Abstract - Cancer is the second leading cause of death globally and among all the cancers, Lung cancer is considered to be the major reason for all cancer-related deaths. Detecting lung cancer at early stages can improve the survival rate, but it is a difficult task as lung cancer shows very few symptoms. To diagnose lung cancers at an earlier stage, Computed Tomography (CT) images are used. Nowadays medical decision-making is performed using only CT scan images and an automatic assessment of Computer-Aided Diagnosis (CAD) system. Computer Aided Diagnosis (CAD) system is a
medical diagnosis tool that is very useful for today’s medical imaging practicality. The primary aim of this work is to develop an advanced computer-aided diagnosis (CAD) system using deep learning algorithms that will efficiently extract data from CT scan images and provide precise and timely diagnosis of lung cancer. The work is divided into three phases - segmentation, feature extraction and classification. The CT scan images are segmented using (OTSU) Thresholding. This work focuses on utilizing the deep learning techniques, namely Convolutional Neural Network (CNN) for feature extraction and recurrent neural network (RNN-LSTM) for lung cancer classification and obtain high accuracy.
Keywords - Lung Cancer Diagnosis, CT Scan Images, Deep Learning, Convolutional Neural Network, Recurrent Neural Network and Long Short Term Memory.
|
DOI:
10.17148/IJARCCE.2021.10756
