An estimate of the longitudinal pace of aging from a single brain scan predicts dementia conversion, morbidity, and mortality DOI Creative Commons

Ethan T. Whitman,

Maxwell L. Elliott, Annchen R. Knodt

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 22, 2024

To understand how aging affects functional decline and increases disease risk, it is necessary to develop accurate reliable measures of fast a person aging. Epigenetic clocks measure but require DNA methylation data, which many studies lack. Using data from the Dunedin Study, we introduce an for rate longitudinal derived cross-sectional brain MRI: Pace Aging Calculated NeuroImaging or DunedinPACNI. Exporting this Alzheimer's Disease Neuroimaging Initiative UK Biobank neuroimaging datasets revealed that faster DunedinPACNI predicted participants' cognitive impairment, accelerated atrophy, conversion diagnosed dementia. Underscoring close links between body brain, also physical frailty, poor health, future chronic diseases, mortality in older adults. Furthermore, followed expected socioeconomic health gradient. When compared age gap, existing MRI biomarker, was similarly more strongly related clinical outcomes. 'next generation' will be made publicly available research community help accelerate evaluate effectiveness dementia prevention anti-aging strategies.

Language: Английский

Examining the reliability of brain age algorithms under varying degrees of participant motion DOI Creative Commons
Jamie L. Hanson,

Dorthea J Adkins,

Eva Bacas

et al.

Brain Informatics, Journal Year: 2024, Volume and Issue: 11(1)

Published: April 4, 2024

Abstract Brain age algorithms using data science and machine learning techniques show promise as biomarkers for neurodegenerative disorders aging. However, head motion during MRI scanning may compromise image quality influence brain estimates. We examined the effects of on predictions in adult participants with low, high, no scans ( Original N = 148; Analytic 138 ). Five popular were tested: brainageR, DeepBrainNet, XGBoost, ENIGMA, pyment. Evaluation metrics, intraclass correlations (ICCs), Bland–Altman analyses assessed reliability across conditions. Linear mixed models quantified effects. Results demonstrated significantly impacted estimates some algorithms, ICCs dropping low 0.609 errors increasing up to 11.5 years high scans. DeepBrainNet pyment showed greatest robustness (ICCs 0.956–0.965). XGBoost brainageR had largest (up 13.5 RMSE) bias motion. Findings indicate artifacts significant ways. Furthermore, our results suggest certain like be preferable deployment populations where acquisition is likely. Further optimization validation critical use a biomarker relevant clinical outcomes.

Language: Английский

Citations

5
Rate of brain aging associates with future executive function in Asian children and older adults DOI Open Access
Susan F. Cheng, Wan Lin Yue, Kwun Kei Ng

et al.

Published: Feb. 4, 2025

Brain age has emerged as a powerful tool to understand neuroanatomical aging and its link health outcomes like cognition. However, there remains lack of studies investigating the rate brain relationship Furthermore, most models are trained tested on cross-sectional data from primarily Caucasian, adult participants. It is thus unclear how well these generalize non-Caucasian participants, especially children. Here, we previously published deep learning model Singaporean elderly participants (55 − 88 years old) children (4 11 old). We found that directly generalized but finetuning was necessary for After finetuning, change in gap associated with future executive function performance both further lateral ventricles frontal areas contributed prediction while white matter posterior regions were more important predicting Taken together, our results suggest potential generalizing diverse populations. Moreover, longitudinal reflects developing processes brain, relating cognitive function.

Language: Английский

Citations

0
Considerations on brain age predictions from repeatedly sampled data across time DOI Creative Commons
Max Korbmacher, Mengyun Wang, Rune Eikeland

et al.

Brain and Behavior, Journal Year: 2023, Volume and Issue: 13(10)

Published: Aug. 16, 2023

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

Language: Английский

Citations

9
Feasibility of brain age predictions from clinical T1-weighted MRIs DOI Creative Commons
Pedro A. Valdés-Hernández, Chavier Laffitte Nodarse, James H. Cole

et al.

Brain Research Bulletin, Journal Year: 2023, Volume and Issue: 205, P. 110811 - 110811

Published: Nov. 10, 2023

An individual's brain predicted age minus chronological (brain-PAD) obtained from MRIs could become a biomarker of disease in research studies. However, reports clinical are scant despite the rich information hospitals provide. Since MRI protocols meant for specific purposes, performance predictions on data need to be tested. We explored feasibility using DeepBrainNet, deep network previously trained research-oriented MRIs, predict ages 840 patients who visited 15 facilities health system Florida. Anticipating strong prediction bias our sample, we characterized it propose covariate model group-level regressions brain-PAD (recommended avoid Type I, II errors), and tested its generalizability, requirement meaningful new single cases. The best bias-related was scanner-independent linear age, while method estimate bias-free inverse quadratic function. demonstrated detect sex-related differences regression accounting selected model. These were preserved after correction. Mean-Average Error (MAE) independent ∼8 years, 2-3 years greater than whereas an R

Language: Английский

Citations

4
Rate of brain aging associates with future executive function in Asian children and older adults DOI Open Access
Susan F. Cheng, Wan Lin Yue, Kwun Kei Ng

et al.

Published: June 7, 2024

Brain age has emerged as a powerful tool to understand neuroanatomical aging and its link health outcomes like cognition. However, there remains lack of studies investigating the rate brain relationship Furthermore, most models are trained tested on cross-sectional data from primarily Caucasian, adult participants. It is thus unclear how well these generalize non-Caucasian participants, especially children. Here, we previously published deep learning model Singaporean elderly participants (55 − 88 years old) children (4 11 old). We found that directly generalized but finetuning was necessary for After finetuning, change in gap associated with future executive function performance both further lateral ventricles frontal areas contributed prediction while white matter posterior regions were more important predicting Taken together, our results suggest potential generalizing diverse populations. Moreover, longitudinal reflects developing processes brain, relating cognitive function.

Language: Английский

Citations

1
Attention Over Vulnerable Brain Regions Associating Cerebral Palsy Disorder and Biological Markers DOI Creative Commons
Muhammad Hassan,

Jieqong Lin,

Ahmed Ameen Fateh

et al.

Journal of Advanced Research, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 1, 2024

Language: Английский

Citations

1
Considerations on brain age predictions from repeatedly sampled data across time DOI Creative Commons
Max Korbmacher, Mengyun Wang, Rune Eikeland

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: March 31, 2023

Abstract Introduction Brain age, the estimation of a person’s age from magnetic resonance imaging (MRI) parameters, has been used as general indicator health. The marker requires however further validation for application in clinical contexts. Here, we show how brain predictions perform same individual at various time points and validate our findings with age-matched healthy controls. Methods We densly sampled T1-weighted MRI data four individuals (from two datasets) to observe corresponds is influenced by acquision quality parameters. For validation, cross-sectional datasets. was predicted pre-trained deep learning model. Results find small within-subject correlations between age. also evidence influence field strength on which replicated data, inconclusive effects scan quality. Conclusion absence maturation range presented sample, model-bias (including training distribution strength) model error are potential reasons relationships longitudinal data. Future models should account differences intra-individual differences.

Language: Английский

Citations

1
Rate of brain aging associates with future executive function in Asian children and older adults DOI Creative Commons
Susan F. Cheng, Wan Lin Yue, Kwun Kei Ng

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Nov. 28, 2023

ABSTRACT Brain age has emerged as a powerful tool to understand neuroanatomical aging and its link health outcomes like cognition. However, there remains lack of studies investigating the rate brain relationship Furthermore, most models are trained tested on cross-sectional data from primarily Caucasian, adult participants. It is thus unclear how well these generalize non-Caucasian participants, especially children. Here, we previously published deep learning model Singaporean elderly participants (55 − 88 years old) children (4 11 old). We found that directly generalized but finetuning was necessary for After finetuning, change in gap associated with future executive function performance both further lateral ventricles frontal areas contributed prediction while white matter posterior regions were more important predicting Taken together, our results suggest potential generalizing diverse populations. Moreover, longitudinal reflects developing processes brain, relating cognitive function.

Language: Английский

Citations

1
Multimodal brain age prediction using machine learning: combining structural MRI and 5-HT2AR PET derived features DOI Creative Commons
Ruben P. Dörfel,

Joan M. Arenas‐Gomez,

Claus Svarer

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 6, 2024

Abstract To better assess the pathology of neurodegenerative disorders and efficacy neuroprotective interventions, it is necessary to develop biomarkers that can accurately capture age-related biological changes in human brain. Brain serotonin 2A receptors (5-HT2AR) show a particularly profound decline are also reduced disorders, such as Alzheimer’s disease. This study investigates whether 5-HT2AR binding, measured vivo using positron emission tomography (PET), be used biomarker for brain aging. Specifically, we aim 1) predict age binding outcomes, 2) compare 5-HT2AR-based predictions based on gray matter (GM) volume, determined with structural magnetic resonance imaging (MRI), 3) investigate combining GM volume data improves prediction. We PET MR images from 209 healthy individuals aged between 18 85 years (mean=38, std=18), estimated 14 cortical subcortical regions. Different machine learning algorithms were applied chronological combined measures. The mean absolute error (MAE) cross-validation approach evaluation model comparison. find both cerebral (mean MAE=6.63 years, std=0.74 years) MAE=6.95 std=0.83 accurately. Combining two measures prediction further MAE=5.54 std=0.68). In conclusion, might useful improving quantification

Language: Английский

Citations

0
eLife assessment: Rate of brain aging associates with future executive function in Asian children and older adults DOI Open Access
Saâd Jbabdi

Published: June 7, 2024

Brain age has emerged as a powerful tool to understand neuroanatomical aging and its link health outcomes like cognition. However, there remains lack of studies investigating the rate brain relationship Furthermore, most models are trained tested on cross-sectional data from primarily Caucasian, adult participants. It is thus unclear how well these generalize non-Caucasian participants, especially children. Here, we previously published deep learning model Singaporean elderly participants (55 − 88 years old) children (4 11 old). We found that directly generalized but finetuning was necessary for After finetuning, change in gap associated with future executive function performance both further lateral ventricles frontal areas contributed prediction while white matter posterior regions were more important predicting Taken together, our results suggest potential generalizing diverse populations. Moreover, longitudinal reflects developing processes brain, relating cognitive function.

Language: Английский

Citations

0