Abstract

Paper Title/ Authors Name Download View

DEVELOPING POST OPERATIVE SURVIVAL PREDICTION MODELS FOR LUNG CANCER PATIENTS USING RECURSIVE PARTITIONING TECHNIQUES

Corey Stone


The purpose of this report is to analyse the factors that may lead to a greater understanding of the post operative survival rate in lung cancer patients who undergo lung resection surgery. This paper will explore the relationships between data collected prior to the operation, and whether the patient is still alive 12 months after the operation. Cutting edge Recursive Partitioning techniques such as Discriminant Analysis, Decision Tree Method, Random Forest, and Artificial Neural Networks will form the basis of the methodologies used for building a predictive model. A random sample of 224 patients who underwent surgery between the years of 2007 and 2011 in Wrawclaw, Poland will form the data set. Of the 224 sample of patients, 63 did not survive more than 12 months after the operation, the remaining 161 survived. Artificial Neural Network and Random Forest intelligence methods are found to achieve the most accurate classification rates. These models can be used by medical professionals to provide an accurate survival outcome for lung cancer patients who elect to undergo lung resection surgery.