Mechanical Embolus Removal in Cerebral Ischemia (MERCI) has been supported by medical trials as an improved method of treating ischemic stroke past the safe window of time for administering clot-busting drugs, and was released for medical use in 2004. The importance of analyzing real-world data collected from MERCI clinical trials is key to providing insights on the effectiveness of MERCI. Most of the existing data analysis on MERCI results has thus far employed conventional statistical analysis techniques. To the best of our knowledge, advanced data analytics and data mining techniques have not yet been systematically applied. To address the issue in this thesis, we conduct a comprehensive study on employing state of the art machine learning algorithms to generate prediction criteria for the outcome of MERCI patients. Specifically, we investigate the issue of how to choose the most significant attributes of a data set with limited instance examples. We propose a few search algorithms to identify the significant attributes, followed by a thorough performance analysis for each algorithm. Finally, we apply our proposed approach to the real-world, de-identified patient data provided by Erlanger Southeast Regional Stroke Center, Chattanooga, TN. Our experimental results have demonstrated that our proposed approach performs well.