Applying Supervised Learning Algorithms and a New Feature Selection Method to Predict Coronary Artery Disease

TL;DR

This paper explores predicting coronary artery disease using genetic SNP data, comparing supervised learning algorithms and introducing a new feature selection method, demonstrating superior results with Random Forest and MTD Feature Selection.

Contribution

It introduces the Mass Transportation Distance (MTD) feature selection method and compares its effectiveness with Random Projections in predicting coronary artery disease.

Findings

MTD Feature Selection with Random Forest achieves 66.6% accuracy.

Area under ROC curve reaches 0.8562, surpassing previous results.

Random Forest outperforms k-NN and Random Projections in this context.

Abstract

From a fresh data science perspective, this thesis discusses the prediction of coronary artery disease based on genetic variations at the DNA base pair level, called Single-Nucleotide Polymorphisms (SNPs), collected from the Ontario Heart Genomics Study (OHGS). First, the thesis explains two commonly used supervised learning algorithms, the k-Nearest Neighbour (k-NN) and Random Forest classifiers, and includes a complete proof that the k-NN classifier is universally consistent in any finite dimensional normed vector space. Second, the thesis introduces two dimensionality reduction steps, Random Projections, a known feature extraction technique based on the Johnson-Lindenstrauss lemma, and a new method termed Mass Transportation Distance (MTD) Feature Selection for discrete domains. Then, this thesis compares the performance of Random Projections with the k-NN classifier against MTD Feature Selection and Random Forest, for predicting artery disease based on accuracy, the F-Measure, and area under the Receiver Operating Characteristic (ROC) curve. The comparative results demonstrate that MTD Feature Selection with Random Forest is vastly superior to Random Projections and k-NN. The Random Forest classifier is able to obtain an accuracy of 0.6660 and an area under the ROC curve of 0.8562 on the OHGS genetic dataset, when 3335 SNPs are selected by MTD Feature Selection for classification. This area is considerably better than the previous high score of 0.608 obtained by Davies et al. in 2010 on the same dataset.