MLFF-GNN: A Multi-Level Feature-Fusion Graph Neural Network for CXR-Image-Based Pneumonia Classification

Ruichao Tian

doi:10.62051/ijcsit.v5n3.12

Authors

Ruichao Tian

DOI:

https://doi.org/10.62051/ijcsit.v5n3.12

Keywords:

Multi-level feature-fusion graph neural network, LSTM-enhanced graph network block, Graph neural network, Transfer learning, Synergistic information

Abstract

Pneumonia does great harm to children's health. Although Graph Neural Network (GNN) has made great progress in the classification of pneumonia images, they usually use the permutation-invariant message-passing mechanism to capture the information between nodes, which is difficult to capture the interaction information between nodes. In addition, the quality of chest X-ray (CXR) images of pneumonia was uneven, and the classification model was affected by the redundant information in the pneumonia images, which could not effectively capture key information to classification. To address these problems, we proposed the multi-level feature-fusion graph neural network (MLFF-GNN) and a novel graph neural network block named the LSTM-enhanced graph network (LEGN) block. The LEGN block is designed to capture interaction information among nodes, including synergistic LEGN information within the interactions. Simultaneously, we introduce an information theory framework to explain the capture of interaction between nodes, including synergistic information within these interactions. In addition, we proposed a novel multi-level feature extraction (MLFE) module that utilizes a Convolutional Neural Network (CNN) pre-training model to extract multi-level feature maps from pneumonia CXR images and employs transfer learning to capture key features. On the two-category pneumonia classification dataset, MLFF-GNN surpasses 10 state-of-the-art (SOTA) models, with a notable 1.35% increase in accuracy. It achieves a sensitivity, precision, and accuracy of 99.30%, 99.30%, and 98.97%, respectively. On the two-category pneumonia classification dataset, MLFF-GNN maintains a high-performance standard with sensitivity, precision, and accuracy rates of 97.78%, demonstrating superior performance compared to existing models. The experimental outcomes demonstrate that the proposed model can extract the key features of CXR images of pneumonia, and can effectively capture the interaction information between nodes, which is superior to the existing methods. The proposed MLFF-GNN significantly aids clinicians by providing robust analytical support for the diagnosis of pneumonia from CXR images.

Downloads

Download data is not yet available.

References

[1] Ahmed M S, Rahman A, AlGhamdi F, et al. Joint Diagnosis of Pneumonia, COVID-19, and Tuberculosis from Chest X-ray Images: A Deep Learning Approach [J]. Diagnostics, 2023, 13(15): 2562. doi: 10.3390/diagnostics13152562.

[2] Meng Y, Bridge J, Addison C, et al. Bilateral adaptive graph convolutional network on CT based Covid-19 diagnosis with uncertainty-aware consensus-assisted multiple instance learning [J]. Medical Image Analysis, 2023, 84: 102722. doi: 10.1016/j.media.2022.102722.

[3] Jie C, Jiming C, Ying S, et al. A pyramid GNN model for CXR-based COVID-19 classification [J]. The Journal of Supercomputing, 2024, 80(4): 5490-5508. doi: 10.1007/s11227-023-05633-1.

[4] Zhang W, Kong L, Lee S, et al. Detecting mental and physical disorders using multi-task learning equipped with knowledge graph attention network [J]. Artificial Intelligence in Medicine, 2024, 149: 102812. doi: 10.1016/j.artmed.2024.102812.

[5] Cui H, Ju F, Li J. Joint Multi-view Feature Network for Automatic Diagnosis of Pneumonia with CT Images [C]//International Conference on Frontier Computing. Springer, 2023: 169-180. doi: 10.1007/978-981-99-9342-0_19.

[6] Wang H, Huang G, Zhao Z, et al. CCF-GNN: A Unified Model Aggregating Appearance, Microenvironment, and Topology for Pathology Image Classification [J]. IEEE Transactions on Medical Imaging, 2023, 42(11): 3179-3193. doi: 10.1109/TMI.2023.3249343.

[7] Gilmer J, Schoenholz S S, Riley P F, et al. Neural Message Passing for Quantum Chemistry [C]//Precup D, Teh Y W. Proceedings of the 34th International Conference on Machine Learning: Vol. 70. PMLR, 2017: 1263-1272. https://proceedings.mlr.press/v70/gilmer17a.html.

[8] Shen W, Wei Z, Ren Q, et al. Rotation-Equivariant Quaternion Neural Networks for 3D Point Cloud Processing [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024: 1-17. doi: 10.1109/TPAMI.2023.3346383.

[9] Luo T, Liu Y, Pan S J. Collaborative Sequential Recommendations via Multi-View GNN-Transformers [J]. ACM Trans. Inf. Syst., 2024. https://doi.org/10.1145/3649436. doi: 10.1145/3649436.

[10] Chatzianastasis M, Lutzeyer J, Dasoulas G, et al. Graph Ordering Attention Networks [J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2023, 37(6): 7006-7014. doi: 10.1609/aaai.v37i6.25856.

[11] Bai Y, Mi Y, Su Y, et al. A Scalable Graph-Based Framework for Multi-Organ Histology Image Classification [J]. IEEE Journal of Biomedical and Health Informatics, 2022, 26(11): 5506-5517. doi: 10.1109/JBHI.2022.3199110.

[12] He K, Zhang X, Ren S, et al. Deep Residual Learning for Image Recognition [C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA: IEEE, 2016: 770-778. http://ieeexplore.ieee.org/document/7780459/. doi: 10.1109/CVPR.2016.90.

[13] Li Z, Zheng Y, Shan D, et al. ScribFormer: Transformer Makes CNN Work Better for Scribble-based Medical Image Segmentation [J]. IEEE Transactions on Medical Imaging, 2024: 1-1. doi: 10.1109/TMI.2024.3363190.

[14] Pati P, Jaume G, Fernandes L A, et al. HACT-Net: A Hierarchical Cell-to-Tissue Graph Neural Network for Histopathological Image Classification [C]//Sudre C H, Fehri H, Arbel T, et al. Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, and Graphs in Biomedical Image Analysis. Cham: Springer International Publishing, 2020: 208-219. doi: 10.1007/ 978-3-030-60365-6_20.

[15] Lei Z, Wang Y, Li Z, et al. Attention based multilayer feature fusion convolutional neural network for unsupervised monocular depth estimation [J]. Neurocomputing, 2021, 423: 343-352. doi: 10.1016/j.neucom.2020.11.002.

[16] Wang L, He K, Feng X, et al. Multilayer feature fusion with parallel convolutional block for fine-grained image classification [J]. Applied Intelligence, 2022, 52(3): 2872-2883. doi: 10.1007/s10489-021-02573-2.