Keywords:
Multicentre study, Not applicable, Retrospective, Dementia, Computer Applications-General, MR, Neuroradiology brain, Computer applications, Artificial Intelligence, Artificial Intelligence and Machine Learning
Authors:
S. Kaliyugarasan, A. Lundervold, A. S. Lundervold; Bergen/NO
DOI:
10.26044/ecr2020/C-05555
Purpose
The brain is constantly changing throughout our lifespan. For normal aging, some progression of substance loss and change in a variety of brain structures are expected, e.g. volumetric, morphometric and signal changes. The biological age of the brain thus reflects an individual’s chronological age. In individuals with neurodegenerative disorders, such as Alzheimer’s disease, one expects biological and cognitive deviations from normal aging.
These deviations can partly be detected as a gap between “brain age” and chronological age. Having a method that is able to estimate brain age can potentially be clinically useful, providing a biomarker, the “brain age gap”, that relates to the risk of age-related brain disease.
It has been shown that machine learning methods applied to brain imaging data is a useful approach, yielding relatively robust biomarkers for aging. Specifically, it is possible to accurately predict the chronological age of a healthy individual from machine learning models trained on brain imaging data, and then use the deviation in the predicted brain age and the actual chronological age as a sensitive marker of various brain diseases.
With inspiration from the work by Cole et al. [1], we have developed a fast (i.e., seconds) and accurate end-to-end pipeline for brain age prediction directly from T1-weighted DICOM MRI recordings from healthy subjects, using state-of-the-art deep learning techniques (see Fig. 1).