Alzheimer’s disease (AD), a predominant form of dementia that accounts for 60 to 70 percent of cases in the elderly population. AD significantly affects daily functioning, memory, cognition, and behaviour, presenting a substantial global health challenge with approximately 50 million dementia cases worldwide and an annual incidence of 10 million new cases. Using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort, we conduct a systematic analysis of several machine learning models to predict genetic variance linked to Late-Onset Alzheimer’s Disease (LOAD). Our experimental results demonstrate that the most effective models achieve an impressive 72 percent area under the Receiver Operating Characteristic (ROC) curve in the evaluation of LOAD genetic risk. This highlights the promise of machine learning models as valuable tools for assessing the genetic susceptibility to LOAD. Furthermore, our exploration into the strategic selection of learning models unveils the potential for identifying novel genetic markers linked to the disease. This improves our ability to anticipate outcomes and advances our understanding of the fundamental processes behind Alzheimer’s disease. The findings presented herein contribute to the evolving field of precision medicine by offering insights into the application of machine learning in understanding and predicting the genetic factors associated with LOAD.
Keywords: Machine learning; alzheimer’s disease; alzheimer’s disease neuroimaging initiative; late-onset alzheimer’s disease machine learning; alzheimer’s disease; alzheimer’s disease neuroimaging initiative; late-onset alzheimer’s disease.
Citation : Rekha Gangula and Dayakar Thalla, “Machine Learning in Predicting Alzheimer’s Disease: Exploring Applications and Advancements”, Journal of Computer Allied Intelligence (JCAI), vol.02, no.01, pp.1-7,2024.