scholarly journals Prediction of tau accumulation in prodromal Alzheimer’s disease using an ensemble machine learning approach

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Jaeho Kim ◽  
Yuhyun Park ◽  
Seongbeom Park ◽  
Hyemin Jang ◽  
Hee Jin Kim ◽  
...  
Keyword(s):  
Machine Learning ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Clinical Data ◽  
Cortical Thickness ◽  
Parietal Lobe ◽  
Occipital Lobe ◽  
Gradient Boosting ◽  
Prodromal Ad ◽  
The Impact

AbstractWe developed machine learning (ML) algorithms to predict abnormal tau accumulation among patients with prodromal AD. We recruited 64 patients with prodromal AD using the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset. Supervised ML approaches based on the random forest (RF) and a gradient boosting machine (GBM) were used. The GBM resulted in an AUC of 0.61 (95% confidence interval [CI] 0.579–0.647) with clinical data (age, sex, years of education) and a higher AUC of 0.817 (95% CI 0.804–0.830) with clinical and neuropsychological data. The highest AUC was 0.86 (95% CI 0.839–0.885) achieved with additional information such as cortical thickness in clinical data and neuropsychological results. Through the analysis of the impact order of the variables in each ML classifier, cortical thickness of the parietal lobe and occipital lobe and neuropsychological tests of memory domain were found to be more important features for each classifier. Our ML algorithms predicting tau burden may provide important information for the recruitment of participants in potential clinical trials of tau targeting therapies.

scholarly journals A metabolite-based machine learning approach to diagnose Alzheimer’s-type dementia in blood: Results from the European Medical Information Framework for Alzheimer’s Disease biomarker discovery cohort

2019 ◽  
Author(s):  
Daniel Stamate ◽  
Min Kim ◽  
Petroula Proitsi ◽  
Sarah Westwood ◽  
Alison Baird ◽  
...  
Keyword(s):  
Machine Learning ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Medical Information ◽  
Biomarker Discovery ◽  
Gradient Boosting ◽  
Plasma Metabolites ◽  
Csf Biomarkers ◽  
Extreme Gradient Boosting ◽  
And Gender

AbstractINTRODUCTIONMachine learning (ML) may harbor the potential to capture the metabolic complexity in Alzheimer’s Disease (AD). Here we set out to test the performance of metabolites in blood to categorise AD when compared to CSF biomarkers.METHODSThis study analysed samples from 242 cognitively normal (CN) people and 115 with AD-type dementia utilizing plasma metabolites (n=883). Deep Learning (DL), Extreme Gradient Boosting (XGBoost) and Random Forest (RF) were used to differentiate AD from CN. These models were internally validated using Nested Cross Validation (NCV).RESULTSOn the test data, DL produced the AUC of 0.85 (0.80-0.89), XGBoost produced 0.88 (0.86-0.89) and RF produced 0.85 (0.83-0.87). By comparison, CSF measures of amyloid, p-tau and t-tau (together with age and gender) produced with XGBoost the AUC values of 0.78, 0.83 and 0.87, respectively.DISCUSSIONThis study showed that plasma metabolites have the potential to match the AUC of well-established AD CSF biomarkers in a relatively small cohort. Further studies in independent cohorts are needed to validate whether this specific panel of blood metabolites can separate AD from controls, and how specific it is for AD as compared with other neurodegenerative disorders

Simulating the outcome of amyloid treatments in Alzheimer’s Disease from multi-modal imaging and clinical data

2020 ◽  
Author(s):  
Clément Abi Nader ◽  
Nicholas Ayache ◽  
Giovanni B. Frisoni ◽  
Philippe Robert ◽  
Marco Lorenzi ◽  
...  
Keyword(s):  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Clinical Data ◽  
Treatment Strategies ◽  
Disease Evolution ◽  
Clinical Endpoints ◽  
Clinical Biomarkers ◽  
Optimal Intervention ◽  
Models Of Disease ◽  
The Impact

AbstractRecent failures of clinical trials in Alzheimer’s Disease underline the critical importance of identifying optimal intervention time to maximize cognitive benefit. While several models of disease progression have been proposed, we still lack quantitative approaches simulating the effect of treatment strategies on the clinical evolution. In this work, we present a data-driven method to model dynamical relationships between imaging and clinical biomarkers. Our approach allows simulating intervention at any stage of the pathology by modulating the progression speed of the biomarkers, and by subsequently assessing the impact on disease evolution. When applied to multi-modal imaging and clinical data from the Alzheimer’s Disease Neuroimaging Initiative our method enables to generate hypothetical scenarios of amyloid lowering interventions. Our results show that in a study with 1000 individuals per arm, accumulation should be completely arrested at least 5 years before Alzheimer’s dementia diagnosis to lead to statistically powered improvement of clinical endpoints.

scholarly journals A Machine Learning Method to Identify Genetic Variants Potentially Associated With Alzheimer’s Disease

2021 ◽  
Vol 12 ◽  
Author(s):  
Bradley Monk ◽  
Andrei Rajkovic ◽  
Semar Petrus ◽  
Aleks Rajkovic ◽  
Terry Gaasterland ◽  
...  
Keyword(s):  
Machine Learning ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
At Risk ◽  
Genetic Variants ◽  
Genetic Linkage ◽  
Machine Learning Method ◽  
Genomic Information ◽  
Effect Measure ◽  
The Impact

There is hope that genomic information will assist prediction, treatment, and understanding of Alzheimer’s disease (AD). Here, using exome data from ∼10,000 individuals, we explore machine learning neural network (NN) methods to estimate the impact of SNPs (i.e., genetic variants) on AD risk. We develop an NN-based method (netSNP) that identifies hundreds of novel potentially protective or at-risk AD-associated SNPs (along with an effect measure); the majority with frequency under 0.01. For case individuals, the number of “protective” (or “at-risk”) netSNP-identified SNPs in their genome correlates positively (or inversely) with their age of AD diagnosis and inversely (or positively) with autopsy neuropathology. The effect measure increases correlations. Simulations suggest our results are not due to genetic linkage, overfitting, or bias introduced by netSNP. These findings suggest that netSNP can identify SNPs associated with AD pathophysiology that may assist with the diagnosis and mechanistic understanding of the disease.

scholarly journals Regional hyperperfusion in older adults with objectively-defined subtle cognitive decline

2020 ◽  
pp. 0271678X2093517 ◽  
Author(s):  
Kelsey R Thomas ◽  
Jessica R Osuna ◽  
Alexandra J Weigand ◽  
Emily C Edmonds ◽  
Alexandra L Clark ◽  
...  
Keyword(s):  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Cognitive Decline ◽  
Orbitofrontal Cortex ◽  
Arterial Spin Labeling ◽  
Parietal Lobe ◽  
A Priori ◽  
Regions Of Interest ◽  
Group Differences ◽  
Prodromal Ad

Although cerebral blood flow (CBF) alterations are associated with Alzheimer’s disease (AD), CBF patterns across prodromal stages of AD remain unclear. Therefore, we investigated patterns of regional CBF in 162 Alzheimer’s Disease Neuroimaging Initiative participants characterized as cognitively unimpaired (CU; n = 80), objectively-defined subtle cognitive decline (Obj-SCD; n = 31), or mild cognitive impairment (MCI; n = 51). Arterial spin labeling MRI quantified regional CBF in a priori regions of interest: hippocampus, inferior temporal gyrus, inferior parietal lobe, medial orbitofrontal cortex, and rostral middle frontal gyrus. Obj-SCD participants had increased hippocampal and inferior parietal CBF relative to CU and MCI participants and increased inferior temporal CBF relative to MCI participants. CU and MCI groups did not differ in hippocampal or inferior parietal CBF, but CU participants had increased inferior temporal CBF relative to MCI participants. There were no CBF group differences in the two frontal regions. Thus, we found an inverted-U pattern of CBF signal across prodromal AD stages in regions susceptible to early AD pathology. Hippocampal and inferior parietal hyperperfusion in Obj-SCD may reflect early neurovascular dysregulation, whereby higher CBF is needed to maintain cognitive functioning relative to MCI participants, yet is also reflective of early cognitive inefficiencies that distinguish Obj-SCD from CU participants.

scholarly journals A comparison of machine learning methods for survival analysis of high-dimensional clinical data for dementia prediction

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Annette Spooner ◽  
Emily Chen ◽  
Arcot Sowmya ◽  
Perminder Sachdev ◽  
Nicole A. Kochan ◽  
...  
Keyword(s):  
Machine Learning ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Survival Analysis ◽  
Clinical Data ◽  
Machine Learning Algorithms ◽  
High Dimensional ◽  
Complex Data ◽  
Detection And Tracking ◽  
Measure Of Performance

AbstractData collected from clinical trials and cohort studies, such as dementia studies, are often high-dimensional, censored, heterogeneous and contain missing information, presenting challenges to traditional statistical analysis. There is an urgent need for methods that can overcome these challenges to model this complex data. At present there is no cure for dementia and no treatment that can successfully change the course of the disease. Machine learning models that can predict the time until a patient develops dementia are important tools in helping understand dementia risks and can give more accurate results than traditional statistical methods when modelling high-dimensional, heterogeneous, clinical data. This work compares the performance and stability of ten machine learning algorithms, combined with eight feature selection methods, capable of performing survival analysis of high-dimensional, heterogeneous, clinical data. We developed models that predict survival to dementia using baseline data from two different studies. The Sydney Memory and Ageing Study (MAS) is a longitudinal cohort study of 1037 participants, aged 70–90 years, that aims to determine the effects of ageing on cognition. The Alzheimer's Disease Neuroimaging Initiative (ADNI) is a longitudinal study aimed at identifying biomarkers for the early detection and tracking of Alzheimer's disease. Using the concordance index as a measure of performance, our models achieve maximum performance values of 0.82 for MAS and 0.93 For ADNI.

scholarly journals Cardiovascular Risk and Hippocampal Thickness in Alzheimer’s Disease

2013 ◽  
Vol 2013 ◽  
pp. 1-7 ◽  
Author(s):  
Markus Donix ◽  
Maria Scharf ◽  
Kira Marschner ◽  
Annett Werner ◽  
Cathrin Sauer ◽  
...  
Keyword(s):  
Risk Factors ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Cardiovascular Risk ◽  
Cardiovascular Risk Factors ◽  
Temporal Lobe ◽  
Cortical Thickness ◽  
Medial Temporal Lobe ◽  
Cortical Thinning ◽  
The Impact

Cardiovascular risk factors influence onset and progression of Alzheimer’s disease. Among cognitively healthy people, changes in brain structure and function associated with high blood pressure, diabetes, or other vascular risks suggest differential regional susceptibility to neuronal damage. In patients with Alzheimer’s disease, hippocampal and medial temporal lobe atrophy indicate early neuronal loss preferentially in key areas for learning and memory. We wanted to investigate whether this regional cortical thinning would be modulated by cardiovascular risk factors. We utilized high-resolution magnetic resonance imaging and a cortical unfolding technique to determine the cortical thickness of medial temporal subregions in 30 patients with Alzheimer’s disease. Cardiovascular risk was assessed using a sex-specific multivariable risk score. Greater cardiovascular risk was associated with cortical thinning in the hippocampus CA2/3/dentate gyrus area but not other hippocampal and medial temporal subregions. APOE genotype, a family history of Alzheimer’s disease, and age did not influence cortical thickness. Alzheimer’s disease-related atrophy could mask the influence of genetic risk factors or age on regional cortical thickness in medial temporal lobe regions, whereas the impact of vascular risk factors remains detectable. This highlights the importance of cardiovascular disease prevention and treatment in patients with Alzheimer’s disease.

MRI-Based Alzheimer’s Disease-Resemblance Atrophy Index in the Detection of Preclinical and Prodromal Alzheimer’s Disease

2020 ◽  
Author(s):  
Wanting Liu ◽  
Lisa Wing Chi Au ◽  
Jill Abrigo ◽  
Yishan Luo ◽  
Adrian Wong ◽  
...  
Keyword(s):  
Machine Learning ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Hippocampal Volume ◽  
Prodromal Stage ◽  
Preclinical Ad ◽  
Volume Sensitivity ◽  
Prodromal Ad ◽  
Prodromal Alzheimer’S Disease ◽  
Higher Sensitivity

Abstract Background: We aimed to validate the performance of an MRI-based machine learning derived Alzheimer’s Disease-resemblance atrophy index (AD-RAI) in detecting preclinical and prodromal AD. Methods: A total of 62 subjects (mild cognitive impairment [MCI]=25, cognitively unimpaired [CU]=37) underwent MRI, 11C- PIB, and 18F-T807 PET. We investigated the performance of AD-RAI at the pre-specified cutoff of ≥ 0.5 in detecting preclinical and prodromal AD and compared its performance with that of visual and volumetric hippocampal measures. Results: AD-RAI achieved the best metrics among all subjects (sensitivity 0.73, specificity 0.91, accuracy 87.10%) and among MCI subgroup (sensitivity 0.91, specificity 0.79, accuracy 84.00%) in detecting A+T+ subjects over other measures. Among CU subgroup, hippocampal volume (sensitivity 0.75, specificity 0.88, accuracy 86.49%) achieved a higher sensitivity than AD-RAI (sensitivity 0.25, specificity 0.97, accuracy 89.19%) in detecting preclinical AD.Conclusions: AD-RAI aids the detection of early AD, in particular at the prodromal stage.

scholarly journals The Evaluation of Tau Deposition with [18F]PI-2620 by Using a Semiquantitative Method in Cognitively Normal Subjects and Patients with Mild Cognitive Impairment and Alzheimer’s Disease

2021 ◽  
Vol 2021 ◽  
pp. 1-8
Author(s):  
Attapon Jantarato ◽  
Sira Vachatimanont ◽  
Natphimol Boonkawin ◽  
Sukanya Yaset ◽  
Anchisa Kunawudhi ◽  
...  
Keyword(s):  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Cognitive Impairment ◽  
Mild Cognitive Impairment ◽  
Parietal Lobe ◽  
Standardized Uptake Value ◽  
Occipital Lobe ◽  
Normal Subjects ◽  
Semiquantitative Analysis ◽  
Cognitively Normal

Background. Some studies have reported the effectiveness of [18F]PI-2620 as an effective tau-binding radiotracer; however, few reports have applied semiquantitative analysis to the tracer. Therefore, this study’s aim was to perform a semiquantitative analysis of [18F]PI-2620 in individuals with normal cognition and patients with mild cognitive impairment (MCI) and Alzheimer’s disease (AD). Methods. Twenty-six cognitively normal (CN) subjects, 7 patients with AD, and 36 patients with MCI were enrolled. A dynamic positron emission tomography (PET) scan was performed 30–75 min postinjection. PET and T1-weighted magnetic resonance imaging scans were coregistered. The standardized uptake value ratio (SUVr) was used for semiquantitative analysis. The P-Mod software was applied to create volumes of interest. The ANOVA and post hoc Tukey HSD were used for statistical analysis. Results. In the AD group, the occipital lobe had a significantly higher mean SUVr ( 1.46 ± 0.57 ) than in the CN and MCI groups. Compared with the CN group, the AD group showed significantly higher mean SUVr in the fusiform gyrus ( 1.06 ± 0.09 vs. 1.49 ± 0.86 ), inferior temporal ( 1.07 ± 0.07 vs. 1.46 ± 0.08 ), parietal lobe, lingual gyrus, and precuneus regions. Similarly, the AD group demonstrated a higher mean SUVr than the MCI group in the precuneus, lingual, inferior temporal, fusiform, supramarginal, orbitofrontal, and superior temporal regions. The remaining observed regions, including the striatum, basal ganglia, thalamus, and white matter, showed a low SUVr across all groups with no statistically significant differences. Conclusion. A significantly higher mean SUVr of [18F]PI-2620 was observed in the AD group; a significant area of the brain in the AD group demonstrated tau protein deposit in concordance with Braak Stages III–V, providing useful information to differentiate AD from CN and MCI. Moreover, the low SUVr in the deep striatum and thalamus could be useful for excluding primary tauopathies.

scholarly journals Predictability of Mortality in Patients With Myocardial Injury After Noncardiac Surgery Based on Perioperative Factors via Machine Learning: Retrospective Study

10.2196/32771 ◽  
2021 ◽  
Vol 9 (10) ◽  
pp. e32771
Author(s):  
Seo Jeong Shin ◽  
Jungchan Park ◽  
Seung-Hwa Lee ◽  
Kwangmo Yang ◽  
Rae Woong Park
Keyword(s):  
Machine Learning ◽  
Clinical Data ◽  
Myocardial Injury ◽  
Learning Algorithms ◽  
Noncardiac Surgery ◽  
Machine Learning Algorithms ◽  
Gradient Boosting ◽  
Data Set ◽  
Extreme Gradient Boosting ◽  
The Impact

Background Myocardial injury after noncardiac surgery (MINS) is associated with increased postoperative mortality, but the relevant perioperative factors that contribute to the mortality of patients with MINS have not been fully evaluated. Objective To establish a comprehensive body of knowledge relating to patients with MINS, we researched the best performing predictive model based on machine learning algorithms. Methods Using clinical data from 7629 patients with MINS from the clinical data warehouse, we evaluated 8 machine learning algorithms for accuracy, precision, recall, F1 score, area under the receiver operating characteristic (AUROC) curve, and area under the precision-recall curve to investigate the best model for predicting mortality. Feature importance and Shapley Additive Explanations values were analyzed to explain the role of each clinical factor in patients with MINS. Results Extreme gradient boosting outperformed the other models. The model showed an AUROC of 0.923 (95% CI 0.916-0.930). The AUROC of the model did not decrease in the test data set (0.894, 95% CI 0.86-0.922; P=.06). Antiplatelet drugs prescription, elevated C-reactive protein level, and beta blocker prescription were associated with reduced 30-day mortality. Conclusions Predicting the mortality of patients with MINS was shown to be feasible using machine learning. By analyzing the impact of predictors, markers that should be cautiously monitored by clinicians may be identified.

Sign in / Sign up

Export Citation Format

Share Document