Related Experiment Video
Updated: Oct 13, 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
14.4K
A new deep learning pipeline to detect Covid-19 on chest X-ray images using local binary pattern, dual tree complex
1Ministry of Health of Republic of Turkey, Ankara, Turkey.
Applied Intelligence (Dordrecht, Netherlands)
|November 12, 2021
Summary
This study developed deep learning models using convolutional neural networks (CNNs) for early Covid-19 detection from X-ray images. The models achieved high accuracy, demonstrating AI
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computer-Aided Diagnosis
Background:
- Early and accurate diagnosis of Covid-19 is crucial for effective patient management and disease control.
- Chest X-ray imaging is a widely accessible tool for respiratory illness assessment.
- Deep learning offers advanced capabilities for image analysis and pattern recognition.
Purpose of the Study:
- To develop and evaluate deep learning models for the early detection of Covid-19 using chest X-ray images.
- To compare the performance of different convolutional neural network (CNN) architectures and image preprocessing techniques.
- To assess the efficacy of artificial intelligence in classifying Covid-19 positive and negative X-ray images.
Main Methods:
- Utilized deep learning, specifically convolutional neural networks (CNNs), for image classification.
- Developed and trained 23-layer and 54-layer CNN architectures.
- Employed image preprocessing techniques including Dual Tree Complex Wavelet Transform (DT-CWT) and Local Binary Pattern (LBP).
- Applied k-fold cross-validation (k=23 and k=2) for robust model evaluation.
- Integrated results from multiple algorithms using novel pipeline approaches.
Main Results:
- Achieved high performance metrics across two distinct datasets and a combined dataset.
- Sensitivity, specificity, accuracy, F-1 score, and AUC values consistently exceeded 0.97 in most configurations.
- The combined dataset (556 images) yielded an accuracy of 0.9906 and perfect specificity (1.0000).
- DT-CWT and LBP preprocessing enhanced classification performance compared to direct image analysis.
Conclusions:
- Deep learning models, particularly CNNs, are highly effective for automated Covid-19 detection from chest X-rays.
- The developed AI models demonstrate significant potential for supporting early diagnosis and clinical decision-making.
- Image preprocessing techniques can further improve the diagnostic accuracy of AI-based systems for Covid-19.
- The study highlights the promise of AI in augmenting radiological assessments for infectious diseases.
