Home

Network Biology, 2020, 10(3): 62-76
[XML] [EndNote] [RefManager] [BibTex] [ Full PDF (1870K)] [Comment/Review Article]

Article

Fusion of Deep Convolutional Neural Network with PCA and Logistic Regression for diagnosis of pediatric pneumonia on chest X-Rays

Nahida Habib1,2, Md. Mahmodul Hasan1, Mohammad Motiur Rahman1
1Department of Computer Science and Engineering (CSE), Mawlana Bhashani Science and Technology University (MBSTU), Santosh, Tangail-1902, Bangladesh
2Department of Computer Science and Engineering (CSE), Ranada Prasad Shaha University (RPSU), Narayanganj-1400, Bangladesh

Received 4 May 2020;Accepted 31 May 2020;Published 1 September 2020
IAEES

Abstract
Consistent headway in machine learning technology is gradually substantiating its significance in many areas of medical research. Pneumonia is a disease caused due to acute respiratory infection affecting one or both lungs. Diagnosis and treatment of pneumonia at early stage can increase the survivability of suffering patients. Computer Aided Diagnosis (CAD) techniques are bridging up the gap of medical science and computer science by successfully diagnosing diseases such as tumor, cancer, pneumonia etc. This paper proposes a fusion of Deep Convolutional Neural Network Model with Principal Component Analysis (PCA) feature extraction model and Logistic Regression (LR) classifiers for the diagnosis of pneumonia from chest X-ray images. In this study, fine-tuned pre-trained CheXNet model is used as Convolutional Neural Network (CNN) model on standard pneumonia dataset collected from Guangzhou Women and Children's Medical Center, Guangzhou. The proposed model is capable of detecting pneumonia with an accuracy which outperforms the existing methods from 0.8% to 21.9% approx. Comparison with existing models and methods reveal that the proposed model delivers superior results than others according to precision, f1-score, accuracy and AUC values. This research can be a great subsidiary for radiologists or medical researchers for diagnosis of pediatric pneumonia from chest X-ray images.

Keywords AUC;CheXNet;computer aided diagnosis;Convolutional Neural Network;Logistic Regression;PCA;pneumonia;ROC.



International Academy of Ecology and Environmental Sciences. E-mail: office@iaees.org
Copyright © 2009-2024 International Academy of Ecology and Environmental Sciences. All rights reserved.
Web administrator: office@iaees.org, website@iaees.org; Last modified: 2024/3/29


Translate page to: