Leveraging Deep Learning to Establish the Brain Age Gap as a Critical Diagnostic Biomarker for Neurological Disorders

Abstract

This thesis presents a novel deep learning-based framework for brain age classification and prediction using neuroimaging data. Brain age is a crucial biomarker for detecting neurological disorders such as Alzheimer s disease (AD), Mild cognitive impairment (MCI), and Parkinson s disease (PD). Structural magnetic resonance imaging (MRI) has been used to analyze age-related neuroanatomical changes in the brain, and the discrepancy between predicted brain age and chronological age offers valuable insights into brain health and potential abnormalities. Accelerated brain aging and atypical patterns are linked to neurological decline, necessitating accurate methods for brain age estimation. Manual interpretation of neuroimaging data is both time-consuming and subject to inter-observer variability. To address this, deep learning (DL) techniques have emerged as powerful tools in medical image analysis, particularly for brain age prediction. This thesis proposes an innovative deep learning framework that integrates convolutional neural networks (CNNs) and pretrained models for feature extraction from MRI images, enhancing the accuracy of brain age estimation. Additionally, multi-kernel functions are incorporated to improve generalization and robustness, while ensemble learning models are applied to enhance predictive performance. Our experimental results demonstrate that the proposed framework outperforms traditional machine learning algorithms in both brain age classification and estimation, achieving high accuracy. This research contributes to developing an automated, reliable, and efficient solution for the early detection of neurological diseases by mapping with brain age gap, improving patient outcomes, aiding healthcare practitioners, and reducing healthcare costs.