imageRecognitionDetectLungCancerPython

TABLE OF CONTENTS

• Abstract
• Additional Resources
• Picking The Algorithm
• Dependencies Required
• Steps Required
• The Findings
• Methodology
• General Findings

ABSTRACT

• This repo is dedicated to using data science methods in Python to analyze low-dose CT images of patient lungs to attempt to determine whether cancer will develop within a year.
• This data comes from the annual Kaggle competition that took place in 2017 and is linked
• We use various machine learning methods such as K-Nearest Neighbors, Neural Networks, Principal Component Analysis and Decision Trees/Random Forests. We will look closely at each of these methods and choose the best method based on accuracy + insight into the relationships between expalnatory variables
• First we sample these methods on one of the most used, well-known set of images out there: The MNIST data set. Each method is used and analyzed for effectiveness (currently the KNN code is not working)
• After we dip our feet in each of these methods, we create a database to extract as much data as we choose (the data set is 66.88 GB after all...) and then go HAM on Machine Learning!!!

ADDITIONAL RESOURCES

We used the following documentation to further educate ourselves through the process of this project.

• The MNIST program was built using the following resources:

• Principal Component Analysis

• Neural Networks

Introduction

-This project originated from the UCSB Project Group in Winter 2017. The members that contributed to this are