Dataset Information

Structural MRI and Amyloid PET Imaging for Prediction of Conversion to Alzheimer's Disease in Patients with Mild Cognitive Impairment: A Meta-Analysis.

ABSTRACT:

Objective

The aim of this study was to explore the prognostic values of biomarkers of neurodegeneration as measured by magnetic resonance imaging (MRI) and amyloid burden as measured by amyloid positron emission tomography (PET) in predicting conversion to Alzheimer's disease (AD) in patients with mild cognitive impairment (MCI).

Methods

PubMed and EMBASE databases were searched for structural MRI or amyloid PET imaging studies published between January 2000 and July 2014 that reported conversion to AD in patients with MCI. Means and standard deviations or individual numbers of biomarkers with positive or negative status at baseline and corresponding numbers of patients who had progressed to AD at follow-up were retrieved from each study. The effect size of each biomarker was expressed as Hedges's g.

Results

Twenty-four MRI studies and 8 amyloid PET imaging studies were retrieved. 674 of the 1741 participants (39%) developed AD. The effect size for predicting conversion to AD was 0.770 [95% confidence interval (CI) 0.607-0.934] for across MRI and 1.316 (95% CI 0.920-1.412) for amyloid PET imaging (p<0.001). The effect size was 1.256 (95% CI 0.902-1.609) for entorhinal cortex volume from MRI.

Conclusion

Our study suggests that volumetric MRI measurement may be useful for the early detection of AD.

SUBMITTER: Seo EH

PROVIDER: S-EPMC5355020 | biostudies-literature | 2017 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Structural MRI and Amyloid PET Imaging for Prediction of Conversion to Alzheimer's Disease in Patients with Mild Cognitive Impairment: A Meta-Analysis.

Seo Eun Hyun EH Park Woon Yeong WY Choo Il Han IH

Psychiatry investigation 20170306 2

<h4>Objective</h4>The aim of this study was to explore the prognostic values of biomarkers of neurodegeneration as measured by magnetic resonance imaging (MRI) and amyloid burden as measured by amyloid positron emission tomography (PET) in predicting conversion to Alzheimer's disease (AD) in patients with mild cognitive impairment (MCI).<h4>Methods</h4>PubMed and EMBASE databases were searched for structural MRI or amyloid PET imaging studies published between January 2000 and July 2014 that rep ...[more]

PMID: 28326120

Similar Datasets

Project description:Background Growing evidence indicates an association between cerebral microhemorrhages (MHs) and amyloid β accumulation in Alzheimer disease (AD), but to the knowledge of the authors the association with tau burden is unknown. Purpose To investigate the association between cerebral MH load and tau pathologic structure measured in healthy older individuals and individuals along the AD spectrum, stratified by using the A (amyloid β)/T (tau)/N (neurodegeneration) biomarker classification system. Materials and methods In this prospective cohort study, participants from the AD Neuroimaging Initiative were included (healthy control participants, participants with mild cognitive impairment, and participants with AD dementia; data from October 2005 to January 2019). T2*-weighted gradient-echo MRI was performed to quantify MH, fluorine 18 (18F) flortaucipir (AV-1451) PET was performed to quantify tau, and 18F-florbetaben/18F- florbetapir (AV45) PET was performed to quantify amyloid β to study associations of MH with regional and global tau and amyloid β load. Associations with cerebrospinal fluid (CSF) biomarkers (amyloid β1-42, total tau, phosphorylated tau 181) were also assessed. Analysis of covariance and Spearman rank correlation test for cross-sectional analysis and Wilcoxon signed rank test for longitudinal analyses were used, controlling for multiple comparisons (Bonferroni significance threshold, P < .008). Results Evaluated were 343 participants (mean age, 75 years ± 7; 186 women), including 205 participants who were A-TN- (mean age, 73 years ± 7; 115 women), 80 participants who were A+TN- (mean age, 76 years ± 7; 38 women), and 58 participants who were A+TN+ (mean age, 77 ± 8; 34 women). MH count was associated with global (Spearman ρ = 0.27; P = .004) and frontal (ρ = 0.27; P = .005) amyloid β load and global tau load (ρ = 0.31; P = .001). In a longitudinal analysis, MH count increased significantly over approximately 5 years in the entire cohort (T-1, 81 [range, 0-6 participants]; T0, 214 [range, 0-58 participants]; P < .001), in A+TN+ (T-1, 20 [range, 0-5 participants]; T0, 119 [range, 1-58 participants]; P < .001), A+TN- (T-1, 31 [range, 0-6 participants]; T0, 43 [range, 0-8 participants]; P = .03), and A-TN- (T-1, 30 [range, 0-4 participants]; T0, 52 [range, 0-6 participants]; P = .007). A higher MH count was associated with higher future global (ρ = 0.29; P = .008) and parietal (ρ = 0.31; P = .005) amyloid β and parietal tau load (ρ = 0.31; P = .005). Conclusion Cerebral microhemorrhage load is associated spatially with tau accumulation, both cross-sectionally and longitudinally. © RSNA, 2020 Online supplemental material is available for this article.

Project description:Accurately identifying the patients that have mild cognitive impairment (MCI) who will go on to develop Alzheimer's disease (AD) will become essential as new treatments will require identification of AD patients at earlier stages in the disease process. Most previous work in this area has centred around the same automated techniques used to diagnose AD patients from healthy controls, by coupling high dimensional brain image data or other relevant biomarker data to modern machine learning techniques. Such studies can now distinguish between AD patients and controls as accurately as an experienced clinician. Models trained on patients with AD and control subjects can also distinguish between MCI patients that will convert to AD within a given timeframe (MCI-c) and those that remain stable (MCI-s), although differences between these groups are smaller and thus, the corresponding accuracy is lower. The most common type of classifier used in these studies is the support vector machine, which gives categorical class decisions. In this paper, we introduce Gaussian process (GP) classification to the problem. This fully Bayesian method produces naturally probabilistic predictions, which we show correlate well with the actual chances of converting to AD within 3 years in a population of 96 MCI-s and 47 MCI-c subjects. Furthermore, we show that GPs can integrate multimodal data (in this study volumetric MRI, FDG-PET, cerebrospinal fluid, and APOE genotype with the classification process through the use of a mixed kernel). The GP approach aids combination of different data sources by learning parameters automatically from training data via type-II maximum likelihood, which we compare to a more conventional method based on cross validation and an SVM classifier. When the resulting probabilities from the GP are dichotomised to produce a binary classification, the results for predicting MCI conversion based on the combination of all three types of data show a balanced accuracy of 74%. This is a substantially higher accuracy than could be obtained using any individual modality or using a multikernel SVM, and is competitive with the highest accuracy yet achieved for predicting conversion within three years on the widely used ADNI dataset.

Project description:BACKGROUND:Progression of mild cognitive impairment (MCI) to Alzheimer's disease (AD) dementia can be predicted by clinical features and a combination of biomarkers may increase the predictive power. In the present study, we investigated whether the combination of olfactory function and plasma neuronal-derived exosome (NDE) A?1-42 can best predict progression to AD dementia. METHODS:87 MCI patients were enrolled and received the cognitive assessment at 2-year and 3-year follow-up to reevaluate cognition. In the meanwhile, 80 healthy controls and 88?AD dementia patients were enrolled at baseline as well to evaluate the diagnose value in cross-section. Olfactory function was evaluated with the sniffin sticks (SS-16) and A?1-42 levels in NDEs were determined by ELISA. Logistic regression was performed to evaluate the risk factors for cognitive decline in MCI at 2-year and 3-year revisits. RESULTS:In the cross cohort, lower SS-16 scores and higher A?1-42 levels in NDEs were found in MCI and AD dementia compared to healthy controls. For the longitudinal set, 8 MCI individuals developed AD dementia within 2?years, and 16 MCI individuals developed AD dementia within 3?years. The two parameter-combination of SS-16 scores and A?1-42 level in NDEs showed better prediction in the conversion of MCI to AD dementia at 2-year and 3-year revisit. Moreover, after a 3-year follow-up, SS-16 scores also significantly predicted the conversion to AD dementia, where lower scores were associated with a 10-fold increased risk of developing AD dementia (p?=?0.006). Similarly, higher A?1-42 levels in NDEs in patients with MCI increased the risk of developing AD dementia by 8.5-fold (p?=?0.002). CONCLUSION:A combination of two biomarkers of NDEs (A?1-42) and SS-16 predicted the conversion of MCI to AD dementia more accurately in combination. These findings have critical implications for understanding the pathophysiology of AD dementia and for developing preventative treatments for cognitive decline.

Project description:BackgroundWhile there is no cure for Alzheimer's disease (AD), early diagnosis and accurate prognosis of AD may enable or encourage lifestyle changes, neurocognitive enrichment, and interventions to slow the rate of cognitive decline. The goal of our study was to develop and evaluate a novel deep learning algorithm to predict mild cognitive impairment (MCI) to AD conversion at three years after diagnosis using longitudinal and whole-brain 3D MRI.MethodsThis retrospective study consisted of 320 normal cognition (NC), 554 MCI, and 237 AD patients. Longitudinal data include T1-weighted 3D MRI obtained at initial presentation with diagnosis of MCI and at 12-month follow up. Whole-brain 3D MRI volumes were used without a priori segmentation of regional structural volumes or cortical thicknesses. MRIs of the AD and NC cohort were used to train a deep learning classification model to obtain weights to be applied via transfer learning for prediction of MCI patient conversion to AD at three years post-diagnosis. Two (zero-shot and fine tuning) transfer learning methods were evaluated. Three different convolutional neural network (CNN) architectures (sequential, residual bottleneck, and wide residual) were compared. Data were split into 75% and 25% for training and testing, respectively, with 4-fold cross validation. Prediction accuracy was evaluated using balanced accuracy. Heatmaps were generated.ResultsThe sequential convolutional approach yielded slightly better performance than the residual-based architecture, the zero-shot transfer learning approach yielded better performance than fine tuning, and CNN using longitudinal data performed better than CNN using a single timepoint MRI in predicting MCI conversion to AD. The best CNN model for predicting MCI conversion to AD at three years after diagnosis yielded a balanced accuracy of 0.793. Heatmaps of the prediction model showed regions most relevant to the network including the lateral ventricles, periventricular white matter and cortical gray matter.ConclusionsThis is the first convolutional neural network model using longitudinal and whole-brain 3D MRIs without extracting regional brain volumes or cortical thicknesses to predict future MCI to AD conversion at 3 years after diagnosis. This approach could lead to early prediction of patients who are likely to progress to AD and thus may lead to better management of the disease.

Dataset Information

Structural MRI and Amyloid PET Imaging for Prediction of Conversion to Alzheimer's Disease in Patients with Mild Cognitive Impairment: A Meta-Analysis.

Objective

Methods

Results

Conclusion

Publications

Structural MRI and Amyloid PET Imaging for Prediction of Conversion to Alzheimer's Disease in Patients with Mild Cognitive Impairment: A Meta-Analysis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure