Open AccessDeep Learning Technique for Detecting NSCLC
Author(s) -
Bhargav Hegde*,
Mahesh Hegde,
C Chetan,
P Dayananda
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.c6540.098319
Subject(s) - lung cancer , computer science , artificial intelligence , artificial neural network , epoch (astronomy) , cancer , deep learning , machine learning , oncology , computational biology , data mining , medicine , biology , stars , computer vision
The lung cancer is one of the major cancers in the world. In lung cancer we have two main types. They are small cell lung cancer and non-small cell lung cancer. In this paper we mainly concentrated on the detection of non-small cell lung cancer. There are several types in NSCLC and we have several stages in NSCLC. The flow of proposed paper consist the following steps: (1) Background: Here we describe the different types of lung cancer and mainly about NSCLC; (2) Methods: To find the NSCLC, we are using the Recurrent Neural Network (RNN); (3) Results: After the training and prediction of the model, we will get the final result as weather the given patient suffering from NSCLC or not; and (4) Conclusions: The given model is working for all the possible datasets and the training accuracy is 88%. The accuracy of the model is mainly depends on the epoch value. For ideal epoch value the accuracy of the model is high. Dataset: The datasets are taken from the NCBI website. We have used the nucleotide datasets of the NCBI website. The datasets are open source and easily accessible. We have used the DNA sequence data of the human genome data. All the NSCLC patients data are taken as positive data and human reference gene data are taken as negative data