Detection of Neurocognitive Impairment in the Elderly Using TabNet Algorithm

As the global population ages, neurocognitive disorders like the Alzheimer's disease which is a form of dementia, and several other forms of dementia are becoming more commonplace. These conditions significantly impair the quality of life for patients and present a growing challenge for healthcare systems worldwide. Early detection is critical to improving patient outcomes, yet traditional diagnostic methods often struggle to catch on to the very early and subtle signs of decline in cognitive function. In this paper, a deep learning algorithm, TabNet, is used on clinical data to detect neurocognitive impairment. Before applying the algorithm, Exploratory Data Analysis is also performed to get a better understanding of the dataset. This research is mainly focused on people with family history. The TabNet model, designed especially for tabular data easily and effectively learns the importance of each feature and their inter-dependencies. In future work, it is aimed to make this research more holistic by using Vision Transformer models to analyze MRI images. This study currently focuses on clinical data, which includes demographic, lifestyle, and cognitive assessments, and demonstrates how TabNet can be used to identify signs of cognitive impairment. Working with critical features, this study also focuses on data pre-processing and feature selection. This project's main aim is to utilize AI in healthcare to help with diagnoses that are otherwise somewhat difficult to make.