Scholarlink

Background <p>Bacterial diseases include any illness caused by bacteria. Bacteria are tiny microorganisms that can only be seen with a microscope. Other types of microorganisms include viruses, fungi, and some parasites. Millions of bacteria normally live on the skin, in the intestines, and on the genitalia. The vast majority of bacteria do not cause disease, and many bacteria are helpful and necessary for good health. Bacteria are single-celled microorganisms that can cause many diseases and are found everywhere in the environment.</p> Aim <p>This paper aims in developing an automated method for the prediction of the bacterial image i.e. Pneumonia, TB employing a chest X-ray scan using ensemble learning, Convolution Neural Network (CNN), and machine learning algorithms.</p> Objective <p>In this paper, the authors introduce a set of novel methods to detect different bacterial diseases using Chest X-ray images of TB, and Pneumonia. Pneumonia is an infectious respiratory illness that affects the lungs. It can be caused by various infectious agents, including bacteria, viruses, fungi, and parasites. Pneumonia occurs when these pathogens cause inflammation in the air sacs in one or both lungs, leading to symptoms such as cough, fever, difficulty breathing, chest pain, and fatigue.</p> Methods <p>In order to diagnose bacterial disease using a chest X-ray scan, this study presents a GUI-based method known as Detection of Bacterial Disease using a Convolution Neural Network (DBDCN). Later, machine learning, ensemble learning, and additional datasets are used to validate the model.</p> Results <p>The accuracy of the model was assessed using various pre-trained models. When the VGG-19 model is trained using the proposed DBDCN technique, accuracy is 95.58% considering 80% of the data is set aside for training and the remaining 20% is set aside for testing. 96.28% accuracy was attained using ensemble learning, while 90% accuracy was achieved using neural networks.</p> Conclusion <p>A GUI-based system of DBDCN was developed to detect Pneumonia and TB considering the VGG19 model. This GUI model can help doctors as a second opinion tool.</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An ensemble learning approach for detecting bacterial infections using chest X-ray imaging

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月17日

查看全文
  • Himanshu Jindal,
  • Monika Bharti,
  • Shruti Jain

摘要

Background

Bacterial diseases include any illness caused by bacteria. Bacteria are tiny microorganisms that can only be seen with a microscope. Other types of microorganisms include viruses, fungi, and some parasites. Millions of bacteria normally live on the skin, in the intestines, and on the genitalia. The vast majority of bacteria do not cause disease, and many bacteria are helpful and necessary for good health. Bacteria are single-celled microorganisms that can cause many diseases and are found everywhere in the environment.

Aim

This paper aims in developing an automated method for the prediction of the bacterial image i.e. Pneumonia, TB employing a chest X-ray scan using ensemble learning, Convolution Neural Network (CNN), and machine learning algorithms.

Objective

In this paper, the authors introduce a set of novel methods to detect different bacterial diseases using Chest X-ray images of TB, and Pneumonia. Pneumonia is an infectious respiratory illness that affects the lungs. It can be caused by various infectious agents, including bacteria, viruses, fungi, and parasites. Pneumonia occurs when these pathogens cause inflammation in the air sacs in one or both lungs, leading to symptoms such as cough, fever, difficulty breathing, chest pain, and fatigue.

Methods

In order to diagnose bacterial disease using a chest X-ray scan, this study presents a GUI-based method known as Detection of Bacterial Disease using a Convolution Neural Network (DBDCN). Later, machine learning, ensemble learning, and additional datasets are used to validate the model.

Results

The accuracy of the model was assessed using various pre-trained models. When the VGG-19 model is trained using the proposed DBDCN technique, accuracy is 95.58% considering 80% of the data is set aside for training and the remaining 20% is set aside for testing. 96.28% accuracy was attained using ensemble learning, while 90% accuracy was achieved using neural networks.

Conclusion

A GUI-based system of DBDCN was developed to detect Pneumonia and TB considering the VGG19 model. This GUI model can help doctors as a second opinion tool.

Graphical Abstract