Automatic Detection of Pulmonary Nodules using Three-dimensional Chain Coding and Optimized Random Forest

May Phu Paing,Supan Tungjitkusolmun,Sarinporn Visitsattapongse,Kazuhiko Hamamoto,Chuchart Pintavirooj

doi:10.3390/app10072346

Abstract

The detection of pulmonary nodules on computed tomography scans provides a clue for the early diagnosis of lung cancer. Manual detection mandates a heavy radiological workload as it identifies nodules slice-by-slice. This paper presents a fully automated nodule detection with three significant contributions. First, an automated seeded region growing is designed to segment the lung regions from the tomography scans. Second, a three-dimensional chain code algorithm is implemented to refine the border of the segmented lungs. Lastly, nodules inside the lungs are detected using an optimized random forest classifier. The experiments for our proposed detection are conducted using 888 scans from a public dataset, and achieves a favorable result of 93.11% accuracy, 94.86% sensitivity, and 91.37% specificity, with only 0.0863 false positives per exam.

Highlights

Lung cancer is a disease that mainly affects the lungs, which is the principal organ of the respiratory system and performs vital activities for human survival
Cross-validation iterates the training and testing processes k times so that each of the k subsets have cross-validation iterates the training and testing processes k times so that each of the k subsets have appeared as a test set exactly once in order to validate the performance of the proposed method
Appeared as a test set exactly once in order to validate the performance of the proposed method

Summary

Introduction

Lung cancer is a disease that mainly affects the lungs, which is the principal organ of the respiratory system and performs vital activities for human survival. It usually starts once the epithelial cells of the bronchioles or alveoli grow abnormally. If the lungs do not carry out their functions properly, due to lung cancer, other parts of the body can be affected, which can eventually lead to death. This fact makes the mortality rate of lung cancer higher than those of other types of cancer. Effective ways to reduce the mortality of lung cancer include preventions, early detections, and precise treatments

Objectives

Methods

Results

Conclusion