Background:
Lung cancer has become a major cause of cancer-related deaths. Detection
of potentially malignant lung nodules is essential for the early diagnosis and clinical management
of lung cancer. In clinical practice, the interpretation of Computed Tomography (CT) images is
challenging for radiologists due to a large number of cases. There is a high rate of false positives in
the manual findings. Computer aided detection system (CAD) and computer aided diagnosis systems
(CADx) enhance the radiologists in accurately delineating the lung nodules.
Objectives:
The objective is to analyze CAD and CADx systems for lung nodule detection. It is
necessary to review the various techniques followed in CAD and CADx systems proposed and implemented
by various research persons. This study aims at analyzing the recent application of various
concepts in computer science to each stage of CAD and CADx.
Methods:
This review paper is special in its own kind because it analyses the various techniques
proposed by different eminent researchers in noise removal, contrast enhancement, thorax removal,
lung segmentation, bone suppression, segmentation of trachea, classification of nodule and nonnodule
and final classification of benign and malignant nodules.
Results:
A comparison of the performance of different techniques implemented by various researchers
for the classification of nodule and non-nodule has been tabulated in the paper.
Conclusion:
The findings of this review paper will definitely prove to be useful to the research
community working on automation of lung nodule detection.
Lung Nodule Detection from Feature Engineering to Deep Learning in Thoracic CT Images: a Comprehensive Review
