SUVRs derived from PERSI-WM and other reference regions were evaluated by effect size and receiver-operator characteristic curve analyses.

RESULTS

The SUVRs derived from PERSI-WM showed significantly higher trace retention in the frontal, parietal, temporal, and occipital lobes, as well as in the anterior cingulate, posterior cingulate, precuneus, and global cortex in the AD Aβ-positive (+) group (mean: +43.3%±5.4%, P<0.01) and MCI Aβ+ group (mean: +29.6%±5.3%, P<0.01). For the global cortex, PERSI-WM had the greatest Cohen's d effect size compared with the HC Aβ-negative (-) group (AD Aβ+ and MCI Aβ+: 3.02, AD Aβ+: 3.56, MCI Aβ+: 2.34), and the highest area under the curve (AUC) between the HC Aβ- and AD Aβ+ groups (AUC: 0.983, 95% confidence interval: 0.978-0.998).

CONCLUSIONS

PERSI-WM could mitigate the influence of PVE and improve the semi-quantification of 18F-AV-45 images; therefore, it could be used for large-scale clinical application in the nuclear medicine domain.