Related Experiment Video
Updated: Aug 16, 2025
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
Stacked ensemble learning based on deep convolutional neural networks for pediatric pneumonia diagnosis using chest
J Arun Prakash1, Vinayakumar Ravi2, V Sowmya1
1Center for Computational Engineering and Networking (CEN), Amrita School of Engineering, Amrita Vishwa Vidyapeetham, Coimbatore, India.
Insights
A new deep learning model accurately detects pediatric pneumonia from chest X-rays, improving diagnosis speed and cost-efficiency. This AI tool aids physicians in identifying pneumonia in children, potentially saving lives.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Pediatrics
Background:
- Pneumonia is a leading cause of child mortality, particularly in developing nations.
- Current diagnostic methods like chest X-rays are time-consuming and challenging in pediatric cases due to low radiation levels.
- There is a critical need for rapid, accurate, and cost-effective diagnostic tools for pediatric pneumonia.
Purpose of the Study:
- To develop and evaluate a novel computer-aided detection (CAD) model for pediatric pneumonia classification.
- To leverage deep learning and ensemble methods for enhanced diagnostic accuracy and efficiency.
- To provide a reliable tool for assisting radiologists and physicians in diagnosing pediatric pneumonia.
Main Methods:
- A stacked ensemble learning approach was employed, utilizing deep learning features extracted from a fine-tuned Xception model.
- Kernel Principal Component Analysis (KPCA) was used for dimensionality reduction of the extracted features.
- A two-stage stacking classifier, including Random Forest, KNN, Logistic Regression, XGBoost, SVC, Nu-SVC, and MLP, was trained and validated using Stratified K-fold cross-validation.
Main Results:
- The proposed model achieved high performance metrics on a public pediatric pneumonia dataset.
- Key performance indicators included 98.3% accuracy, 99.29% precision, 98.36% recall, 98.83% F1-score, and 98.24% AUC score.
- The results demonstrate the model's potential for reliable real-time deployment.
Conclusions:
- The developed stacked ensemble deep learning model offers a highly accurate and efficient solution for pediatric pneumonia detection.
- This AI-powered tool can significantly assist healthcare professionals in diagnosing pneumonia, potentially reducing mortality rates.
- The model's performance suggests its suitability for integration into clinical workflows for improved pediatric care.
Abstract:
Pneumonia is an acute respiratory infection caused by bacteria, viruses, or fungi and has become very common in children ranging from 1 to 5 years of age. Common symptoms of pneumonia include difficulty breathing due to inflamed or pus and fluid-filled alveoli. The United Nations Children's Fund reports nearly 800,000 deaths in children due to pneumonia. Delayed diagnosis and overpriced tests are the prime reason for the high mortality rate, especially in underdeveloped countries. A time and cost-efficient diagnosis tool: Chest X-rays, was thus accepted as the standard diagnostic test for pediatric pneumonia. However, the lower radiation levels for diagnosis in children make the task much more onerous and time-consuming. The mentioned challenges initiate the need for a computer-aided detection model that is instantaneous and accurate. Our work proposes a stacked ensemble learning of deep learning-based features for pediatric pneumonia classification. The extracted features from the global average pooling layer of the fine-tuned Xception model pretrained on ImageNet weights are sent to the Kernel Principal Component Analysis for dimensionality reduction. The dimensionally reduced features are further trained and validated on the stacking classifier. The stacking classifier consists of two stages; the first stage uses the Random-Forest classifier, K-Nearest Neighbors, Logistic Regression, XGB classifier, Support Vector Classifier (SVC), Nu-SVC, and MLP classifier. The second stage operates on Logistic Regression using the first stage predictions for the final classification with Stratified K-fold cross-validation to prevent overfitting. The model was tested on the publicly available pediatric pneumonia dataset, achieving an accuracy of 98.3%, precision of 99.29%, recall of 98.36%, F1-score of 98.83%, and an AUC score of 98.24%. The performance shows its reliability for real-time deployment in assisting radiologists and physicians.
