Considerations on brain age predictions from repeatedly sampled data across time

Abstract

IntroductionBrain age, the estimation of a person's age from magnetic resonance imaging (MRI) parameters, has been used as a general indicator of health. The marker requires however further validation for application in clinical contexts. Here, we show how brain age predictions perform for the same individual at various time points and validate our findings with age‐matched healthy controls.MethodsWe used densely sampled T1‐weighted MRI data from four individuals (from two densely sampled datasets) to observe how brain age corresponds to age and is influenced by acquisition and quality parameters. For validation, we used two cross‐sectional datasets. Brain age was predicted by a pretrained deep learning model.ResultsWe found small within‐subject correlations between age and brain age. We also found evidence for the influence of field strength on brain age which replicated in the cross‐sectional validation data and inconclusive effects of scan quality.ConclusionThe absence of maturation effects for the age range in the presented sample, brain age model bias (including training age distribution and field strength), and model error are potential reasons for small relationships between age and brain age in densely sampled longitudinal data. Clinical applications of brain age models should consider of the possibility of apparent biases caused by variation in the data acquisition process.

The intra‐individual reliability of 1H‐MRS measurement in the anterior cingulate cortex across 1 year

“…With the aim of exploring the reliability of MRI measurements, this study specifically for the MRS measurements belongs to the BBSC Project (Korbmacher et al, 2023;Wang et al, 2022Wang et al, , 2023. There are three major findings.…”

“…In this study, a longitudinal MRS dataset is constructed, where three participants were repeatedly scanned for a year with an even interval. With the aim of exploring the reliability of MRI measurements, this study specifically for the MRS measurements belongs to the BBSC Project (Korbmacher et al, 2023 ; Wang et al, 2022 , 2023 ). There are three major findings.…”

“…However, the present new data collection method, where the same participants are scanned repeatedly for a period of time, does not necessarily need a very large sample. But it can provide invaluable insights that conventional studies cannot offer (Korbmacher et al, 2023 ; Wang et al, 2022 , 2023 ). Lastly, the absence of female participants may affect the conclusions of this study, since there is evidence suggesting that the menstrual cycle could influence the GABA level (Harada et al, 2011 ) and other metabolites level (Hjelmervik et al, 2018 ).…”

“…Three participants (Table 1) were scanned in this study, which is part of a precision brain mapping project titled BBSC Project (Korbmacher et al, 2023;Wang et al, 2022Wang et al, , 2023. In this project, we aim to chart the individual brain organizations and explore the reliability of the MRI data by repeatedly scanning three subjects for a year including resting state fMRI data, MRS data, and structural brain imaging data (Wang et al, 2022(Wang et al, , 2023.…”

“…Therefore, in this study, three subjects were repeatedly scanned in an MRI scanner across 1 year, resulting in at least 25 MRS sessions for each subject. To be noticed, this study is part of a project named Bergen breakfast scanning club (BBSC) where we also collected structural and functional MRI data (Korbmacher et al, 2023;Wang et al, 2022Wang et al, , 2023. The aims of the study are threefold.…”