Braintrian: An Ai-Based Application To Support Mild Cognitive Impairment (Mci) Patients

Abstract

This thesis covers the development of a mobile game application intended to stimulate the cognitive functions of people with Mild Cognitive Impairment (MCI). Mild cognitive impairment is a condition that affects memory and other cognitive abilities. It is sometimes regarded as a transitional stage between normal age-related cognitive decline and more serious illnesses such as Alzheimer’s disease. The fundamental goal of this project is to improve cognitive capacities and give a useful tool for people who are having cognitive issues. By immersing users in cognitive exercises, the game hopes to delay the progression of cognitive decline and improve general mental functioning. This project expands on the existing Brain Train app by offering a new game mode for language training. Language skills are essential for good communication and overall cognitive health, and the new game mode focuses exclusively on them. The methodology uses advanced word embedding techniques to create a word association game that promotes semantic memory and improves language processing skills. Semantic memory, or the recall of broad world information, including concepts and facts, is a critical area damaged by MCI. Improving this element of memory can have a substantial impact on people’s daily lives and interactions. The Word2Vec technique, which was originally created by Google, is used in the word embedding process. Word2Vec is a sophisticated model that converts words into continuous vector representations while preserving semantic links between them. Using this technology, the game may provide relevant and demanding word association activities that promote cognitive processing. The Word2Vec model can be trained on various corpuses of literature to ensure accurate and relevant word associations. The final model and its use in developing a word semantic relationship game produce positive results. However, further improvement is required. Future enhancements may include increasing the Word2Vec training corpus, improving the user interface, and introducing variable difficulty levels. This thesis enhances mobile technology for cognitive rehabilitation by providing a fun and engaging approach for people with MCI to exercise their minds, while also demonstrating the effectiveness of word embedding techniques in cognitive training.