Neural and Contextual Predictors of Long-Term Physical Activity in Aging With Cardiovascular Disease
Poster No:
717
Submission Type:
Abstract Submission
Authors:
Naga Thovinakere1, Satra Ghosh2, Yasser Iturria-Medina3, Maiya Geddes1
Institutions:
1McGill University, Montreal, Quebec, 2Massachusetts Institute of Technology, Cambridge, MA, 3McGill, Montreal, Quebec
First Author:
Co-Author(s):
Introduction:
Physical activity is crucial for preventing cognitive decline, stroke and dementia in older adults1. An understanding of factors that influence physical activity engagement are therefore needed to improve health outcomes and decrease the burden on healthcare systems. A new cardiovascular diagnosis represents a pivotal moment for initiating lifestyle changes2. However, sustaining long-term physical activity is challenging, and its underlying mechanisms remain poorly known. This study identifies neural and behavioral predictors of sustained physical activity following a new cardiovascular diagnosis.
Methods:
We applied support vector machine (SVM) learning to predict changes in moderate-to-vigorous physical activity (MVPA) over four years in 295 cognitively unimpaired older adults from the UK Biobank, a large longitudinal population cohort. Inclusion criteria were: 1) cognitively unimpaired at enrollment; 2) reported a new cardiovascular diagnosis (i.e., hypertension, type II diabetes, dyslipidemia, cardiac angina or myocardial infarction) between baseline (T1; 2014) and follow-up over four years later (T2; 2019) (mean duration 4.2 years, SD 1.1); 3) did not meet the World Health Organization recommendation of 150 minutes/week of MVPA at baseline; and 4) age >= 60. Self-reported MVPA was measured using the Lifetime Total Physical Activity Questionnaire. Demographic variables, including age, sex, and years of education were included as covariates.
We evaluated three SVM models to predict future MVPA behavior as a continuous outcome: (i) demographic, cognitive, and contextual factors; (ii) baseline resting-state functional connectivity (RSFC) MRI; and (iii) a combined multimodal model integrating all predictors. We used Kernel SHAP (SHapley Additive exPlanations)3 to interpret feature contributions by calculating mean absolute SHAP values, weighted by model performance.
We evaluated three SVM models to predict future MVPA behavior as a continuous outcome: (i) demographic, cognitive, and contextual factors; (ii) baseline resting-state functional connectivity (RSFC) MRI; and (iii) a combined multimodal model integrating all predictors. We used Kernel SHAP (SHapley Additive exPlanations)3 to interpret feature contributions by calculating mean absolute SHAP values, weighted by model performance.
Results:
The combined multimodal model demonstrated the highest predictive power (r = 0.28, p = 0.001). Key predictors included access to greenspace, social support, retirement status, executive function, and RSFC features: between-network connectivity in the default mode and frontoparietal control networks, and enhanced within-network connectivity within the default mode network.
·Baseline RSFC features that predict future increase in MVPA after a new cardiovascular diagnosis in aging
Conclusions:
Our findings underscore the importance of social and structural health determinants of health while uncovering neural mechanisms that may support lifestyle modifications. This research advances understanding of successful physical activity behavior change and informs the design of interventions and health policies aimed at mitigating the negative health effects of cardiovascular disease and cognitive decline in later life.
Emotion, Motivation and Social Neuroscience:
Emotion and Motivation Other
Higher Cognitive Functions:
Decision Making 1
Lifespan Development:
Aging
Modeling and Analysis Methods:
fMRI Connectivity and Network Modeling 2
Keywords:
Cerebrovascular Disease
FUNCTIONAL MRI
Machine Learning
Other - Physical activity
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
2. Lane-Cordova, A. D., Jerome, G. J., Paluch, A. E., Bustamante, E. E., LaMonte, M. J., Pate, R. R., Weaver, R. G., Webber-Ritchey, K. J., Gibbs, B. B., & on behalf of the Committee on Physical Activity of the American Heart Association Council on Lifestyle and Cardiometabolic Health. (2022). Supporting Physical Activity in Patients and Populations During Life Events and Transitions: A Scientific Statement From the American Heart Association. Circulation, 145(4), e117–e128. https://doi.org/10.1161/CIR.0000000000001035
3. Lundberg, S. (2017). A unified approach to interpreting model predictions. arXiv preprint arXiv:1705.07874.