My research project focuses on detecting Alzheimer's disease by analyzing language and voice patterns using machine learning, and is supervised by Professor Dean Ho, Head of the Department of Biomedical Engineering at the National University of Singapore.

According to the Institute of Mental Health, 1 in 10 individuals aged 60 and above in Singapore suffers from dementia. Doctors are on the lookout for ways in which they can identify Alzheimer's as soon as possible to help treat it. My research focuses on the analysis of language and speech patterns using artificial intelligence that could provide doctors with indicators of Alzheimer's and help patients seek treatment sooner.

This four-year research project involves writing a perspective paper that reviews the literature on the subject, writing a study design, identifying a relevant population group, taking audio recordings, doing AI modelling and analysis, and writing a research paper.

I have completed the CITI Program courses under requirements set by National University of Singapore: Health Information Privacy and Security (HIPS) - Basic Course and Good Clinical Practice - Basic Course. I’ve also completed the Ethics of AI short course by the London School of Economics.

Abstract of Perspective Paper (Draft)

Alzheimer's Disease (AD) is a progressive neurodegenerative disease associated with severe memory loss with language and speech impairment appearing as initial signs of the onset of AD. As there is no current cure for AD, early diagnosis is the only way to receive adequate treatment. The ability to detect AD automatically in a non-invasive manner will help identify patients with AD. 

This perspective paper reviews the literature on AD detection by analyzing language and speech impairment, highlighting which AI models differentiate healthy individuals from those with AD. It outlines how researchers have collected relevant data, processed it, and analyzed it with artificial intelligence. It lists which models have proved most accurate in identifying AD. The paper also identifies gaps in the current research. One gap is the need for doing such research in Singapore by addressing the fact that Singaporeans often speak Singlish. The paper concludes with recommendations for a study design in Singapore to take advantage of the encouraging use of language and speech analysis for AD identification.

Keywords : Alzheimer’s disease; language analysis; speech analysis; linguistic features; prosodic features; machine learning; Singlish; speech processing