Related Experiment Video
Updated: Aug 20, 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
14.3K
Automated Lung-Related Pneumonia and COVID-19 Detection Based on Novel Feature Extraction Framework and Vision
Chiagoziem C Ukwuoma1, Zhiguang Qin1, Md Belal Bin Heyat2,3,4
1School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu 610054, China.
Bioengineering (Basel, Switzerland)
|November 24, 2022
Summary
This study introduces a novel deep learning framework for enhanced lung disease classification from chest X-rays, achieving high accuracy. The model effectively addresses image quality issues, improving diagnostic reliability for machine learning models.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computer Vision
Background:
- Image quality issues (blur, low contrast) reduce machine learning model accuracy in image recognition.
- Chest X-rays are crucial for medical diagnostics but challenging to interpret.
- Accurate lung disease classification using machine learning on chest X-rays is essential.
Purpose of the Study:
- To develop a robust deep learning model for high-accuracy lung disease classification from chest X-ray images.
- To investigate an advanced framework that enhances feature extraction and analysis of chest X-ray data.
Main Methods:
- An ensemble technique was used to derive richer features from chest X-ray images.
- Global second-order pooling was applied to extract higher-level global features.
- A vision transformer approach analyzed image patches after applying position embedding.
Main Results:
- The model achieved 98.00% accuracy, 96.01% sensitivity, and 96.20% precision on the COVID-19 Radiography Dataset.
- On the Covid-ChestX-ray-15k dataset, the model obtained 97.84% accuracy, 96.76% sensitivity, and 96.80% precision.
- Experimental results demonstrate superior performance compared to traditional deep learning models and existing state-of-the-art methods.
Conclusions:
- The proposed deep learning framework significantly improves the accuracy of lung disease classification from chest X-rays.
- The novel approach effectively handles image complexities, offering a reliable tool for medical diagnostics.
- This method shows promise for advancing AI-driven medical image analysis in identifying lung pathologies.
