Introduction: Alzheimer’s disease, a progressive and debilitating neurological disorder, significantly impacts the quality of life, particularly in the elderly. Given the increasing prevalence of this disease, developing accurate methods for early prediction and diagnosis is crucial. This study aims to identify key factors influencing Alzheimer’s disease prediction using novel feature selection techniques and machine learning models. The primary objective of this study is to contribute to the development of more accurate diagnostic tools, thereby improving the management and treatment of this disease.
Methods: In this study, we employed ten wrapper-based feature selection methods to identify the most accurate and relevant features of Alzheimer’s disease. The performance of these models was evaluated using popular machine learning algorithms and standard evaluation metrics such as accuracy, precision, recall, F1-score, and ROC curve analysis. All evaluations were conducted on the ADNI standard Alzheimer’s disease dataset.
Results: The influential features included cognitive test results (e.g., Mini-Mental State Examination), functional assessments, patient-reported memory and behavioral problems, and activities of daily living scores, which were identified as key indicators for Alzheimer’s disease diagnosis.
Discussion: The results demonstrate that employing novel feature selection techniques and machine learning algorithms can lead to the development of more accurate models for predicting Alzheimer’s disease. These findings can contribute to improving early diagnosis and management of this diseas.
Type of Study:
Research |
Subject:
Special Received: 2024/09/6 | Accepted: 2024/12/14 | Published: 2024/12/20