Related Experiment Video
Updated: Jul 25, 2025
07:53
Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer
Published on: October 13, 2023
1.5K
Detection of Lung Opacity and Treatment Planning with Three-Channel Fusion CNN Model
1Department of Computer Engineering, Çankırı Karatekin University, 18100 Çankırı, Turkey.
Arabian Journal for Science and Engineering
|June 26, 2023
Summary
This study introduces a deep learning model for detecting and classifying lung opacities. The novel three-channel CNN approach achieves high accuracy, aiding physicians in diagnosis and patient monitoring.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computer Science
Background:
- Lung opacities require careful monitoring due to potential irreversible patient consequences.
- Accurate diagnosis and classification of lung opacity are crucial for effective patient management.
- Deep learning offers promising solutions for lung opacity detection, classification, and segmentation.
Purpose of the Study:
- To develop and evaluate a deep learning model for detecting and classifying lung opacity.
- To assess the effectiveness of a three-channel fusion Convolutional Neural Network (CNN) model.
- To provide physicians with a tool that offers cost and time advantages in diagnosing lung opacity.
Main Methods:
- A balanced dataset was compiled from public sources for training and validation.
- A three-channel fusion CNN model was designed, integrating MobileNetV2, InceptionV3, and VGG19 architectures.
- ResNet architecture was utilized for feature transfer between layers.
Main Results:
- The proposed model achieved high accuracy rates on a newly compiled dataset for lung opacity classification.
- Accuracy for two, three, four, and five classes were 92.52%, 92.44%, 87.12%, and 91.71%, respectively.
- The model demonstrated significant cost and time advantages for physicians.
Conclusions:
- The three-channel fusion CNN model is effective for detecting and classifying lung opacity.
- This deep learning approach can assist physicians in improving diagnostic accuracy and efficiency.
- The model shows potential for widespread clinical application in lung opacity monitoring.
