Related Experiment Video
Updated: Jun 5, 2025
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
1.7K
From Binary to Multi-Class Classification: A Two-Step Hybrid CNN-ViT Model for Chest Disease Classification Based on
Yousra Hadhoud1, Tahar Mekhaznia1, Akram Bennour1
1LAMIS Laboratory, Larbi Tebessi University, Tebessa 12002, Algeria.
Diagnostics (Basel, Switzerland)
|December 17, 2024
Summary
A new hybrid model combining Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) accurately detects Tuberculosis and distinguishes between Pneumonia types from chest X-rays. This Computer-Aided Diagnosis (CAD) system shows high accuracy, aiding in resource-limited settings.
Area of Science:
- Artificial Intelligence in Medical Imaging
- Deep Learning for Diagnostic Systems
- Radiographic Image Analysis
Background:
- Chest disease identification, particularly for Tuberculosis and Pneumonia, faces diagnostic challenges due to overlapping radiographic features.
- Limited availability of expert radiologists exacerbates diagnostic difficulties, especially in developing countries.
- A need exists for objective and consistent analysis of chest X-ray images to reduce human error in diagnosis.
Purpose of the Study:
- To develop a Computer-Aided Diagnosis (CAD) system for analyzing chest X-ray images.
- To accurately detect Tuberculosis and differentiate between Tuberculosis and Pneumonia using a hybrid AI model.
- To leverage the strengths of Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) for enhanced diagnostic performance.
Main Methods:
- A two-step hybrid model integrating ResNet-50 CNN with the ViT-b16 architecture was designed.
- Transfer learning was employed using datasets from Guangzhou Women's and Children's Medical Center (Pneumonia) and Qatar/Dhaka universities (Tuberculosis).
- The model combines CNNs' hierarchical feature extraction with ViTs' self-attention mechanisms for improved classification.
Main Results:
- The hybrid CNN-ViT model achieved 98.97% accuracy in binary classification for Tuberculosis detection.
- For multi-class classification (Tuberculosis, viral Pneumonia, bacterial Pneumonia), the model reached 96.18% accuracy.
- These results indicate significant potential for improving diagnostic accuracy and reliability in chest disease classification.
Conclusions:
- The proposed hybrid CNN-ViT model shows substantial potential for advancing CAD systems in chest disease diagnosis.
- Integrating CNN and ViT architectures enhances diagnostic precision, offering a robust solution for complex radiographic analyses.
- This approach can alleviate healthcare burdens in resource-limited settings and improve patient outcomes for chest diseases.
Related Concept Videos
01:20Imaging Studies for Cardiovascular System III: X-Ray
142
The most common cardiovascular diagnostic test is an X-ray. It produces images of the heart, blood vessels, and adjacent structures.
Definition and Purpose
An X-ray, or radiograph, is a non-invasive method that uses ionizing radiation to take images of internal structures. It is mainly used in cardiac imaging to examine the heart, lungs, and major blood vessels, aiming to identify abnormalities in the heart's size, shape, and position, such as heart failure, congenital defects, and vascular...
Definition and Purpose
An X-ray, or radiograph, is a non-invasive method that uses ionizing radiation to take images of internal structures. It is mainly used in cardiac imaging to examine the heart, lungs, and major blood vessels, aiming to identify abnormalities in the heart's size, shape, and position, such as heart failure, congenital defects, and vascular...
142
01:28Imaging Studies for Cardiovascular System V: CT
1
Cardiac computed tomography (CT) scanning is an advanced cardiac imaging technique that utilizes CT technology, with or without intravenous (IV) contrast, to produce accurate cross-sectional virtual slices of specific areas of the heart, coronary circulation, and major blood vessels such as the aorta, pulmonary veins, and arteries. The computer processes these slices to generate three-dimensional images. Multidetector CT (MDCT) is a rapid form of CT scanning that captures multiple slices...
1