Predicting Age From Brain EEG Signals—A Machine Learning Approach

Abstract

Objective: The brain age gap estimate (BrainAGE) is the difference between the estimated age and the individual chronological age. BrainAGE was studied primarily using MRI techniques. EEG signals in combination with machine learning (ML) approaches were not commonly used for the human age prediction, and BrainAGE. We investigated whether age-related changes are affecting brain EEG signals, and whether we can predict the chronological age and obtain BrainAGE estimates using a rigorous ML framework with a novel and extensive EEG features extraction.Methods: EEG data were obtained from 468 healthy, mood/anxiety, eating and substance use disorder participants (297 females) from the Tulsa-1000, a naturalistic longitudinal study based on Research Domain Criteria framework. Five sets of preprocessed EEG features across channels and frequency bands were used with different ML methods to predict age. Using a nested-cross-validation (NCV) approach and stack-ensemble learning from EEG features, the predicted age was estimated. The important features and their spatial distributions were deduced.Results: The stack-ensemble age prediction model achieved R2 = 0.37 (0.06), Mean Absolute Error (MAE) = 6.87(0.69) and RMSE = 8.46(0.59) in years. The age and predicted age correlation was r = 0.6. The feature importance revealed that age predictors are spread out across different feature types. The NCV approach produced a reliable age estimation, with features consistent behavior across different folds.Conclusion: Our rigorous ML framework and extensive EEG signal features allow a reliable estimation of chronological age, and BrainAGE. This general framework can be extended to test EEG association with and to predict/study other physiological relevant responses.

Automated cot-side tracking of functional brain age in preterm infants

“…This multivariable prediction of age from the EEG enables the estimation of functional brain maturity to within 1-2 weeks of PMA; an accuracy that generalized to an independent validation dataset acquired under a considerably different EEG recording environment. The margin of error is far lower than similar predictions in preterm infants based on functional neuroimaging with fMRI (31), and orders of magnitude lower than what is achieved over later stages of life using EEG or MRI (error margins of 5-10 years) (32)(33)(34). Our findings are also comparable to an array of somatic anatomical methods over similar preterm age ranges based on measures of femur length, head circumference, weight, and structural MRI (cortical folding, thickness) (35)(36)(37)(38).…”

Combination of an Automated 3D Field Phenotyping Workflow and Predictive Modelling for High-Throughput and Non-Invasive Phenotyping of Grape Bunches

“…The multivariate regression models on transformed data dropped substantially in performance for the predicted traits berry number and total volume, whereas the svm poly showed the most stable and the best predictive performance for all predicted traits. In contrast, svms are characterized by seeking to minimize the impact of outliers, having the ability for a good generalization [33,62], and have been proven in many different disciplines [56,[63][64][65].…”

Investigating systematic bias in brain age estimation with application to post‐traumatic stress disorders

“…Brain age can be predicted in individuals based on high‐dimensional neuroimaging data using machine‐learning techniques (Al Zoubi et al, ; Chung et al, ; Cole et al, ; Franke, Luders, May, Wilke, & Gaser, ; Liem et al, ). The predicted brain age can differ from the individual chronological age; the difference between the predicted age and the chronological age, termed the brain age gap (Franke, Ziegler, Klöppel, & Gaser, ) or predicted age difference (Cole, Leech, & Sharp, ), can be used to examine and capture any disease‐related deviations from natural aging.…”