Automatic Lung Cancer Detection Using Artificial Intelligence

Session

Computer Science and Communication Engineering

Description

By far, lung cancer is the prominent cause of cancer deaths for both men and women around the world. In 2018, statistics for WHO (World Health Organization) showed that of 2.09 million people diagnosed with this disease, 1.76 million people have died. The survival rate increases if it is detected in its early stages. Taking into consideration the complexity of the problem, many computer-aided diagnosis systems that increase the survival rate have been proposed and developed. Driven by the notable success of deep learning in the area of complex image classification problems, this paper presents the use ResNet34, ResNet50, VGG16, and VGG19 convolutional neural network architectures or classifying images of patients with cancer. Moreover, to compare the performance evaluation Accuracy, Precision, Area Under Curve, and F1 score were calculated. In conclusion, ResNet50 architecture was estimated with the best result for this classification problem, with 87.04% Accuracy and 85.71% F1 score. The strategy of using pre-trained deep learning models proved to be pertinent to this problem.

Keywords:

Lung Cancer, Artificial Intelligence, Deep-Learning, Convolutional Neural Networks, ResNet50, Classification.

Session Chair

Bertan Karahoda

Session Co-Chair

Krenare Pireva

Proceedings Editor

Edmond Hajrizi

ISBN

978-9951-550-19-2

Location

Pristina, Kosovo

Start Date

26-10-2019 1:30 PM

End Date

26-10-2019 3:00 PM

DOI

10.33107/ubt-ic.2019.268

This document is currently not available here.

Share

COinS
 
Oct 26th, 1:30 PM Oct 26th, 3:00 PM

Automatic Lung Cancer Detection Using Artificial Intelligence

Pristina, Kosovo

By far, lung cancer is the prominent cause of cancer deaths for both men and women around the world. In 2018, statistics for WHO (World Health Organization) showed that of 2.09 million people diagnosed with this disease, 1.76 million people have died. The survival rate increases if it is detected in its early stages. Taking into consideration the complexity of the problem, many computer-aided diagnosis systems that increase the survival rate have been proposed and developed. Driven by the notable success of deep learning in the area of complex image classification problems, this paper presents the use ResNet34, ResNet50, VGG16, and VGG19 convolutional neural network architectures or classifying images of patients with cancer. Moreover, to compare the performance evaluation Accuracy, Precision, Area Under Curve, and F1 score were calculated. In conclusion, ResNet50 architecture was estimated with the best result for this classification problem, with 87.04% Accuracy and 85.71% F1 score. The strategy of using pre-trained deep learning models proved to be pertinent to this problem.