SVM Based AutoEncoder for Detecting Dementia in Young Adults

Dementia's impact on cognitive function necessitates timely diagnosis for effective intervention. Understanding the need for timely detection, the proposed work integrates SVM's decision boundary determination and autoencoder's noise reduction capabilities. The proposed work advances in dementia detection in young adult. Results indicate promising performance, with the model achieving high accuracy around 85.33%. The ROC curve illustrates a balanced trade-off between sensitivity and specificity, while the precision-recall curve highlights effective classification. Importantly, the model surpasses existing literature, underscoring its practical utility. While acknowledging limitations, such as parameter fine-tuning, this study lays the groundwork for refining and expanding this innovative methodology. In summary, this research contributes to the urgent field of early dementia detection, potentially transforming patient care and intervention strategies.