With the ever-increasing amount of data that is collected in medicine, it is time to take advantage of sophisticated machine learning algorithms to analyze past data to pave the way for better treatment planning and decision making in the future. This course will provide the attendees with an introductory, hands-on experience in building and training a versatile machine learning algorithm, namely, random forest. This algorithm is particularly advantageous in identifying at risk patients as it allows for easily interpretable results. Specifically, the goal is to use a dataset of patients to classify at risk patients and further evaluate the patterns to determine the most important contributing factors to complications. Modeling techniques to improve model performance will be discussed. The objectives of the course are as follows:

• To provide introductory information about machine learning algorithms and classification techniques
• To discuss various applications of the techniques in medicine
• To develop and train random forest classifiers with respect to a response variable
• To identify the most important contributing factors giving rise to the observed values of the response variable
• To learn techniques to improve model performance
• To analyze the results and draw actionable insights