Brain Aging in Pediatric Development and Neurological Conditions: A Machine Learning Study
Poster No:
882
Submission Type:
Abstract Submission
Authors:
Timur Latypov1, Hrishikesh Suresh2, Karim Mithani2, Flavia Venetucci Gouveia1, George Ibrahim1
Institutions:
1Hospital for Sick Children, Toronto, Ontario, 2University of Toronto, Toronto, Ontario
First Author:
Co-Author(s):
Introduction:
Brain age, derived from medical imaging, is an intuitive and inherently personalized metric of brain health. In adults, deviations between predicted brain age and chronological age, known as the brain age gap, are associated with various neurological and psychiatric conditions (Cole, 2017). Applying this concept to pediatric populations presents unique challenges due to the dynamic, non-linear nature of brain development. Unlike adults, where gray matter (GM) volume decreases steadily, pediatric brains follow a bell curve pattern, peaking around ages 4–6 and then gradually declining as pruning and reorganization processes enhance efficiency (Fig. 1A) (Bethlehem, 2022). These developmental trajectories vary across brain regions, necessitating precise, age-specific modeling. In this study, we aimed to train a robust machine learning (ML) model to predict brain age in pediatric and adult populations and applied it to evaluate brain aging patterns and brain age gap (BAGap) in children with cerebral palsy and temporal lobe epilepsy.
Methods:
We utilized publicly available datasets, including the Calgary Preschool MRI dataset, the Cambridge Centre for Ageing and Neuroscience (CamCAN) dataset, and the Healthy Brain Network dataset, comprising 2,719 participants aged 1.9–90 years (Mean 20.85 y, Median 11.83 y, 62% M) (Taylor, 2017; Reynolds, 2020; Richie-Halford, 2022). T1-weighted MR imaging data were processed using Freesurfer 7 to extract cortical and subcortical GM metrics (Dale, 1999). An XGBoost regression model was trained to predict chronological age, with performance evaluated using 10-fold cross-validation and external validation on a cohort of healthy controls acquired in our hospital (n=363, age 1.9-18.1 y, 47% M)
To assess utility, the model was applied to the data from children with cerebral palsy (n=87, age 4-17.6 y, 56% M) and temporal lobe epilepsy (n=53, age 2.8-17.9 y, 61%M). Brain age gaps (BAGap=predicted age - chronological age) were computed and compared to age- and sex-matched healthy controls from the local healthy controls' cohort (Fig. 1B).
To assess utility, the model was applied to the data from children with cerebral palsy (n=87, age 4-17.6 y, 56% M) and temporal lobe epilepsy (n=53, age 2.8-17.9 y, 61%M). Brain age gaps (BAGap=predicted age - chronological age) were computed and compared to age- and sex-matched healthy controls from the local healthy controls' cohort (Fig. 1B).
Results:
The brain age prediction model achieved a mean absolute error (MAE) of 3.51 years and an R² of 0.93 on the test subset of the development data. External validation on unseen cohort yielded an MAE of 2.16 years, confirming the generalizability and robustness of the model (Fig. 1C).
Pediatric patients with cerebral palsy and temporal lobe epilepsy demonstrated significantly larger (p<0.0001) brain age gaps than age- and sex-matched healthy controls (BAGapTLE=7.7 y, BAGapCP=3.15 y). Positive gap suggests that brains of affected individuals appear older than their chronological age. This effect was more pronounced in the temporal lobe epilepsy cohort, likely reflecting the ongoing impact of epilepsy on brain structure and function (Fig. 1D,E).
Pediatric patients with cerebral palsy and temporal lobe epilepsy demonstrated significantly larger (p<0.0001) brain age gaps than age- and sex-matched healthy controls (BAGapTLE=7.7 y, BAGapCP=3.15 y). Positive gap suggests that brains of affected individuals appear older than their chronological age. This effect was more pronounced in the temporal lobe epilepsy cohort, likely reflecting the ongoing impact of epilepsy on brain structure and function (Fig. 1D,E).
Conclusions:
This study demonstrates that brain age modeling effectively captures typical and atypical developmental trajectories in pediatric populations. Larger brain age gaps in cerebral palsy and epilepsy suggest loss of gray matter, highlighting the potential of brain age metrics as biomarkers for neurodevelopmental health. Future efforts should incorporate additional clinical and genetic data to refine the model and explore therapeutic interventions.
Disorders of the Nervous System:
Neurodevelopmental/ Early Life (eg. ADHD, autism)
Lifespan Development:
Aging 1
Early life, Adolescence, Aging
Normal Brain Development: Fetus to Adolescence
Modeling and Analysis Methods:
Classification and Predictive Modeling 2
Keywords:
Aging
Computational Neuroscience
Computing
Cortex
DISORDERS
Epilepsy
Machine Learning
STRUCTURAL MRI
Other - Cerebral Palsy
1|2Indicates the priority used for review
·Figure 1. Study summary
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.
Bethlehem, R. A. I., Seidlitz, J., White, S. R., Vogel, J. W., Anderson, K. M., Adamson, C., Adler, S., Alexopoulos, G. S., Anagnostou, E., Areces-Gonzalez, A., Astle, D. E., Auyeung, B., Ayub, M., Bae, J., Ball, G., Baron-Cohen, S., Beare, R., Bedford, S. A., Benegal, V., Beyer, F., … Alexander-Bloch, A. F. (2022). Brain charts for the human lifespan. Nature, 604(7906), 525–533. https://doi.org/10.1038/s41586-022-04554-y
Taylor, J. R., Williams, N., Cusack, R., Auer, T., Shafto, M. A., Dixon, M., Tyler, L. K., Cam-Can, & Henson, R. N. (2017). The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) data repository: Structural and functional MRI, MEG, and cognitive data from a cross-sectional adult lifespan sample. NeuroImage, 144(Pt B), 262–269. https://doi.org/10.1016/j.neuroimage.2015.09.018
Reynolds, J. E., Long, X., Paniukov, D., Bagshawe, M., & Lebel, C. (2020). Calgary Preschool magnetic resonance imaging (MRI) dataset. Data in brief, 29, 105224. https://doi.org/10.1016/j.dib.2020.105224
Richie-Halford, A., Cieslak, M., Ai, L., Caffarra, S., Covitz, S., Franco, A. R., Karipidis, I. I., Kruper, J., Milham, M., Avelar-Pereira, B., Roy, E., Sydnor, V. J., Yeatman, J. D., Fibr Community Science Consortium, Satterthwaite, T. D., & Rokem, A. (2022). An analysis-ready and quality controlled resource for pediatric brain white-matter research. Scientific data, 9(1), 616. https://doi.org/10.1038/s41597-022-01695-7
Dale, A. M., Fischl, B., & Sereno, M. I. (1999). Cortical surface-based analysis. I. Segmentation and surface reconstruction. NeuroImage, 9(2), 179–194. https://doi.org/10.1006/nimg.1998.0395