Biological Brain Age Prediction Using Machine Learning on Structural Neuroimaging Data: Multi-Cohort Validation Against Biomarkers of Alzheimer’s Disease and Neurodegeneration stratified by sex

García-Prat, Marina; Operto, Grégory; Falcón, Carles; Shekari, Mahnaz; Cacciaglia, Raffaele; Milà‐Alomà, Marta; Lorenzini, Luigi; Ingala, Silvia; Wink, Alle Meije; Mutsaerts, Henk‐Jan; Minguillón, Carolina; Fauria, Karine; Molinuevo, José Luís; Haller, Sven; Chételat, Gaël; Waldman, Adam; Schwarz, Adam J.; Barkhof, Frederik; Suridjan, Ivonne; Kollmorgen, Gwendlyn; Bayfield, Anna; Zetterberg, Henrik; Blennow, Kaj; Suárez‐Calvet, Marc; Gispert, Juan Domingo

doi:10.1101/2022.06.23.22276492

Cited by 4 publications

(9 citation statements)

References 86 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We found that predicted brain ages generated by a densely connected neural network using 3 distinct sets of neuroimaging features (FW-corrected dMRI, T1-weighted MRI, combined FW+T1) all showed high correlation with actual age in baseline CU participants, which confirms findings from previous literature that have accurately predicted chronological age of healthy adults using neuroimaging-derived measures with machine learning approaches including deep learning 11,12,17,26–29 . Importantly, the top-contributing neuroimaging features identified for each model ( Figure 2 ) provide biological interpretability as they include brain regions that have been associated with both normal aging and AD neuropathology.…”

Section: Discussionsupporting

confidence: 86%

“…Moreover, individuals with a higher predicted brain age at baseline were more likely to convert from MCI to AD 42 or develop dementia later in life 18 . Studies generating predicted age difference from structural MRI scans of healthy controls have also found correlations with performance on traditional screening tools for AD (e.g., Mini-Mental State Examination, Clinical Dementia Ratio), anatomical measurements such as cortical thickness and hippocampal volume 43 , AD neuropathology such as β-amyloid positivity 16,26 , and AD risk factors such as APOE-4 carrier status 16,26 . We also found that all predicted brain ages were robustly associated with cross-sectional and longitudinal cognitive function including baseline memory and executive function scores and longitudinal memory and executive function slopes.…”

Section: Predicted Age Is a More Sensitive Measure Than Actual Age An...mentioning

confidence: 99%

See 1 more Smart Citation

A deep neural network estimation of brain age is sensitive to cognitive impairment and decline

Yang,

Sathe,

Schilling

et al. 2023

Preprint

View full text Add to dashboard Cite

The greatest known risk factor for Alzheimer's disease (AD) is age. While both normal aging and AD pathology involve structural changes in the brain, their trajectories of atrophy are not the same. Recent developments in artificial intelligence have encouraged studies to leverage neuroimaging-derived measures and deep learning approaches to predict brain age, which has shown promise as a sensitive biomarker in diagnosing and monitoring AD. However, prior efforts primarily involved structural magnetic resonance imaging and conventional diffusion MRI (dMRI) metrics without accounting for partial volume effects. To address this issue, we post-processed our dMRI scans with an advanced free-water (FW) correction technique to compute distinct FW-corrected fractional anisotropy (FAFWcorr) and FW maps that allow for the separation of tissue from fluid in a scan. We built 3 densely connected neural networks from FW-corrected dMRI, T1-weighted MRI, and combined FW+T1 features, respectively, to predict brain age. We then investigated the relationship of actual age and predicted brain ages with cognition. We found that all models accurately predicted actual age in cognitively unimpaired (CU) controls (FW: r=0.66,p=1.62x10-32; T1: r=0.61,p=1.45x10-26, FW+T1: r=0.77,p=6.48x10-50) and distinguished between CU and mild cognitive impairment participants (FW:p=0.006; T1:p=0.048; FW+T1:p=0.003), with FW+T1-derived age showing best performance. Additionally, all predicted brain age models were significantly associated with cross-sectional cognition (memory, FW: β=-1.094,p=6.32x10-7; T1: β=-1.331,p=6.52x10-7; FW+T1: β=-1.476,p=2.53x10-10; executive function, FW: β=-1.276,p=1.46x10-9; T1: β=-1.337,p=2.52x10-7; FW+T1: β=-1.850,p=3.85x10-17) and longitudinal cognition (memory, FW: β=-0.091,p=4.62x10-11; T1: β=-0.097,p=1.40x10-8; FW+T1: β=-0.101,p=1.35x10-11; executive function, FW: β=-0.125,p=1.20x10-10; T1: β=-0.163,p=4.25x10-12; FW+T1: β=-0.158,p=1.65x10-14). Our findings provide evidence that both T1-weighted MRI and dMRI measures improve brain age prediction and support predicted brain age as a sensitive biomarker of cognition and cognitive decline.

show abstract

Section: Discussionsupporting

confidence: 86%

Section: Predicted Age Is a More Sensitive Measure Than Actual Age An...mentioning

confidence: 99%

A deep neural network estimation of brain age is sensitive to cognitive impairment and decline

Yang,

Sathe,

Schilling

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The precise prediction of both chronological and biological age in PD constitutes a clinically significant task. Recent machine learning studies have leveraged extensive datasets, such as the UK Biobank brain imaging data, to construct brain-age models and investigate various aging-related hypotheses [23,24]. One approach involves calculating the brain-age delta by subtracting chronological age from the estimated brain age [25].…”

Section: Related Workmentioning

confidence: 99%

Analyzing wav2vec embedding in Parkinson’s disease speech: A study on cross-database classification and regression tasks

Klempir,

Krupicka

2024

Preprint

View full text Add to dashboard Cite

Advancements in deep learning speech representations have facilitated the effective use of extensive datasets comprised of unlabeled speech signals, and have achieved success in modeling tasks associated with Parkinson's disease (PD) with minimal annotated data. This study focuses on PD non-fine-tuned wav2vec 1.0 architecture. Utilizing features derived from wav2vec embedding, we develop machine learning models tailored for clinically relevant PD speech diagnosis tasks, such as cross-database classification and regression to predict demographic and articulation characteristics, for instance, modeling the subjects' age and number of characters per second. The primary aim is to conduct feature importance analysis on both classification and regression tasks, investigating whether latent discrete speech representations in PD are shared across models, particularly for related tasks. The proposed wav2vec-based models were evaluated on PD versus healthy controls using three multi-language-task PD datasets. Results indicated that wav2vec accurately detected PD based on speech, outperforming feature extraction using mel-frequency cepstral coefficients in the proposed cross-database scenarios. Furthermore, wav2vec proved effective in regression, modeling various quantitative speech characteristics related to intelligibility and aging. Subsequent analysis of important features, obtained using scikit-learn feature importance built-in tools and the Shapley additive explanations method, examined the presence of significant overlaps between classification and regression models. The feature importance experiments discovered shared features across trained models, with increased sharing for related tasks, further suggesting that wav2vec contributes to improved generalizability. In conclusion, the study proposes wav2vec embedding as a promising step toward a speech-based universal model to assist in the evaluation of PD.

show abstract

“…A brain age estimation higher than that of healthy agematched peers has already been linked to AD [13][14][15][16], Parkinson's disease [17], Schizophrenia [16,[18][19][20], Multiple Sclerosis [21][22][23], as well as to life expectancy [24]. Various brain age models exist [25], but tend to be rather complex in their interpretation.…”

Section: Introductionmentioning

confidence: 99%

Brain age as a biomarker for pathological versus healthy ageing – a REMEMBER study

Wittens,

Denissen,

Sima

et al. 2024

Alz Res Therapy

View full text Add to dashboard Cite

Objectives This study aimed to evaluate the potential clinical value of a new brain age prediction model as a single interpretable variable representing the condition of our brain. Among many clinical use cases, brain age could be a novel outcome measure to assess the preventive effect of life-style interventions. Methods The REMEMBER study population (N = 742) consisted of cognitively healthy (HC,N = 91), subjective cognitive decline (SCD,N = 65), mild cognitive impairment (MCI,N = 319) and AD dementia (ADD,N = 267) subjects. Automated brain volumetry of global, cortical, and subcortical brain structures computed by the CE-labeled and FDA-cleared software icobrain dm (dementia) was retrospectively extracted from T1-weighted MRI sequences that were acquired during clinical routine at participating memory clinics from the Belgian Dementia Council. The volumetric features, along with sex, were combined into a weighted sum using a linear model, and were used to predict ‘brain age’ and ‘brain predicted age difference’ (BPAD = brain age–chronological age) for every subject. Results MCI and ADD patients showed an increased brain age compared to their chronological age. Overall, brain age outperformed BPAD and chronological age in terms of classification accuracy across the AD spectrum. There was a weak-to-moderate correlation between total MMSE score and both brain age (r = -0.38,p < .001) and BPAD (r = -0.26,p < .001). Noticeable trends, but no significant correlations, were found between BPAD and incidence of conversion from MCI to ADD, nor between BPAD and conversion time from MCI to ADD. BPAD was increased in heavy alcohol drinkers compared to non-/sporadic (p = .014) and moderate (p = .040) drinkers. Conclusions Brain age and associated BPAD have the potential to serve as indicators for, and to evaluate the impact of lifestyle modifications or interventions on, brain health.

show abstract

Biological Brain Age Prediction Using Machine Learning on Structural Neuroimaging Data: Multi-Cohort Validation Against Biomarkers of Alzheimer’s Disease and Neurodegeneration stratified by sex

Cited by 4 publications

References 86 publications

A deep neural network estimation of brain age is sensitive to cognitive impairment and decline

A deep neural network estimation of brain age is sensitive to cognitive impairment and decline

Analyzing wav2vec embedding in Parkinson’s disease speech: A study on cross-database classification and regression tasks

Brain age as a biomarker for pathological versus healthy ageing – a REMEMBER study

Contact Info

Product

Resources

About