Hippocampal subfield surface deformity in non-semantic primary progressive aphasia

Diagnostic performance of MRI radiomics for classification of Alzheimer's disease, mild cognitive impairment, and normal subjects: a systematic review and meta-analysis

BACKGROUND

Alzheimer's disease (AD) is a debilitating neurodegenerative disease. Early diagnosis of AD and its precursor, mild cognitive impairment (MCI), is crucial for timely intervention and management. Radiomics involves extracting quantitative features from medical images and analyzing them using advanced computational algorithms. These characteristics have the potential to serve as biomarkers for disease classification, treatment response prediction, and patient stratification. Of note, Magnetic resonance imaging (MRI) radiomics showed a promising result for diagnosing and classifying AD, and MCI from normal subjects. Thus, we aimed to systematically evaluate the diagnostic performance of the MRI radiomics for this task.

METHODS AND MATERIALS

A comprehensive search of the current literature was conducted using relevant keywords in PubMed/MEDLINE, Embase, Scopus, and Web of Science databases from inception to August 5, 2023. Original studies discussing the diagnostic performance of MRI radiomics for the classification of AD, MCI, and normal subjects were included. Method quality was evaluated with the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) and the Radiomics Quality Score (RQS) tools.

RESULTS

We identified 13 studies that met the inclusion criteria, involving a total of 5448 participants. The overall quality of the included studies was moderate to high. The pooled sensitivity and specificity of MRI radiomics for differentiating AD from normal subjects were 0.92 (95% CI [0.85; 0.96]) and 0.91 (95% CI [0.85; 0.95]), respectively. The pooled sensitivity and specificity of MRI radiomics for differentiating MCI from normal subjects were 0.74 (95% CI [0.60; 0.85]) and 0.79 (95% CI [0.70; 0.86]), respectively. Also, the pooled sensitivity and specificity of MRI radiomics for differentiating AD from MCI were 0.73 (95% CI [0.64; 0.80]) and 0.79 (95% CI [0.64; 0.90]), respectively.

CONCLUSION

MRI radiomics has promising diagnostic performance in differentiating AD, MCI, and normal subjects. It can potentially serve as a non-invasive and reliable tool for early diagnosis and classification of AD and MCI.

The unique effect of TDP-43 on hippocampal subfield morphometry and cognition

•We explored postmortem TDP-43 burden and antemortem hippocampal surface deformation. •TDP-43 was uniquely associated with inward deformation in the hippocampus. •Deformation patterns account for co-existing disease showing TDP-43′s unique effect. •Deformation was significantly correlated with cognition scores.

Verbal and Nonverbal Memory in Neurodegenerative and Stroke Aphasia: Evidence From the Turkish Version of the Three Words Three Shapes Test

BACKGROUND

Although language impairment is the most salient feature of cognitive impairment in both primary progressive aphasia (PPA) and stroke aphasia (SA), memory can also be impaired in both patient populations.

OBJECTIVE

To identify distinctive features of verbal and nonverbal memory processing in individuals with PPA and those with SA.

METHOD

We gave individuals with PPA (n = 14), those with SA (n = 8), and healthy controls (HC; n = 13) a comprehensive neuropsychological test battery and the Turkish version of the Three Words Three Shapes Test (3W3S-Turkish). The 3W3S-Turkish Test includes five subtests: Copy, Incidental Recall, Acquisition, Delayed Recall, and Recognition. High-resolution brain scans were performed in a subset of individuals with PPA and those with SA. Lesion distribution was limited to the dorsal language areas in the SA group, whereas peak atrophy areas in the PPA group extended beyond the language network, including the medial temporal lobe, precuneus, and posterior/medial portions of the cingulate cortex.

RESULTS

Both the PPA and SA groups showed impairment in incidental recall, and the PPA group showed additional impairment in delayed recall. Greater impairment for verbal stimuli suggestive of material-specific memory impairment was evident in the PPA group's scores on the Incidental Recall and Delayed Recall subtests. Both aphasia groups retained the acquired information regardless of material type.

CONCLUSION

Although both aphasia groups shared similarities in the involvement of the dorsal prefrontal working memory/attention network, the PPA group showed greater impairment in delayed recall compared with the SA group.

Clinical, Imaging, and Pathologic Characteristics of Patients With Right vs Left Hemisphere-Predominant Logopenic Progressive Aphasia

OBJECTIVE

To assess and compare demographic, clinical, neuroimaging, and pathologic characteristics of a cohort of patients with right hemisphere-predominant vs left hemisphere-predominant logopenic progressive aphasia (LPA).

METHODS

This is a case-control study of patients with LPA who were prospectively followed at Mayo Clinic and underwent [¹⁸F]-fluorodeoxyglucose (FDG) PET scan. Patients were classified as rLPA if right temporal lobe metabolism was ≥1 SD lower than left temporal lobe metabolism. Patients with rLPA were frequency-matched 3:1 to typical left-predominant LPA based on degree of asymmetry and severity of temporal lobe metabolism. Patients were compared on clinical, imaging (MRI, FDG-PET, β-amyloid, and tau-PET), and pathologic characteristics.

RESULTS

Of 103 prospectively recruited patients with LPA, 8 (4 female) were classified as rLPA (7.8%); all patients with rLPA were right-handed. Patients with rLPA had milder aphasia based on the Western Aphasia Battery-Aphasia Quotient (p = 0.04) and less frequent phonologic errors (p = 0.015). Patients with rLPA had shorter survival compared to typical LPA: hazard ratio 4.0 (1.2-12.9), p = 0.02. There were no other differences in demographics, handedness, genetics, or neurologic or neuropsychological tests. Compared to the 24 frequency-matched patients with typical LPA, patients with rLPA showed greater frontotemporal hypometabolism of the nondominant hemisphere on FDG-PET and less atrophy in amygdala and hippocampus of the dominant hemisphere. Autopsy evaluation revealed a similar distribution of pathologic findings in both groups, with Alzheimer disease pathologic changes being the most frequent pathology.

CONCLUSIONS

rLPA is associated with less severe aphasia but has shorter survival from reported symptom onset than typical LPA, possibly related to greater involvement of the nondominant hemisphere.

Automated Hippocampal Subfield Volumetric Analyses in Atypical Alzheimer's Disease

BACKGROUND

Posterior cortical atrophy (PCA) and logopenic progressive aphasia (LPA) are two of the most common variants of atypical Alzheimer's disease (AD). Both PCA and LPA are associated with relative sparing of hippocampus compared to neocortex, although hippocampal atrophy is observed. It is unclear whether regional patterns of hippocampal subfield involvement differ between PCA and LPA, and whether they differ from typical AD.

OBJECTIVE

To assess volume of specific subfields of the hippocampus in PCA, LPA, and typical AD.

METHODS

Fifty-nine patients with PCA and 77 patients with LPA were recruited and underwent T1-weighted MRI and Pittsburgh Compound B (PiB) PET at Mayo Clinic. Thirty-six probable AD patients and 100 controls were identified from the Alzheimer's Disease Neuroimaging Initiative. Hippocampal subfield volumes were calculated using Freesurfer, and volumes were compared between PCA, LPA, AD, and controls using Kruskal-Wallis and Dunn tests.

RESULTS

The LPA and PCA groups both showed the most striking abnormalities in CA4, presubiculum, molecular layer of the hippocampus, molecular and granule cell layers of the dentate gyrus, and the hippocampal-amygdala transition area, although atrophy was left-sided in LPA. PCA showed smaller volume of right presubiculum compared to LPA, with trends for smaller volumes of right parasubiculum and fimbria. LPA showed a trend for smaller volumes of left CA1 compared to PCA. The AD group showed smaller volumes of the right subiculum, CA1, and presubiculum compared to LPA.

CONCLUSION

Patterns of hippocampal subfield atrophy differ across the different syndromic variants of AD.

Differences in neuroimaging features of early- versus late-onset nonfluent/agrammatic primary progressive aphasia

This study investigated distinct neuroimaging features measured by cortical thickness and subcortical structural shape abnormality in early-onset (EO, onset age <65 years) and late-onset (LO, onset age ≥65 years) nonfluent/agrammatic variant of primary progressive aphasia (nfvPPA) patients. Cortical thickness and subcortical structural shape analyses were performed using a surface-based method from 38 patients with nfvPPA and 76 cognitively normal individuals. To minimize the effects of physiological aging, we used W-scores in comparisons between the groups. The EO-nfvPPA group exhibited more extensive cortical thickness reductions predominantly in the left perisylvian, lateral and medial prefrontal, temporal, posterior cingulate, and precuneus regions than the LO-nfvPPA group. The EO-nfvPPA group also exhibited significantly greater subcortical structural shape abnormality than the LO-nfvPPA group, mainly in the left striatum, hippocampus, and amygdala. Our findings suggested that there were differences in neuroimaging features between these groups by the age of symptom onset, which might be explained by underlying heterogeneous neuropathological differences or the age-related brain reserve hypothesis.

Grant Report on PREDICT-ADFTD: Multimodal Imaging Prediction of AD/FTD and Differential Diagnosis

We report on the ongoing project "PREDICT-ADFTD: Multimodal Imaging Prediction of AD/FTD and Differential Diagnosis" describing completed and future work supported by this grant. This project is a multi-site, multi-study collaboration effort with research spanning seven sites across the US and Canada. The overall goal of the project is to study neurodegeneration within Alzheimer's Disease, Frontotemporal Dementia, and related neurodegenerative disorders, using a variety of brain imaging and computational techniques to develop methods for the early and accurate prediction of disease and its course. The overarching goal of the project is to develop the earliest and most accurate biomarker that can differentiate clinical diagnoses to inform clinical trials and patient care. In its third year, this project has already completed several projects to achieve this goal, focusing on (1) structural MRI (2) machine learning and (3) FDG-PET and multimodal imaging. Studies utilizing structural MRI have identified key features of underlying pathology by studying hippocampal deformation that is unique to clinical diagnosis and also post-mortem confirmed neuropathology. Several machine learning experiments have shown high classification accuracy in the prediction of disease based on Convolutional Neural Networks utilizing MRI images as input. In addition, we have also achieved high accuracy in predicting conversion to DAT up to five years in the future. Further, we evaluated multimodal models that combine structural and FDG-PET imaging, in order to compare the predictive power of multimodal to unimodal models. Studies utilizing FDG-PET have shown significant predictive ability in the prediction and progression of disease.

In vivo hippocampal subfield shape related to TDP-43, amyloid beta, and tau pathologies

Despite advances in the development of biomarkers for Alzheimer's disease (AD), accurate ante-mortem diagnosis remains challenging because a variety of neuropathologic disease states can coexist and contribute to the AD dementia syndrome. Here, we report a neuroimaging study correlating hippocampal deformity with regional AD and transactive response DNA-binding protein of 43 kDA pathology burden. We used hippocampal shape analysis of ante-mortem T1-weighted structural magnetic resonance imaging images of 42 participants from two longitudinal cohort studies conducted by the Rush Alzheimer's Disease Center. Surfaces were generated for the whole hippocampus and zones approximating the underlying subfields using a previously developed automated image-segmentation pipeline. Multiple linear regression models were constructed to correlate the shape with pathology measures while accounting for covariates, with relationships mapped out onto hippocampal surface locations. A significant relationship existed between higher paired helical filaments-tau burden and inward hippocampal shape deformity in zones approximating CA1 and subiculum which persisted after accounting for coexisting pathologies. No significant patterns of inward surface deformity were associated with amyloid-beta or transactive response DNA-binding protein of 43 kDA after including covariates. Our findings indicate that hippocampal shape deformity measures in surface zones approximating CA1 may represent a biomarker for postmortem AD pathology.

Radiomic Features of Hippocampal Subregions in Alzheimer's Disease and Amnestic Mild Cognitive Impairment

Alzheimer's disease (AD) is characterized by progressive dementia, especially in episodic memory, and amnestic mild cognitive impairment (aMCI) is associated with a high risk of developing AD. Hippocampal atrophy/shape changes are believed to be the most robust magnetic resonance imaging (MRI) markers for AD and aMCI. Radiomics, a method of texture analysis, can quantitatively examine a large set of features and has previously been successfully applied to evaluate imaging biomarkers for AD. To test whether radiomic features in the hippocampus can be employed for early classification of AD and aMCI, 1692 features from the caudal and head parts of the bilateral hippocampus were extracted from 38 AD patients, 33 aMCI patients and 45 normal controls (NCs). One way analysis of variance (ANOVA) showed that 111 features exhibited statistically significant group differences (P < 0.01, Bonferroni corrected). Among these features, 98 were significantly correlated with Mini-Mental State Examination (MMSE) scores in AD and aMCI subjects (P < 0.01). The support vector machine (SVM) model demonstrated that radiomic features allowed us to distinguish AD from NC with an accuracy of 86.75% (specificity = 88.89% and sensitivity = 84.21%) and an area under curve (AUC) of 0.93. In conclusion, these findings provide evidence showing that radiomic features are beneficial in detecting early cognitive decline, and SVM classification analysis provides encouraging evidence for using hippocampal radiomic features as a potential biomarker for clinical applications in AD.

Systematic comparison of different techniques to measure hippocampal subfield volumes in ADNI2

Objective

Subfield-specific measurements provide superior information in the early stages of neurodegenerative diseases compared to global hippocampal measurements. The overall goal was to systematically compare the performance of five representative manual and automated T1 and T2 based subfield labeling techniques in a sub-set of the ADNI2 population.

Methods

The high resolution T2 weighted hippocampal images (T2-HighRes) and the corresponding T1 images from 106 ADNI2 subjects (41 controls, 57 MCI, 8 AD) were processed as follows. A. T1-based: 1. Freesurfer + Large-Diffeomorphic-Metric-Mapping in combination with shape analysis. 2. FreeSurfer 5.1 subfields using in-vivo atlas. B. T2-HighRes: 1. Model-based subfield segmentation using ex-vivo atlas (FreeSurfer 6.0). 2. T2-based automated multi-atlas segmentation combined with similarity-weighted voting (ASHS). 3. Manual subfield parcellation. Multiple regression analyses were used to calculate effect sizes (ES) for group, amyloid positivity in controls, and associations with cognitive/memory performance for each approach.

Results

Subfield volumetry was better than whole hippocampal volumetry for the detection of the mild atrophy differences between controls and MCI (ES: 0.27 vs 0.11). T2-HighRes approaches outperformed T1 approaches for the detection of early stage atrophy (ES: 0.27 vs.0.10), amyloid positivity (ES: 0.11 vs 0.04), and cognitive associations (ES: 0.22 vs 0.19).

Conclusions

T2-HighRes subfield approaches outperformed whole hippocampus and T1 subfield approaches. None of the different T2-HghRes methods tested had a clear advantage over the other methods. Each has strengths and weaknesses that need to be taken into account when deciding which one to use to get the best results from subfield volumetry.

Selective verbal recognition memory impairments are associated with atrophy of the language network in non-semantic variants of primary progressive aphasia

Primary progressive aphasia (PPA) is clinically defined by an initial loss of language function and preservation of other cognitive abilities, including episodic memory. While PPA primarily affects the left-lateralized perisylvian language network, some clinical neuropsychological tests suggest concurrent initial memory loss. The goal of this study was to test recognition memory of objects and words in the visual and auditory modality to separate language-processing impairments from retentive memory in PPA. Individuals with non-semantic PPA had longer reaction times and higher false alarms for auditory word stimuli compared to visual object stimuli. Moreover, false alarms for auditory word recognition memory were related to cortical thickness within the left inferior frontal gyrus and left temporal pole, while false alarms for visual object recognition memory was related to cortical thickness within the right-temporal pole. This pattern of results suggests that specific vulnerability in processing verbal stimuli can hinder episodic memory in PPA, and provides evidence for differential contributions of the left and right temporal poles in word and object recognition memory.

Deformed Subcortical Structures Are Related to Past HIV Disease Severity in Youth With Perinatally Acquired HIV Infection

BACKGROUND

Combination antiretroviral therapy has led to increased survival among youth with perinatally acquired HIV (PHIV). However, cognitive deficits continue to be common. Histopathological studies in adults have found HIV concentrated in subcortical structures, which are involved in sensory processing, movement, and higher-order cognition that emerges with development.

METHODS

We conducted magnetic resonance imaging and cognitive testing in 40 youth with PHIV at one site of the Adolescent Master Protocol of the Pediatric HIV/AIDS Cohort Study. We collected HIV disease-severity measures and substance-use reports. Subcortical volume and shape deformation were generated with FreeSurfer-Initiated Large Deformation Diffeomorphic Metric Mapping. Inward shape deformation was defined as negative displacement. We evaluated associations of subcortical shape deformation with past HIV severity after adjustment for sex, age at neuroimaging, age at HIV severity marker, and substance use. We examined associations between subcortical deformation and cognitive function.

RESULTS

Negative correlations between shape deformation and peak HIV viral load (VL) were found in clusters in the caudate tail, globus pallidus, lateral putamen, and anterior and medial thalamus. Positive correlations between shape deformation and nadir CD4-positive T-lymphocyte percentage (CD4%) were found in clusters in the medial and posterior thalamus. Inward deformation in caudate and thalamic clusters correlated with worse cognition.

CONCLUSIONS

Youth with PHIV have demonstrable subcortical shape deformation related to past HIV severity and cognition; inward deformation was associated with higher peak VL, lower nadir CD4%, and worse cognition. Identifying subcortical deformation may inform clinical practice for early intervention to help improve cognitive outcomes and assess the neuroefficacy of combination antiretroviral therapy in youth with PHIV.

Bayesian longitudinal segmentation of hippocampal substructures in brain MRI using subject-specific atlases

The hippocampal formation is a complex, heterogeneous structure that consists of a number of distinct, interacting subregions. Atrophy of these subregions is implied in a variety of neurodegenerative diseases, most prominently in Alzheimer's disease (AD). Thanks to the increasing resolution of MR images and computational atlases, automatic segmentation of hippocampal subregions is becoming feasible in MRI scans. Here we introduce a generative model for dedicated longitudinal segmentation that relies on subject-specific atlases. The segmentations of the scans at the different time points are jointly computed using Bayesian inference. All time points are treated the same to avoid processing bias. We evaluate this approach using over 4700 scans from two publicly available datasets (ADNI and MIRIAD). In test-retest reliability experiments, the proposed method yielded significantly lower volume differences and significantly higher Dice overlaps than the cross-sectional approach for nearly every subregion (average across subregions: 4.5% vs. 6.5%, Dice overlap: 81.8% vs. 75.4%). The longitudinal algorithm also demonstrated increased sensitivity to group differences: in MIRIAD (69 subjects: 46 with AD and 23 controls), it found differences in atrophy rates between AD and controls that the cross sectional method could not detect in a number of subregions: right parasubiculum, left and right presubiculum, right subiculum, left dentate gyrus, left CA4, left HATA and right tail. In ADNI (836 subjects: 369 with AD, 215 with early cognitive impairment - eMCI - and 252 controls), all methods found significant differences between AD and controls, but the proposed longitudinal algorithm detected differences between controls and eMCI and differences between eMCI and AD that the cross sectional method could not find: left presubiculum, right subiculum, left and right parasubiculum, left and right HATA. Moreover, many of the differences that the cross-sectional method already found were detected with higher significance. The presented algorithm will be made available as part of the open-source neuroimaging package FreeSurfer.