Yusuf Musa Malgwi, Ibraim Goni* and Bamanga Mahmud Ahmad

Department of Computer Science, Faculty of Physical Science, Modibbo Adama University, Yola, Nigeria

*Corresponding Author: Ibraim Goni, Department of Computer Science, Faculty of Physical Science, Modibbo Adama University, Yola, Nigeria.

Received: July 08, 2022; Published: August 04, 2022

Abstract

Lung cancer is one of the deadly diseases in recent years. However, research proved that detection in an early stage improved the chances of survival. The disease is identified using nodules attached to lung walls and lung parenchyma. Nodules on the lungs are the major sign and symptoms for identifying lung cancer. The aim of this research work was to detect lung cancer using convolutional neural network. CT scanned images were obtained and form the datasets for training and testing the models then nodules are classified as benign or malign. The model helps in improving accuracy in identifying nodules in the lungs. Different classifiers such as Multilayered Perceptron and CNN classifiers are used in comparative analysis. Based on these findings it was conclude that the approach of feature extraction with CNN decreases the false positive rate significantly compared to the existing classification methods.

Keywords: Lung Nodules; CNN; Lung Cancer; Multilayered Perceptron