Brain age prediction: A comparison between machine learning models using region- and voxel-based morphometric data

Optimal Blood Pressure Keeps Our Brains Younger

Background: Elevated blood pressure (BP) is a major health risk factor and the leading global cause of premature death. Hypertension is also a risk factor for cognitive decline and dementia. However, when elevated blood pressure starts impacting cerebral health is less clear. We addressed this gap by estimating how a validated measure of brain health relates to changes in BP over a period of 12 years. Methods: Middle-age (44-46 years at baseline, n = 335, 52% female) and older-age (60-64 years, n = 351, 46% female) cognitively intact individuals underwent up to four brain scans. Brain health was assessed using a machine learning approach to produce an estimate of "observed" age (BrainAGE), which can be contrasted with chronological age. Longitudinal associations between blood pressures and BrainAGE were assessed with linear mixed-effects models. Results: A progressive increase in BP was observed over the follow up (MAP = 0.8 mmHg/year, SD = 0.92; SBP = 1.41 mmHg/year, SD = 1.49; DBP = 0.61 mmHg/year, SD = 0.78). In fully adjusted models, every additional 10 mmHg increase in blood pressure (above 90 for mean, 114 for systolic, and 74 for diastolic blood pressure) was associated with a higher BrainAGE by 65.7 days for mean, and 51.1 days for systolic/diastolic blood pressure. These effects occurred across the blood pressure range and were not exclusively driven by hypertension. Conclusion: Increasing blood pressure is associated with poorer brain health. Compared to a person becoming hypertensive, somebody with an ideal BP is predicted to have a brain that appears more than 6 months younger at midlife.

Using normative modelling to detect disease progression in mild cognitive impairment and Alzheimer's disease in a cross-sectional multi-cohort study

Normative modelling is an emerging method for quantifying how individuals deviate from the healthy populational pattern. Several machine learning models have been implemented to develop normative models to investigate brain disorders, including regression, support vector machines and Gaussian process models. With the advance of deep learning technology, the use of deep neural networks has also been proposed. In this study, we assessed normative models based on deep autoencoders using structural neuroimaging data from patients with Alzheimer's disease (n = 206) and mild cognitive impairment (n = 354). We first trained the autoencoder on an independent dataset (UK Biobank dataset) with 11,034 healthy controls. Then, we estimated how each patient deviated from this norm and established which brain regions were associated to this deviation. Finally, we compared the performance of our normative model against traditional classifiers. As expected, we found that patients exhibited deviations according to the severity of their clinical condition. The model identified medial temporal regions, including the hippocampus, and the ventricular system as critical regions for the calculation of the deviation score. Overall, the normative model had comparable cross-cohort generalizability to traditional classifiers. To promote open science, we are making all scripts and the trained models available to the wider research community.

Brain age prediction: A comparison between machine learning models using region- and voxel-based morphometric data

Brain morphology varies across the ageing trajectory and the prediction of a person's age using brain features can aid the detection of abnormalities in the ageing process. Existing studies on such “brain age prediction” vary widely in terms of their methods and type of data, so at present the most accurate and generalisable methodological approach is unclear. Therefore, we used the UK Biobank data set (N = 10,824, age range 47–73) to compare the performance of the machine learning models support vector regression, relevance vector regression and Gaussian process regression on whole‐brain region‐based or voxel‐based structural magnetic resonance imaging data with or without dimensionality reduction through principal component analysis. Performance was assessed in the validation set through cross‐validation as well as an independent test set. The models achieved mean absolute errors between 3.7 and 4.7 years, with those trained on voxel‐level data with principal component analysis performing best. Overall, we observed little difference in performance between models trained on the same data type, indicating that the type of input data had greater impact on performance than model choice. All code is provided online in the hope that this will aid future research.