Computing Degree Show 2018

Morphological Brain Age Prediction using Multi-view Brain Networks Derived from Cortical Morphology in Healthy and Disordered Populations

Brain development and ageing are complex and dynamic processes that unfold over years on multiple levels. Recent studies using images of the brain (MRI) have revealed a disparity between the chronological brain age and the ‘data-driven’ brain age. In particular, predicting the ‘brain age’ from connectomic data might help identify relevant connectional biomarkers of neurological disorders that emerge during the lifespan.

While prior brain-age prediction studies have relied exclusively on structural or functional data, here we propose to predict the morphological age of the brain by solely using T1-w images in both healthy and disordered populations.

Our protocol includes the following steps: (i) building multi-view morphological brain networks (M-MBN), (ii) Feature extraction and selection, (iii) training a machine-learning model to predict age from M-MBN data, and (iv) utilising the predicted age to make clinical discoveries related to brain age.

We evaluated our framework on 341 subjects. We demonstrated the proposed protocol outperforms existing approaches in brain-age (and even behaviour) prediction. More importantly, we discovered brain connectional morphological features that fingerprint the age of brain cortical morphology in both disordered and healthy individuals. Lastly, we found that the discrepancy between predicted age and chronological age captures varying disordered connectional brain morphology patterns.

Name: Josh Corps
Supervisor: Dr Islem Rekik

GitHub
LinkedIn
Basira Lab