Deep Learning Analysis of Neuroimaging (MRI) Data

Le, Vu

Student Work

Deep Learning Analysis of Neuroimaging (MRI) Data

Public Deposited

Alzheimer’s disease has become the biggest medical challenge in modern neurology, due to its prevalence and socio-economic burdens. The diseases have affected millions of people, and the number is expected to rise. Alzheimer’s disease can lead to decrease in cognitive ability, memory loss, and daily functioning impairment. Therefore, it is important to have a proper method for early diagnosis and timely intervention for AD. In recent years, Predictive Modeling and Pattern Recognition have become more popular in neuroscience that allows researchers to analyze early symptoms of diseases. This method can be very helpful in providing a snapshot of patents at baseline and tracking changes over time that signal the progression of Alzheimer’s diseases. These predictive cognitive and functional assessments from the pattern can also lead to more evidence to understand the disease and provide better intervention and treatments. This study aims to utilize datasets that focus on baseline measurements and changes over time in cognitive and functional assessments among patients to predict outcomes related to Alzheimer’s disease using publicly available datasets. The goal is to use the cognitive and functional score to find the correlation with the progression of Alzheimer’s disease over time. Another important approach to analyze the early symptoms of Alzheimer’s disease is the incorporation of MRI, which provides a detailed view of the brain structure and functionality. MRI scans are invaluable for capturing the baseline snapshot of patient’s brain health and for monitoring anatomical changes over time, such as the progression of the brain atrophy associated with Alzheimer’s disease. MRI images can reveal the brain structure changes pattern, which combining with the cognitive score analysis, can improve the predictive models, which used to forecast the development of Alzheimer's diseases. This approach can provide deeper insights into the disease and provide important information for treatment. We hope to contribute a broader understanding of how early detection and continuous monitoring influence treatment strategies and patient care.

This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator