Alzheimer’s disease (AD) is an incurable neurodegenerative disorder that affects an increasing number of elderly individuals every year. Symptoms generally begin with mild cognitive impairment (MCI) and worsen as the disease progresses. One way to delay the onset and severity of symptoms of AD is with early diagnosis. This can be difficult with a disease such as AD because sometimes symptoms do not become apparent until the disease is already prevalent in the brain. There is increasing popularity with the use of machine learning techniques such as support vector machines (SVM) to try to predict the onset of the disease before severe cognitive impairment begins. Our goal is to train a SVM on subjects diagnosed with MCI and AD to predict the MMSE scores of certain individuals. We are using a sample of 202 non-Hispanic Caucasians and African Americans from the Oasis-3 data set to train our SVM that have undergone MRI scans as well as mini-mental state exam (MMSE) cognitive testing to confirm diagnosis prediction. A MMSE score of 23 or below is indicative of cognitive impairment along with structural deficits in areas of the medial temporal lobe such as the entorhinal cortex, hippocampus, limbic system, and neocortex (confirmed by MRI scans). We believe with these two factors we can train the machine to accurately predict the MMSE scores of patients with AD and with early diagnosis, help improve the life and care of patients diagnosed.
Powered by Acadiate
© 2011-2024, Acadiate Inc. or its affiliates · Privacy