Predicting progression of amnesic MCI: The integration of episodic memory impairment with perfusion SPECT

Neuroimaging and machine learning for studying the pathways from mild cognitive impairment to alzheimer's disease: a systematic review

BACKGROUND

This systematic review synthesizes the most recent neuroimaging procedures and machine learning approaches for the prediction of conversion from mild cognitive impairment to Alzheimer's disease dementia.

METHODS

We systematically searched PubMed, SCOPUS, and Web of Science databases following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) systematic review guidelines.

RESULTS

Our search returned 2572 articles, 56 of which met the criteria for inclusion in the final selection. The multimodality framework and deep learning techniques showed potential for predicting the conversion of MCI to AD dementia.

CONCLUSION

Findings of this systematic review identified that the possibility of using neuroimaging data processed by advanced learning algorithms is promising for the prediction of AD progression. We also provided a detailed description of the challenges that researchers are faced along with future research directions. The protocol has been registered in the International Prospective Register of Systematic Reviews- CRD42019133402 and published in the Systematic Reviews journal.

Neuroimaging modalities in the detection of Alzheimer's disease-associated biomarkers

Alzheimer's disease (AD) is the most common cause of dementia. Neuropathological changes in AD patients occur up to 10-20 years before the emergence of clinical symptoms. Specific diagnosis and appropriate intervention strategies are crucial during the phase of mild cognitive impairment (MCI) and AD. The detection of biomarkers has emerged as a promising tool for tracking the efficacy of potential therapies, making an early disease diagnosis, and prejudging treatment prognosis. Specifically, multiple neuroimaging modalities, including magnetic resonance imaging (MRI), positron emission tomography, optical imaging, and single photon emission-computed tomography, have provided a few potential biomarkers for clinical application. The MRI modalities described in this review include structural MRI, functional MRI, diffusion tensor imaging, magnetic resonance spectroscopy, and arterial spin labelling. These techniques allow the detection of presymptomatic diagnostic biomarkers in the brains of cognitively normal elderly people and might also be used to monitor AD disease progression after the onset of clinical symptoms. This review highlights potential biomarkers, merits, and demerits of different neuroimaging modalities and their clinical value in MCI and AD patients. Further studies are necessary to explore more biomarkers and overcome the limitations of multiple neuroimaging modalities for inclusion in diagnostic criteria for AD.

Patterns of structure-function association in normal aging and in Alzheimer's disease: Screening for mild cognitive impairment and dementia with ML regression and classification models

Background

The combined analysis of imaging and functional modalities is supposed to improve diagnostics of neurodegenerative diseases with advanced data science techniques.

Objective

To get an insight into normal and accelerated brain aging by developing the machine learning models that predict individual performance in neuropsychological and cognitive tests from brain MRI. With these models we endeavor to look for patterns of brain structure-function association (SFA) indicative of mild cognitive impairment (MCI) and Alzheimer's dementia.

Materials and methods

We explored the age-related variability of cognitive and neuropsychological test scores in normal and accelerated aging and constructed regression models predicting functional performance in cognitive tests from brain radiomics data. The models were trained on the three study cohorts from ADNI dataset-cognitively normal individuals, patients with MCI or dementia-separately. We also looked for significant correlations between cortical parcellation volumes and test scores in the cohorts to investigate neuroanatomical differences in relation to cognitive status. Finally, we worked out an approach for the classification of the examinees according to the pattern of structure-function associations into the cohorts of the cognitively normal elderly and patients with MCI or dementia.

Results

In the healthy population, the global cognitive functioning slightly changes with age. It also remains stable across the disease course in the majority of cases. In healthy adults and patients with MCI or dementia, the trendlines of performance in digit symbol substitution test and trail making test converge at the approximated point of 100 years of age. According to the SFA pattern, we distinguish three cohorts: the cognitively normal elderly, patients with MCI, and dementia. The highest accuracy is achieved with the model trained to predict the mini-mental state examination score from voxel-based morphometry data. The application of the majority voting technique to models predicting results in cognitive tests improved the classification performance up to 91.95% true positive rate for healthy participants, 86.21%-for MCI and 80.18%-for dementia cases.

Conclusion

The machine learning model, when trained on the cases of this of that group, describes a disease-specific SFA pattern. The pattern serves as a "stamp" of the disease reflected by the model.

R4Alz-Revised: A Tool Able to Strongly Discriminate 'Subjective Cognitive Decline' from Healthy Cognition and 'Minor Neurocognitive Disorder'

BACKGROUND

The diagnosis of the minor neurocognitive diseases in the clinical course of dementia before the clinical symptoms' appearance is the holy grail of neuropsychological research. The R4Alz battery is a novel and valid tool that was designed to assess cognitive control in people with minor cognitive disorders. The aim of the current study is the R4Alz battery's extension (namely R4Alz-R), enhanced by the design and administration of extra episodic memory tasks, as well as extra cognitive control tasks, towards improving the overall R4Alz discriminant validity.

METHODS

The study comprised 80 people: (a) 20 Healthy adults (HC), (b) 29 people with Subjective Cognitive Decline (SCD), and (c) 31 people with Mild Cognitive Impairment (MCI). The groups differed in age and educational level.

RESULTS

Updating, inhibition, attention switching, and cognitive flexibility tasks discriminated SCD from HC (p ≤ 0.003). Updating, switching, cognitive flexibility, and episodic memory tasks discriminated SCD from MCI (p ≤ 0.001). All the R4Alz-R's tasks discriminated HC from MCI (p ≤ 0.001). The R4Alz-R was free of age and educational level effects. The battery discriminated perfectly SCD from HC and HC from MCI (100% sensitivity-95% specificity and 100% sensitivity-90% specificity, respectively), whilst it discriminated excellently SCD from MCI (90.3% sensitivity-82.8% specificity).

CONCLUSION

SCD seems to be stage a of neurodegeneration since it can be objectively evaluated via the R4Alz-R battery, which seems to be a useful tool for early diagnosis.

Episodic Memory in Amnestic Mild Cognitive Impairment (aMCI) and Alzheimer’s Disease Dementia (ADD): Using the “Doors and People” Tool to Differentiate between Early aMCI—Late aMCI—Mild ADD Diagnostic Groups

Episodic memory is the type of memory that allows the recollection of personal experiences containing information on what has happened and, also, where and when it happened. Because of its sensitivity to neurodegenerative diseases and the aging of the brain, it is considered a hallmark of Alzheimer’s disease dementia (ADD). The objective of the present study was to examine episodic memory in amnestic mild cognitive impairment (aMCI) and ADD. Patients with the diagnosis of early aMCI, late aMCI, and mild ADD were evaluated using the Doors and People tool which consists of four subtests examining different aspects of episodic memory. The statistical analysis with receiver operating characteristic curves (ROC) showed the discriminant potential and the cutoffs of every subtest. Overall, the evaluation of episodic memory with the Doors and People tool can discriminate with great sensitivity between the different groups of people with AD and, especially, early aMCI, late aMCI, and mild ADD patients.

Chronic hypoperfusion due to intracranial large artery stenosis is not associated with cerebral β-amyloid deposition and brain atrophy

Background:

Insufficient cerebral perfusion is suggested to play a role in the development of Alzheimer disease (AD). However, there is a lack of direct evidence indicating whether hypoperfusion causes or aggravates AD pathology. We investigated the effect of chronic cerebral hypoperfusion on AD-related pathology in humans.

Methods:

We enrolled a group of cognitively normal patients (median age: 64 years) with unilateral chronic cerebral hypoperfusion. Regions of interest with the most pronounced hypoperfusion changes were chosen in the hypoperfused region and were then mirrored in the contralateral hemisphere to create a control region with normal perfusion. ¹¹C-Pittsburgh compound-positron emission tomography standard uptake ratios and brain atrophy indices were calculated from the computed tomography images of each patient.

Results:

The median age of the 10 participants, consisting of 4 males and 6 females, was 64 years (47–76 years). We found that there were no differences in standard uptake ratios of the cortex (volume of interest [VOI]: P = 0.721, region of interest [ROI]: P = 0.241) and grey/white ratio (VOI: P = 0.333, ROI: P = 0.445) and brain atrophy indices (Bicaudate, Bifrontal, Evans, Cella, Cella media, and Ventricular index, P > 0.05) between the hypoperfused regions and contralateral normally perfused regions in patients with unilateral chronic cerebral hypoperfusion.

Conclusion:

Our findings suggest that chronic hypoperfusion due to large vessel stenosis may not directly induce cerebral β-amyloid deposition and neurodegeneration in humans.

The Destiny of Multiple Domain Amnesic Mild Cognitive Impairment: Effect of Alternative Neuropsychological Definitions and Their Adjunctive Role in Respect of Memory Impairment

OBJECTIVE

Mild cognitive impairment is the main risk factor of dementia. Previous evidence has claimed that subjects with memory disturbances associated with impairment of other cognitive domains (multiple domain amnesic MCI) are at the highest risk of developing dementia. To date, a shared definition of amnesic MCI multiple domain (aMCI-MD) is still lacking.

METHOD

163 subjects with aMCI were enrolled and followed-up for 2 years. They underwent a baseline comprehensive neuropsychological battery. The cut-off point for each test was set at 1, 1.5, and 2 SD below the mean obtained in normative studies; aMCI-MD was defined as the occurrence of abnormal scores on at least one, two, or three tests not assessing memory. The Episodic Memory Score (EMS), that measures the severity of memory impairment, was determined. Logistic regressionand Cox's proportional hazard risk models were carried out. The adjunctive effect of the definitions of aMCI-MD on the severity of memory impairment was assessed.

RESULTS

Fifty-four subjects progressed to dementia. Only restrictive definitions of aMCI-MD (at least three tests below 1.5 SD; at least two tests below 2 SD) predicted conversion to dementia in both logistic regression and survival analysis. None of the conditions showed a significant adjunctive effect on the EMS.

CONCLUSIONS

The predictive effect of impairment in tests assessing cognitive domains other than memory depends on its psychometric definition. The use of a restrictive definition would be of some usefulness, but the adjunctive effect of such a definition on an integrated analysis of memory impairment may be questionable.

High-quality brain perfusion SPECT images may be achieved with a high-speed recording using 360° CZT camera

Objective

The aim of this study was to compare brain perfusion SPECT obtained from a 360° CZT and a conventional Anger camera.

Methods

The 360° CZT camera utilizing a brain configuration, with 12 detectors surrounding the head, was compared to a 2-head Anger camera for count sensitivity and image quality on 30-min SPECT recordings from a brain phantom and from ^99mTc-HMPAO brain perfusion in 2 groups of 21 patients investigated with the CZT and Anger cameras, respectively. Image reconstruction was adjusted according to image contrast for each camera.

Results

The CZT camera provided more than 2-fold increase in count sensitivity, as compared with the Anger camera, as well as (1) lower sharpness indexes, giving evidence of higher spatial resolution, for both peripheral/central brain structures, with respective median values of 5.2%/3.7% versus 2.4%/1.9% for CZT and Anger camera respectively in patients (p < 0.01), and 8.0%/6.9% versus 6.2%/3.7% on phantom; and (2) higher gray/white matter contrast on peripheral/central structures, with respective ratio median values of 1.56/1.35 versus 1.11/1.20 for CZT and Anger camera respectively in patients (p < 0.05), and 2.57/2.17 versus 1.40/1.12 on phantom; and (3) no change in noise level. Image quality, scored visually by experienced physicians, was also significantly higher on CZT than on the Anger camera (+ 80%, p < 0.01), and all these results were unchanged on the CZT images obtained with only a 15 min recording time.

Conclusion

The 360° CZT camera provides brain perfusion images of much higher quality than a conventional Anger camera, even with high-speed recordings, thus demonstrating the potential for repositioning brain perfusion SPECT to the forefront of brain imaging.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40658-020-00334-7.

Episodic memory in mild cognitive impairment inversely correlates with the global modularity of the cerebral blood flow network

Cerebral blood flow (CBF) SPECT is an interesting methodology to study brain connectivity in mild cognitive impairment (MCI) since it is accessible worldwide and can be used as a biomarker of neuronal injury in MCI. In CBF SPECT, connectivity is grounded in group-based correlation networks. Therefore, topological metrics derived from the CBF correlation network cannot be used to support diagnosis and prognosis individually. However, methods to extract the individual patient contribution to topological metrics of group-based correlation networks were developed although not yet applied to MCI patients. Here, we investigate whether the episodic memory of 24 amnestic MCI patients correlates with individual patient contributions to topological metrics of the CBF correlation network. We first compared topological metrics of the MCI group network with the network corresponding to 26 controls. Metrics that showed significant differences were then used for the individual patient contribution analysis. We found that the global network modularity was increased while global efficiency decreased in the MCI network compared to the control. Most importantly, we found that episodic memory inversely correlates with the patient contribution to the global network modularity, which highlights the potential of this approach to develop a CBF connectivity-based biomarker at the individual level.