Machine learning approaches for the computer aided diagnosis and prediction of Alzheimer's disease based on clinical data

Alzheimers is an irreversible brain disease that impairs memory, thinking, and behavior and leads, ultimately, to death. It is a major public health problem in the elder population and has a huge impact on society. It is useful to diagnose AD as early as possible, in order to improve the quality of life of the patient and their care takers. In this thesis we analyze the performance of different machine learning methods for the task of classifying different subjects cognitive status as normal (NL), mild cognitive impairment (MCI) or Alzheimers disease (AD). Machine learning methods are used to improve the performance of computer aided diagnosis of Alzheimers disease using clinical data. Also, feature subset selection methods are used to eliminate redundant attributes and retain the most relevant features. The results are very promising and demonstrate the utility of machine learning methods in this domain. The second part of the study is to predict the possible conversion from MCI to AD. We conducted many experiments with various learning algorithms and achieved performance levels comparable to the previously published results.