Identifying Mild Cognitive Impairment with Random Forest by Integrating Multiple MRI Morphological Metrics

Choroid plexus enlargement in mild cognitive impairment on MRI: a large cohort study

OBJECTIVES

Previous studies have shown possible choroid plexus (CP) dysfunction in Alzheimer's disease (AD) and highlighted CP enlargement on magnetic resonance imaging (MRI) as a predictive factor of AD. However, few studies have assessed the relationship between CP volume (CPV) and mild cognitive impairment (MCI). In this large elderly population study, we investigated the changes in CPV in patients with MCI using MRI above 65 years.

METHODS

This cross-sectional study included 2144 participants (median age, 69 years; 60.9% females) who underwent 3T MRI; they were grouped as 218 MCI participants and 1904 cognitively healthy controls. The total intracranial volume (ICV), total brain volume (TBV), CPV, hippocampal volume (HV), and lateral ventricle volume (LVV) were calculated.

RESULTS

CPV/ICV was a significant independent predictor of MCI (p < 0.01) after adjusting for potential confounders (age, sex, hypertension, hyperlipidemia, diabetes, and education level). The CPV/ICV ratio was also a significant independent predictor of MCI after adjusting for the TBV/ICV ratio (p = 0.022) or HV/ICV ratio (p = 0.017), in addition to potential confounders. The CPV was significantly correlated with the LVV (r = 0.97, p < 0.01).

CONCLUSION

We identified a relationship between CPV and MCI, which could not be explained by the degree of brain atrophy. Our results support CP dysfunction in MCI.

CLINICAL RELEVANCE STATEMENT

Choroid plexus volume measurement may serve as a valuable imaging biomarker for diagnosing and monitoring mild cognitive impairment. The enlargement of the choroid plexus, independent of brain atrophy, suggests its potential role in mild cognitive impairment pathology.

KEY POINTS

• The study examines choroid plexus volume in relation to cognitive decline in elderly. • Enlarged choroid plexus volume independently indicates mild cognitive impairment presence. • Choroid plexus volume could be a specific biomarker for early mild cognitive impairment diagnosis.

Cortical and Subcortical Gray Matter Abnormalities in Mild Cognitive Impairment

Gray matter changes are thought to be closely related to cognitive decline in mild cognitive impairment (MCI) patients. The study aimed to explore cortical and subcortical structural alterations in MCI and their association with cognitive assessment. 24 MCI patients and 22 normal controls (NCs) were included. Voxel-based morphometry (VBM), vertex-based shape analysis and surface-based morphometry (SBM) analysis were applied to explore subcortical nuclei volume, shape and cortical morphology. Correlations between structural changes and cognition were explored using spearman correlation analysis. Support vector machine (SVM) classification evaluated MCI identification accuracy. MCI patients showed significant atrophy in the left thalamus, left hippocampus, left amygdala, right pallidum, right hippocampus, along with inward deformation in the left amygdala. SBM analysis revealed that MCI group exhibited shallower sulci depth in the left hemisphere and increased cortical gyrification index (GI) in the right frontal gyrus. Correlation analysis showed the positive correlation between right hippocampus volume and episodic memory, while negative correlation between the altered GI and memory performance in MCI group. SVM analysis demonstrated superior performance of sulci depth and GI derived from SBM in MCI identification. When combined with cortical and subcortical metrics, SVM achieved a peak accuracy of 89% in distinguishing MCI from NC. The study reveals significant gray matter structural changes in MCI, suggesting their potential role in underlying functional differences and neural mechanisms behind memory impairment in MCI.

Identifying Leukoaraiosis with Mild Cognitive Impairment by Fusing Multiple MRI Morphological Metrics and Ensemble Machine Learning

Leukoaraiosis (LA) is strongly associated with impaired cognition and increased dementia risk. Determining effective and robust methods of identifying LA patients with mild cognitive impairment (LA-MCI) is important for clinical intervention and disease monitoring. In this study, an ensemble learning method that combines multiple magnetic resonance imaging (MRI) morphological features is proposed to distinguish LA-MCI patients from LA patients lacking cognitive impairment (LA-nCI). Multiple comprehensive morphological measures (including gray matter volume (GMV), cortical thickness (CT), surface area (SA), cortical volume (CV), sulcus depth (SD), fractal dimension (FD), and gyrification index (GI)) are extracted from MRI to enrich model training on disease characterization information. Then, based on the general extreme gradient boosting (XGBoost) classifier, we leverage a weighted soft-voting ensemble framework to ensemble a data-level resampling method (Fusion + XGBoost) and an algorithm-level focal loss (FL)-improved XGBoost model (FL-XGBoost) to overcome class-imbalance learning problems and provide superior classification performance and stability. The baseline XGBoost model trained on an original imbalanced dataset had a balanced accuracy (Bacc) of 78.20%. The separate Fusion + XGBoost and FL-XGBoost models achieved Bacc scores of 80.53 and 81.25%, respectively, which are clear improvements (i.e., 2.33% and 3.05%, respectively). The fused model distinguishes LA-MCI from LA-nCI with an overall accuracy of 84.82%. Sensitivity and specificity were also well improved (85.50 and 84.14%, respectively). This improved model has the potential to facilitate the clinical diagnosis of LA-MCI.

A radiomics-based brain network in T1 images: construction, attributes, and applications

T1 image is a widely collected imaging sequence in various neuroimaging datasets, but it is rarely used to construct an individual-level brain network. In this study, a novel individualized radiomics-based structural similarity network was proposed from T1 images. In detail, it used voxel-based morphometry to obtain the preprocessed gray matter images, and radiomic features were then extracted on each region of interest in Brainnetome atlas, and an individualized radiomics-based structural similarity network was finally built using the correlational values of radiomic features between any pair of regions of interest. After that, the network characteristics of individualized radiomics-based structural similarity network were assessed, including graph theory attributes, test-retest reliability, and individual identification ability (fingerprinting). At last, two representative applications for individualized radiomics-based structural similarity network, namely mild cognitive impairment subtype discrimination and fluid intelligence prediction, were exemplified and compared with some other networks on large open-source datasets. The results revealed that the individualized radiomics-based structural similarity network displays remarkable network characteristics and exhibits advantageous performances in mild cognitive impairment subtype discrimination and fluid intelligence prediction. In summary, the individualized radiomics-based structural similarity network provides a distinctive, reliable, and informative individualized structural brain network, which can be combined with other networks such as resting-state functional connectivity for various phenotypic and clinical applications.

Identifying Alzheimer's disease and mild cognitive impairment with atlas-based multi-modal metrics

Introduction

Multi-modal neuroimaging metrics in combination with advanced machine learning techniques have attracted more and more attention for an effective multi-class identification of Alzheimer's disease (AD), mild cognitive impairment (MCI) and health controls (HC) recently.

Methods

In this paper, a total of 180 subjects consisting of 44 AD, 66 MCI and 58 HC subjects were enrolled, and the multi-modalities of the resting-state functional magnetic resonance imaging (rs-fMRI) and the structural MRI (sMRI) for all participants were obtained. Then, four kinds of metrics including the Hurst exponent (HE) metric and bilateral hippocampus seed independently based connectivity metrics generated from fMRI data, and the gray matter volume (GMV) metric obtained from sMRI data, were calculated and extracted in each region of interest (ROI) based on a newly proposed automated anatomical Labeling (AAL3) atlas after data pre-processing. Next, these metrics were selected with a minimal redundancy maximal relevance (MRMR) method and a sequential feature collection (SFC) algorithm, and only a subset of optimal features were retained after this step. Finally, the support vector machine (SVM) based classification methods and artificial neural network (ANN) algorithm were utilized to identify the multi-class of AD, MCI and HC subjects in single modal and multi-modal metrics respectively, and a nested ten-fold cross-validation was utilized to estimate the final classification performance.

Results

The results of the SVM and ANN based methods indicated the best accuracies of 80.36 and 74.40%, respectively, by utilizing all the multi-modal metrics, and the optimal accuracies for AD, MCI and HC were 79.55, 78.79 and 82.76%, respectively, in the SVM based method. In contrast, when using single modal metric, the SVM based method obtained a best accuracy of 72.62% with the HE metric, and the accuracies for AD, MCI and HC subjects were just 56.82, 80.30 and 75.86%, respectively. Moreover, the overlapping abnormal brain regions detected by multi-modal metrics were mainly located at posterior cingulate gyrus, superior frontal gyrus and cuneus.

Conclusion

Taken together, the SVM based method with multi-modal metrics could provide effective diagnostic information for identifying AD, MCI and HC subjects.

Predicting mild cognitive impairment progression to Alzheimer's disease based on machine learning analysis of cortical morphological features

PURPOSE

To establish a model for predicting mild cognitive impairment (MCI) progression to Alzheimer's disease (AD) using morphological features extracted from a joint analysis of voxel-based morphometry (VBM) and surface-based morphometry (SBM).

METHODS

We analyzed data from 121 MCI patients from the Alzheimer's Disease Neuroimaging Initiative, 32 of whom progressed to AD during a 4-year follow-up period and were classified as the progression group, while the remaining 89 were classified as the non-progression group. Patients were divided into a training set (n = 84) and a testing set (n = 37). Morphological features measured by VBM and SBM were extracted from the cortex of the training set and dimensionally reduced to construct morphological biomarkers using machine learning methods, which were combined with clinical data to build a multimodal combinatorial model. The model's performance was evaluated using receiver operating characteristic curves on the testing set.

RESULTS

The Alzheimer's Disease Assessment Scale (ADAS) score, apolipoprotein E (APOE4), and morphological biomarkers were independent predictors of MCI progression to AD. The combinatorial model based on the independent predictors had an area under the curve (AUC) of 0.866 in the training set and 0.828 in the testing set, with sensitivities of 0.773 and 0.900 and specificities of 0.903 and 0.747, respectively. The number of MCI patients classified as high-risk for progression to AD was significantly different from those classified as low-risk in the training set, testing set, and entire dataset, according to the combinatorial model (P < 0.05).

CONCLUSION

The combinatorial model based on cortical morphological features can identify high-risk MCI patients likely to progress to AD, potentially providing an effective tool for clinical screening.

Alteration of the cortical morphology in classical trigeminal neuralgia: voxel-, deformation-, and surface-based analysis

OBJECTIVE

This study aimed to combine voxel-based morphometry, deformation-based morphometry, and surface-based morphometry to analyze gray matter volume and cortex shape in classical trigeminal neuralgia patients.

METHODS

This study included 79 classical trigeminal neuralgia patients and age- and sex-matched 81 healthy controls. The aforementioned three methods were used to analyze brain structure in classical trigeminal neuralgia patients. Spearman correlation analysis was used to analyze the correlation of brain structure with the trigeminal nerve and clinical parameters.

RESULTS

The bilateral trigeminal nerve was atrophied, and the ipsilateral trigeminal nerve volume was smaller than the contralateral volume in the classical trigeminal neuralgia. The gray matter volume of Temporal_Pole_Sup_R and Precentral_R was found to be decreased using voxel-based morphometry. The gray matter volume of Temporal_Pole_Sup_R had a positive correlation with disease duration and a negative correlation with the cross-section area of the compression point and the quality-of-life score in trigeminal neuralgia. The gray matter volume of Precentral_R was negatively correlated with the ipsilateral volume of the trigeminal nerve cisternal segment, cross-section area of compression point, and visual analogue scale. The gray matter volume of Temporal_Pole_Sup_L was found to be increased using deformation-based morphometry and had a negative correlation with the self-rating anxiety scale. The gyrification of the middle temporal gyrus_L increased and the Postcentral_L thickness decreased, as detected using surface-based morphometry.

CONCLUSIONS

The gray matter volume and cortical morphology of pain-related brain regions were correlated with clinical and trigeminal nerve parameters. voxel-based morphometry, deformation-based morphometry, and surface-based morphometry complemented each other in analyzing the brain structures of patients with classical trigeminal neuralgia and provided a basis for studying the pathophysiology of classical trigeminal neuralgia.

Hierarchical multi-class Alzheimer’s disease diagnostic framework using imaging and clinical features

Due to the clinical continuum of Alzheimer’s disease (AD), the accuracy of early diagnostic remains unsatisfactory and warrants further research. The objectives of this study were: (1) to develop an effective hierarchical multi-class framework for clinical populations, namely, normal cognition (NC), early mild cognitive impairment (EMCI), late mild cognitive impairment (LMCI), and AD, and (2) to explore the geometric properties of cognition-related anatomical structures in the cerebral cortex. A total of 1,670 participants were enrolled in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, comprising 985 participants (314 NC, 208 EMCI, 258 LMCI, and 205 AD) in the model development set and 685 participants (417 NC, 110 EMCI, 83 LMCI, and 75 AD) after 2017 in the temporal validation set. Four cortical geometric properties for 148 anatomical structures were extracted, namely, cortical thickness (CTh), fractal dimension (FD), gyrification index (GI), and sulcus depth (SD). By integrating these imaging features with Mini-Mental State Examination (MMSE) scores at four-time points after the initial visit, we identified an optimal subset of 40 imaging features using the temporally constrained group sparse learning method. The combination of selected imaging features and clinical variables improved the multi-class performance using the AdaBoost algorithm, with overall accuracy rates of 0.877 in the temporal validation set. Clinical Dementia Rating (CDR) was the primary clinical variable associated with AD-related populations. The most discriminative imaging features included the bilateral CTh of the dorsal part of the posterior cingulate gyrus, parahippocampal gyrus (PHG), parahippocampal part of the medial occipito-temporal gyrus, and angular gyrus, the GI of the left inferior segment of the insula circular sulcus, and the CTh and SD of the left superior temporal sulcus (STS). Our hierarchical multi-class framework underscores the utility of combining cognitive variables with imaging features and the reliability of surface-based morphometry, facilitating more accurate early diagnosis of AD in clinical practice.

A Multi-Modal and Multi-Atlas Integrated Framework for Identification of Mild Cognitive Impairment

Background: Multi-modal neuroimaging with appropriate atlas is vital for effectively differentiating mild cognitive impairment (MCI) from healthy controls (HC). Methods: The resting-state functional magnetic resonance imaging (rs-fMRI) and structural MRI (sMRI) of 69 MCI patients and 61 HC subjects were collected. Then, the gray matter volumes obtained from the sMRI and Hurst exponent (HE) values calculated from rs-fMRI data in the Automated Anatomical Labeling (AAL-90), Brainnetome (BN-246), Harvard–Oxford (HOA-112) and AAL3-170 atlases were extracted, respectively. Next, these characteristics were selected with a minimal redundancy maximal relevance algorithm and a sequential feature collection method in single or multi-modalities, and only the optimal features were retained after this procedure. Lastly, the retained characteristics were served as the input features for the support vector machine (SVM)-based method to classify MCI patients, and the performance was estimated with a leave-one-out cross-validation (LOOCV). Results: Our proposed method obtained the best 92.00% accuracy, 94.92% specificity and 89.39% sensitivity with the sMRI in AAL-90 and the fMRI in HOA-112 atlas, which was much better than using the single-modal or single-atlas features. Conclusion: The results demonstrated that the multi-modal and multi-atlas integrated method could effectively recognize MCI patients, which could be extended into various neurological and neuropsychiatric diseases.

A Reparametrized CNN Model to Distinguish Alzheimer's Disease Applying Multiple Morphological Metrics and Deep Semantic Features From Structural MRI

It is of potential clinical value to improve the accuracy of Alzheimer's disease (AD) recognition using structural MRI. We proposed a reparametrized convolutional neural network (Re-CNN) to discriminate AD from NC by applying morphological metrics and deep semantic features. The deep semantic features were extracted through Re-CNN on structural MRI. Considering the high redundancy in deep semantic features, we constrained the similarity of the features and retained the most distinguishing features utilizing the reparametrized module. The Re-CNN model was trained in an end-to-end manner on structural MRI from the ADNI dataset and tested on structural MRI from the AIBL dataset. Our proposed model achieves better performance over some existing structural MRI-based AD recognition models. The experimental results show that morphological metrics along with the constrained deep semantic features can relatively improve AD recognition performance. Our code is available at: https://github.com/czp19940707/Re-CNN.

Fractal dimension of the brain in neurodegenerative disease and dementia: A systematic review

Sensitive and specific antemortem biomarkers of neurodegenerative disease and dementia are crucial to the pursuit of effective treatments, required both to reliably identify disease and to track its progression. Atrophy is the structural magnetic resonance imaging (MRI) hallmark of neurodegeneration. However in most cases it likely indicates a relatively advanced stage of disease less susceptible to treatment as some disease processes begin decades prior to clinical onset. Among emerging metrics that characterise brain shape rather than volume, fractal dimension (FD) quantifies shape complexity. FD has been applied in diverse fields of science to measure subtle changes in elaborate structures. We review its application thus far to structural MRI of the brain in neurodegenerative disease and dementia. We identified studies involving subjects who met criteria for mild cognitive impairment, Alzheimer's Disease, Vascular Dementia, Lewy Body Dementia, Frontotemporal Dementia, Amyotrophic Lateral Sclerosis, Parkinson's Disease, Huntington's Disease, Multiple Systems Atrophy, Spinocerebellar Ataxia and Multiple Sclerosis. The early literature suggests that neurodegenerative disease processes are usually associated with a decline in FD of the brain. The literature includes examples of disease-related change in FD occurring independently of atrophy, which if substantiated would represent a valuable advantage over other structural imaging metrics. However, it is likely to be non-specific and to exhibit complex spatial and temporal patterns. A more harmonious methodological approach across a larger number of studies as well as careful attention to technical factors associated with image processing and FD measurement will help to better elucidate the metric's utility.

Prediction of Early Alzheimer Disease by Hippocampal Volume Changes under Machine Learning Algorithm

This research was aimed at discussing the application value of different machine learning algorithms in the prediction of early Alzheimer's disease (AD), which was based on hippocampal volume changes in magnetic resonance imaging (MRI). In the research, the 84 cases in American Alzheimer's disease neuroimaging initiative (ADNI) database were selected as the research data. Based on the scoring results of cognitive function, all cases were divided into three groups, including cognitive function normal (normal group), early mild cognitive impairment (e-MCI group), and later mild cognitive impairment (l-MCI group) groups. Each group included 28 cases. The features of hippocampal volume changes in MRI images of the patients in different groups were extracted. The samples of training set and test set were established. Besides, the established support vector machine (SVM), decision tree (DT), and random forest (RF) prediction models were used to predict e-MCI. Metalinear regression was utilized to analyze MRI feature data, and the predictive accuracy, sensitivity, and specificity of different models were calculated. The result showed that the volumes of hippocampal left CA1, left CA2-3, left CA4-DG, left presubiculum, left tail, right CA2-3, right CA4-DG, right presubiculum, and right tail in e-MCI group were all smaller than those in normal group (P < 0.01). The corresponding volume of hippocampal subregions in l-MCI group was remarkably reduced compared with that in normal group (P < 0.001). The volumes of regions left CA1, left CA2-3, left CA4-DG, right CA2-3, right CA4-DG, and right presubiculum were all positively correlated with logical memory test-delay recall (LMT-DR) score (R 2 = 0.1702, 0.3779, 0.1607, 0.1620, 0.0426, and 0.1309; P < 0.001). The predictive accuracy of training set sample by DT, SVM, and RF was 86.67%, 93.33%, and 98.33%, respectively. Based on the changes in the volumes of left CA4-DG, right CA2-3, and right CA4-DG, the predictive accuracy of e-MCI and l-MCI by RF model was both higher than those by DT model (P < 0.01). Besides, the predictive accuracy, sensitivity, and specificity of e-MCI by RF model was all notably higher than those by DT model (P < 0.01). The above results demonstrated that the effective early AD prediction models were established by the volume changes in hippocampal subregions, which was based on RF in the research. The establishment of early AD prediction models offered certain reference basis to the diagnosis and treatment of AD patients.

Using the Alzheimer's Disease Neuroimaging Initiative to improve early detection, diagnosis, and treatment of Alzheimer's disease

INTRODUCTION

The Alzheimer's Disease Neuroimaging Initiative (ADNI) has accumulated 15 years of clinical, neuroimaging, cognitive, biofluid biomarker and genetic data, and biofluid samples available to researchers, resulting in more than 3500 publications. This review covers studies from 2018 to 2020.

METHODS

We identified 1442 publications using ADNI data by conventional search methods and selected impactful studies for inclusion.

RESULTS

Disease progression studies supported pivotal roles for regional amyloid beta (Aβ) and tau deposition, and identified underlying genetic contributions to Alzheimer's disease (AD). Vascular disease, immune response, inflammation, resilience, and sex modulated disease course. Biologically coherent subgroups were identified at all clinical stages. Practical algorithms and methodological changes improved determination of Aβ status. Plasma Aβ, phosphorylated tau181, and neurofilament light were promising noninvasive biomarkers. Prognostic and diagnostic models were externally validated in ADNI but studies are limited by lack of ethnocultural cohort diversity.

DISCUSSION

ADNI has had a profound impact in improving clinical trials for AD.

Human immune deficiency virus-related structural alterations in the brain are dependent on age

Currently, it is still unknown whether human immune deficiency virus (HIV)‐related structural alterations in the brain are dependent on age. With people living with HIV at different ages, we aim to investigate age‐specific structural alterations in HIV patients. Eighty‐three male HIV patients and eighty‐three age‐matched male controls were enrolled, and high‐resolution T1 weighted images were collected and analyzed with four morphological metrics. Then, statistical analyses were respectively conducted to ascertain HIV effects, age effects, and medication effects in brain structure of HIV patients, and the relationship with neuropsychological evaluations were further explored. Finally, discriminative performances of these structural abnormalities were quantitatively testified with three machine learning models. Compared with healthy controls, HIV patients displayed lower gray matter volumes (GMV), lower gyrification index, deeper sulcus depth, and larger cortical thickness (CTH). Age‐specific differences were found in GMV and CTH: young‐aged HIV patients displayed more obvious morphological alterations than middle‐aged HIV patients when comparing corresponding age‐matched healthy controls. Furthermore, age‐specific long‐term medication effect of combination antiretroviral therapy were also presented. Additionally, several subcortical structural changes were negatively associated with language, attention and motor functions. Finally, three machine learning models demonstrated young‐aged HIV patients were easier to be recognized than middle‐aged HIV patients. Our study indicated young‐aged HIV patients were more vulnerable to HIV infection in brain structure than middle‐aged patients, and future studies should not ignore the age effect in studying the HIV‐related abnormalities.