Binary Rao Algorithm-Based Approach for Diagnosing Alzheimer’s Disease

Alzheimer’s disease (AD) is a neurodegenerative disorder that affects millions of people worldwide. It affects the patient’s cognitive abilities such as judgment, memory, and learning abilities. Early and accurate diagnosis of AD is crucial for effective treatment and disease management. Researchers have created a dataset specially to detect AD based on handwriting. This dataset has 450 features. All of the features are not equally important. Therefore, to select the set of relevant features, feature selection techniques need to be introduced here which will improve the classifier performance. Taking the challenge of optimal feature selection for automatic diagnosis of AD on handwriting-based evaluation, this work uses the Rao optimization algorithms (Rao-1, Rao-2, and Rao-3). The algorithms have been modified to solve the problem of binary feature selection by introducing transfer function. The study uses the wrapper method of feature selection and classification accuracy used as the objective function which needs to be maximized. Significant improvement in the accuracy has been achieved by applying the Rao optimization algorithm. The results obtained from the study are also compared with existing work on the dataset. Comparison with existing work also shows the better performance of wrapper method-based Rao algorithm of feature selection for classification of patients with AD.