Pneumonia Detection in Children from Chest X-Ray Images by Executing Network Surgery of Deep Neural Networks

Pneumonia is an infection that is caused by bacteria, viruses, or fungi. It is a form of acute respiratory infection that affects the lungs and is a leading cause of death in developing countries. Pneumonia leads to one in three deaths as per WHO. However, with early detection and medication, it can...

Full description

Saved in:
Bibliographic Details
Published in:SN computer science 2024-02, Vol.5 (2), p.215, Article 215
Main Authors: Dash, Amiya Kumar, Mohapatra, Puspanjali, Ray, Niranjan Kumar
Format: Article
Language:English
Subjects:
Accuracy
AI Perspective
Artificial intelligence
Artificial neural networks
Chest
Classification
Computer Imaging
Computer Science
Computer Systems Organization and Communication Networks
Data Structures and Information Theory
Datasets
Deep learning
Developing countries
Information Systems and Communication Service
Innovation in Smart Things: A Systems
LDCs
Machine learning
Neural networks
Pattern Recognition and Graphics
Pneumonia
Review Article
Security
Software Engineering/Programming and Operating Systems
Vision
X-rays
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Pneumonia is an infection that is caused by bacteria, viruses, or fungi. It is a form of acute respiratory infection that affects the lungs and is a leading cause of death in developing countries. Pneumonia leads to one in three deaths as per WHO. However, with early detection and medication, it can be cured. Chest X-ray is one of the most common methods to detect pneumonia. Several deep learning models have been proposed to detect pneumonia using Chest X-rays, but these models come with their own set of limitations in terms of accuracy and reliability.The challenge is to find the Deep Learning method which satisfies all the standard metrics of performance.In this paper, we compared six different deep learning models like VGG16, VGG19, ResNet50, ResNet152V2, MobileNetV2 and OpticNet by fine-tuning the model architecture. We replace the previously fully connected layers with newly initialized ones and subsequently training the current fully connected layers to predict our input classes. The results show that fine-tuning the pre-trained deep convolutional neural networks (CNN) proves to be effective in pneumonia detection compared to the state-of-the-art models with fewer parameters and less training data.
ISSN:2661-8907
2662-995X
2661-8907
DOI:10.1007/s42979-023-02512-7