Related Experiment Video
Updated: Jun 28, 2025
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
14.2K
Detection of various lung diseases including COVID-19 using extreme learning machine algorithm based on the features
Md Nahiduzzaman1, Md Omaer Faruq Goni1, Md Robiul Islam1
1Department of Electrical & Computer Engineering, Rajshahi University of Engineering & Technology, Rajshahi 6204, Bangladesh.
Biocybernetics and Biomedical Engineering
|April 15, 2024
Summary
This study introduces an AI system using machine learning to detect seven lung diseases from X-rays. The CNN-PCC-ELM model accurately identifies diseases like COVID-19, aiding medical diagnosis.
Area of Science:
- Medical Imaging and Diagnostics
- Artificial Intelligence in Healthcare
- Computational Pathology
Background:
- Lung diseases including pneumonia, tuberculosis (TB), and COVID-19 pose significant global health threats, leading to severe illness and mortality.
- Accurate and timely diagnosis of lung diseases is critical, especially during the COVID-19 pandemic, to enable effective treatment.
- Existing diagnostic methods can be time-consuming and may require specialized expertise, highlighting the need for advanced automated systems.
Purpose of the Study:
- To develop and evaluate an intelligent recognition system for the early detection of seven common lung diseases using machine learning techniques.
- To enhance the diagnostic capabilities available to medical experts by providing an automated tool for analyzing chest X-ray (CXR) images.
- To improve the accuracy and efficiency of lung disease classification, particularly for differentiating COVID-19 from other respiratory conditions.
Main Methods:
- A lightweight Convolutional Neural Network (CNN) was employed to extract salient features from chest X-ray (CXR) images.
- The Pearson Correlation Coefficient (PCC) was utilized to identify the optimal subset of extracted features.
- An Extreme Learning Machine (ELM) classifier was implemented for the multi-class classification of lung diseases, ensuring computational efficiency.
Main Results:
- The proposed CNN-PCC-ELM model achieved a high accuracy of 96.22% and an Area Under Curve (AUC) of 99.48% for eight-class classification.
- The model demonstrated superior performance compared to state-of-the-art (SOTA) methods in detecting COVID-19, pneumonia, and tuberculosis in both binary and multiclass settings.
- For COVID-19 detection in an eight-class scenario, the model achieved 100% precision, 99% recall, 100% F1-score, and 99.99% ROC, underscoring its robustness.
Conclusions:
- The developed CNN-PCC-ELM model offers a robust and accurate solution for the automated detection and differentiation of multiple lung diseases from CXR images.
- This intelligent system has the potential to significantly assist medical physicians in making faster and more precise diagnoses, thereby improving patient treatment outcomes.
- The model's high performance, especially in identifying COVID-19, suggests its value as a supplementary tool in managing respiratory disease outbreaks and routine diagnostics.
