AI-BASED MEDICAL IMAGE ANALYSIS FOR EARLY DETECTION OF NEUROLOGICAL DISORDERS USING DEEP LEARNING

Abstract

Neurological disorders such as Alzheimer’s disease (AD), Parkinson’s disease (PD), and brain tumors remain among the primary contributors to global disability and mortality. Early and precise diagnosis is essential for effective intervention and improved patient outcomes. However, conventional diagnostic methods rely heavily on manual interpretation of neuroimaging by radiologists, which can be time-consuming, subjective, and susceptible to human error. The emergence of artificial intelligence (AI), particularly deep learning (DL), offers a transformative solution through automated and high-accuracy medical image analysis. This study proposes an AI-driven diagnostic framework that leverages EfficientNetB0, a lightweight yet high-performing convolutional neural network (CNN), to classify neurological conditions using brain MRI and CT scans. The model was trained and fine-tuned on a labeled dataset comprising three categories: Alzheimer’s disease, Parkinson’s disease, and healthy controls. It achieved an overall classification accuracy of 95%, demonstrating its effectiveness in differentiating between pathological and non-pathological cases. The model reported a precision, recall, and F1-score of 0.97 for AD, a recall of 0.98 for control cases, and a precision of 0.96 with a recall of 0.85 for PD. Additionally, the area under the ROC curve (AUC) was 0.98 for AD, 0.95 for controls, and 0.92 for PD, indicating strong discriminative performance. These findings highlight the potential of EfficientNetB0 as a scalable, efficient, and accurate tool for supporting early detection and diagnosis of neurological disorders in clinical practice. This work contributes to advancing AI-assisted healthcare solutions aimed at improving diagnostic speed and consistency in neuroimaging analysis.