Predicting Brain Age using Machine Learning and Deep Learning Algorithms a Comprehensive Evaluation

Utupalli Lohith1
Department of Computer Science and
Engineering
Sri Venkateswara College of Engineering,
Karakambadi
Tirupati, India, 517509
lohithutupalli@gmail.com

Dr. S. Sajida2
Department of Computer Science and
Engineering
Sri Venkateswara College of Engineering,
Karakambadi
Tirupati, India, 517509
Sajida.s@svce.edu.i

Abstract — The Brain Age Estimation (BAE) project has developed a machine learning framework that predicts neurological age, or "brain age," using structural MRI data from large neuroimaging datasetslike IXI, ADNI, and UK Biobank. The system models  brain morphology changes, such as grey matter volume and white matter integrity, to estimate brain age and calculate the Brain Age Gap (BAG), thedifference between predicted and chronological age. A positive BAG indicates accelerated aging, linked to disorders like Alzheimer's and Parkinson's, while anegative BAG suggests preserved brain health. The pipeline includes automated feature extraction, machine learning models like Relevance Vector Regression, and cross-site validation. A Dash-based dashboard with SHAP-driven explainability supports clinical interpretation, and real-time prediction is enabled through manual feature input. The system achieves a mean absolute error of 3.8 years and acorrelation of r = 0.94, making it a potential clinicaldecision-support tool. Keywords — Brain Age Estimation, Brain Age Gap,Structural MRI, Relevance Vector Regression, SHAP,CAT12.8, Neuroimaging, Biomarker, Bias Correction,Dash Dashboard.