Related Experiment Video
Updated: Jun 18, 2026
14:08
Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
Published on: April 13, 2013
43.4K
Hybrid deep learning and machine learning framework for automated pneumonia detection in chest X-ray images
Akshay S1, Sneha Kashyap1, Danikk Patel H N1
1Department of Computer Science, School of Computing, Amrita Vishwa Vidyapeetham, Mysuru, Karnataka, India.
Methodsx
|December 17, 2025
Summary
This study introduces a hybrid deep learning approach using VGG16 and ResNet for accurate pneumonia detection from chest X-rays. The method enhances diagnostic interpretability with Random Forest and Support Vector Machine classifiers.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computer Science
Background:
- Pneumonia poses significant health risks, necessitating accurate and timely diagnosis.
- Distinguishing between pneumonic and normal chest X-rays is crucial for effective treatment.
Purpose of the Study:
- To develop a hybrid deep learning system for precise pneumonia detection from chest X-rays.
- To enhance diagnostic accuracy and interpretability compared to single-model approaches.
Main Methods:
- Image preprocessing: histogram equalization, contrast normalization, grayscale conversion.
- Hybrid feature extraction using VGG16 and ResNet, followed by feature fusion.
- Dimensionality reduction with Principal Component Analysis (PCA) retaining 95% variance.
- Classification using Support Vector Machine (SVM) and Random Forest (RF) classifiers.
Main Results:
- The dual-CNN feature fusion approach demonstrated effective pneumonia identification.
- PCA-optimized feature dimensions improved model efficiency and accuracy.
- SVM and RF classifiers provided more interpretable diagnostic outcomes.
Conclusions:
- The proposed hybrid deep learning model offers a robust method for pneumonia detection.
- Combining VGG16 and ResNet features with PCA and interpretable classifiers enhances diagnostic capabilities.
