PREDICTING BRAIN AGE USING MACHINE LEARNING ALGORITHMS A COMPREHENSIVE EVALUATION
Abstract : The rise of machine learning has unlocked new ways of analysing structural neuroimaging data, including brain age prediction. In this stateoftheart review, we provide an introduction to the methods and potential clinical applications of brain age prediction. Studies on brain age typically involve the creation of a regression machine learning model of agerelated neuroanatomical changes in healthy people. This model is then applied to new subjects to predict their brain age. The difference between predicted brain age and chronological age in a given individual is known as ‘brainage gap’
? The main outcome measure in brain age prediction is the difference between an individual’s predicted age and their chronological age, which is referred to as ‘brainage gap’ in the present review.
? Studies of clinical groups typically estimate the mean brainage gap across all patients and then either compare it to the mean brainage gap of a control group or to zero, where predicted and chronological age are equal
? This outcome can be attributed to LSTM’s storing memory and solving the gradient vanishing problem
? CNN canautomatically notice and extract the appropriate internalstructure from a time series dataset to create indepthinput features, using convolution and pooling operations.
? Also, CNN and LSTM algorithms are resilient to noisetolerance and accuracy for timeseries classi?cation.
? Brain age models tend to be affected by regression to the mean, so the age of younger subjects is overestimated and the age of older subjects is underestimated.
? Various statistical approaches have been proposed to correct for this age bias . Whether a study took age bias into account therefore is an important factor for their interpretation.
? Recent studies compared the performance of commonly used models like support vector regression and relevance vector regression to provide guidance on the most suitable model choices for brain age prediction however, as demonstrated by Wolpert in what is known as the ‘no free lunch’ theorem for machine learning, the performance of different models will depend on the characteristics of the datasets, so there is no single best model for a certain task.
? Nevertheless, in direct comparison to the commonly used shallow machine learning approaches like relevance vector regression, deep learning approaches appear to be comparable (MAE 45 years) or superior (MAE 78 years versus MAE 56 years)
