Design of Fusion based Computer Aided Diagnosis Model for Lung Cancer Detection and Classification

Authors

  • N. Venkatesan, Dr. S. Pasupathy, Dr. B. Gobinathan

DOI:

https://doi.org/10.17762/msea.v71i3.262

Abstract

Lung cancer remains a serious illness and results in high mortality rate over the globe. Earlier identification of lung cancer based on computed tomography (CT) images helps to diagnose the disease and increase the survival rate. Computer aided diagnosis (CAD) models can be designed for lung cancer diagnosis using different processes such as preprocessing, segmentation, feature extraction, and classification. With this motivation, this study designs a fusion based CAD model for lung cancer detection and classification (FCAD-LCDC). The goal of the FCAD-LCDC technique is for detecting and classifying different types of lung cancer using CT images. The FCAD-LCDC technique involves Gaussian filtering (GF) as a preprocessing approachfor removing the noise that exists in the CT images. Besides, a fusion of Inceptionv3 based deep features and histogram of gradients (HOG) based handcrafted features take place for feature extraction. Moreover, extreme gradient boosting (XGBoost) classifier is used to allot proper class labels to the applied CT images. In order to showcase the better outcome of the FCAD-LCDC technique, a series of simulations were carried out on the benchmark dataset and the results portrayed the supremacy of the FCAD-LCDC technique over the other CAD models interms of different measures.

Downloads

Published

2022-06-09

How to Cite

N. Venkatesan, Dr. S. Pasupathy, Dr. B. Gobinathan. (2022). Design of Fusion based Computer Aided Diagnosis Model for Lung Cancer Detection and Classification. Mathematical Statistician and Engineering Applications, 71(3), 992 –. https://doi.org/10.17762/msea.v71i3.262

Issue

Section

Articles