Abstract Classification, focusing on early detection and diagnosis of neurodegenerative conditions such as Parkinson's and Alzheimer's diseases. The methodology integrates regression algorithms for brain age estimation from structural MRI scans with convolutional neural networks (CNNs) for disease classification based on brain imaging data. Initially, regression models, including Linear Regression, Ridge Regression, Lasso Regression, Random Forest Regression, and Support Vector Regression, are trained and evaluated. The results demonstrate the efficacy of Ridge and Lasso Regression models in accurately estimating brain age, as indicated by low mean absolute error and high R-squared values. Significant disparities between predicted and actual brain ages serve as indicators of potential neurodegenerative conditions. In cases of notable deviations, a CNN-based classification model is employed to identify and differentiate between Parkinson's disease, Alzheimer's disease, and normal brain aging. The classification model, trained on a diverse dataset of MRI scans annotated with disease labels, achieves high accuracy, precision, recall, and F1-scores for both Alzheimer's and Parkinson's disease, indicating its effectiveness in disease classification. The methodology encompasses MRI data preprocessing, feature extraction, and model training using advanced machine learning techniques. Performance evaluation is conducted using metrics such as accuracy, sensitivity, and specificity. The results highlight the proposed approach's efficacy in accurately estimating brain age and distinguishing between normal aging and neurodegenerative diseases, with the CNN model achieving an accuracy of 99.69% in disease classification. This framework shows promise for early detection and diagnosis of neurodegenerative conditions, potentially facilitating timely interventions and improving patient outcomes. Future research may focus on refining the models, evaluating generalization on unseen data, and exploring interpretability analysis for deeper insights into disease classification decisions.
Keywords: Machine learning , deep learning, regression, classification.
|
DOI:
10.17148/IJARCCE.2024.13857
