Detecting pneumonia from chest X-ray images is a critical and challenging task due to the diverse and sometimes subtle manifestations of lesions. This paper proposes a fusion model that combines a Mixture of Experts (MoE) architecture with Transfer Learning techniques to improve the accuracy and robustness of the diagnostic system. The model consists of local experts, which are trained specifically on the left and right lung regions after a segmentation step, and a global expert that processes the entire image. Each expert is built upon pre-trained Convolutional Neural Network (CNN) architectures and subsequently fine-tuned on X-ray data. Tests on the PneumoniaMNIST dataset reveal that the suggested model might greatly enhance accuracy and recall compared to baseline methods. This opens up new possibilities for automated medical diagnostic support systems.
Research Article | Open Access | Download Full Text
Volume 3 | Issue 1 | Year 2026 | Article Id: DSM-V3I1P102 DOI: https://doi.org/10.59232/DSM-V3I1P102
A Mixture of Experts (MoE) and Transfer Learning Combined Model for Pneumonia Detection in X-Ray Images
Hoang Trinh, Thao Nguyen, Hai Tran
| Received | Revised | Accepted | Published |
|---|---|---|---|
| 12 Dec 2025 | 08 Jan 2026 | 31 Jan 2026 | 07 Feb 2026 |
Citation
Hoang Trinh, Thao Nguyen, Hai Tran. “A Mixture of Experts (MoE) and Transfer Learning Combined Model for Pneumonia Detection in X-Ray Images.” DS Journal of Multidisciplinary, vol. 3, no. 1, pp. 17-24, 2026.
Abstract
Keywords
Pneumonia detection, Mixture of Experts (MoE), Transfer learning, Deep learning, Chest X-Ray Analysis.
References
[1] Jiancheng Yang et al., “MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification,” Scientific Data, vol. 10, no. 1, p. 1-10, 2023.
[Google Scholar] [Publisher Link]
[2] Muhammad Ayaz, Furqan Shaukat, and Gulistan Raja, “Ensemble Learning based Automatic Detection of Tuberculosis in Chest X-Ray Images using Hybrid Feature Descriptors,” Physical and Engineering Sciences in Medicine, vol. 44, no. 1, pp. 183-194, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Evans Kotei, and Ramkumar Thirunavukarasu, “Ensemble Technique Coupled with Deep Transfer Learning Framework for Automatic Detection of Tuberculosis from Chest X-Ray Radiographs,” Healthcare, vol. 10, no. 11, pp. 1-22, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Sourodip Ghosh et al., “Vision Transformers Excel in Chest X-Ray Analysis,” 2025 IEEE Conference on Artificial Intelligence (CAI), Santa Clara, CA, USA, pp. 495-500, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Enes Ayan, Bergen Karabulut, and Halil Murat Ünver, “Diagnosis of Pediatric Pneumonia with Ensemble of Deep Convolutional Neural Networks in Chest X-Ray Images,” Arabian Journal for Science and Engineering, vol. 47, no. 2, pp. 2123-2139, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Yufeng Jiang, and Yiqing Shen, “M4oE: A Foundation Model for Medical Multimodal Image Segmentation with Mixture of Experts,” International Conference on Medical Image Computing and Computer-Assisted Intervention,” pp. 621-631, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Ahmed Iqbal, Muhammad Usman, and Zohair Ahmed, “An Efficient Deep Learning-Based Framework for Tuberculosis Detection using Chest X-Ray Images,” Tuberculosis, vol. 136, 2022. [CrossRef]
[Google Scholar] [Publisher Link]
[8] Mehdhar S.A.M. Al-Gaashani, Fengjun Shang, and Ahmed A. Abd El-Latif, “Ensemble Learning of Lightweight Deep Convolutional Neural Networks for Crop Disease Image Detection,” Journal of Circuits, Systems and Computers, vol. 32, no. 5, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Xavier Alphonse Inbaraj et al., “A Novel Machine Learning Approach for Tuberculosis Segmentation and Prediction using Chest-X-Ray (CXR) Images,” Applied Sciences, vol. 11, no. 19, pp. 1-17, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Yılmaz Kaya et al., “A New Approach to COVID-19 Detection from X-Ray Images using Angle Transformation with GoogleNet and LSTM,” Measurement Science and Technology, vol. 33, no. 12, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Sheetal Rajpal et al., “Using Handpicked Features in Conjunction with Resnet-50 for Improved Detection of Covid-19 from Chest X-Ray Images,” Chaos, Solitons & Fractals, vol. 145, pp. 1-9, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Morteza Heidari et al., “Improving the Performance of CNN to Predict the Likelihood of COVID-19 using Chest X-Ray Images with Preprocessing Algorithms,” International Journal of Medical Informatics, vol. 144, pp. 1-9, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Arpita Vats et al., “The Evolution of Mixture of Experts: A Survey from Basics to Breakthroughs,” Prerpints, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Jiacheng Liu et al., “A Survey on Inference Optimization Techniques for Mixture of Experts Models,” arXiv Preprint, pp. 1-35, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Reyhan Achmad Rizal et al., “Analysis of Tuberculosis (TB) on X-Ray Image using SURF Feature Extraction and the K-Nearest Neighbor (KNN) Classification Method,” Jaict, vol. 5, no. 2, pp. 9-12, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Taewook Kim, “Factors Associated with Predicting Knee Pain using Knee X-Ray and Personal Factors: A Multivariate Logistic Regression and Xgboost Model Analysis from the Nationwide Korean Database (KNHANES),” PloS One, vol. 19, no. 12, pp. 1-16, 2024.
[CrossRef] [Google Scholar] [Publisher Link]