Lung Cancer Stages Detection Using Hybrid Deep Learning Approach

Abstract

Cancer is the underlying cause of a significant number of deaths worldwide, with lung cancer being the cause of the highest mortality rate. Computer tomography scan is used by radiologists to diagnose and monitor the development of cancer in the body. Visual analysis of the database will lead to the diagnosis of cancer at a later stage, leading to late treatment of cancer that only raises the mortality risk of cancer. Image recognition techniques can also be used for early identification of cancer. In this article, a lung cancer identification algorithm is proposed using mathematical morphological operations for segmentation of the lung area of interest, from which Haralick features are extracted and used to identify cancer through artificial neural networks.