Related Experiment Video
Updated: May 4, 2026
10:26
Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
Published on: May 19, 2023
2.9K
Lung cancer detection and classification using optimized CNN features and Squeeze-Inception-ResNeXt model.
1Department of Computer Science and Engineering, Kalasalingam Academy of Research and Education, Krishnankoil, Srivilliputhur, Tamil Nadu, India.
Computational Biology and Chemistry
|March 30, 2025
Summary
A new deep learning model, Squeeze-Inception-ResNeXt, accurately classifies lung diseases from CT scans. This advanced system improves early lung cancer detection, aiding radiologists and potentially reducing mortality rates.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Lung cancer is a leading global cause of mortality, necessitating improved early detection and diagnostic methods.
- Computer-Aided Diagnostic (CAD) systems assist radiologists in identifying lung nodules and malignancies, mitigating human error.
- Accurate classification of lung cancer subtypes is critical for effective treatment strategies and patient outcomes.
Purpose of the Study:
- To develop and evaluate a deep learning approach for classifying lung diseases using chest Computed Tomography (CT) images.
- To enhance the accuracy and efficiency of lung disease diagnosis through advanced image processing and classification techniques.
- To introduce a novel deep learning architecture, Squeeze-Inception-ResNeXt, for improved lung cancer subtype classification.
Main Methods:
- Image pre-processing techniques including color space conversion, data augmentation, resizing, and normalization were applied to CT scans.
- Feature extraction was performed using a Convolutional Neural Network (CNN) optimized with the Slime Mould Algorithm (SMA).
- A hybrid classification model, Squeeze-Inception-ResNeXt, combining Squeeze-Inception V3 and ResNeXt, was developed and trained using SMA.
Main Results:
- The Squeeze-Inception-ResNeXt model achieved high performance in classifying lung diseases, including Adenocarcinoma, Large Cell Carcinoma, and Squamous Cell Carcinoma.
- The proposed model demonstrated superior accuracy (97.7%), sensitivity (98.1%), and specificity (97.4%) compared to traditional models.
- The Squeeze-Inception-ResNeXt model offers reduced computational cost while maintaining high diagnostic performance.
Conclusions:
- The developed deep learning approach, Squeeze-Inception-ResNeXt, shows significant promise for accurate and efficient lung disease classification from CT scans.
- This method has the potential to aid radiologists in early lung cancer detection and diagnosis, contributing to improved patient management.
- The integration of SMA for optimization further enhances the model's effectiveness in clinical applications.