[1]Lü Jia,GONG Xiao,HE Ling.Classification model for pediatric pneumonia built by integrating spatial and multi-axis frequency domain analysis[J].CAAI Transactions on Intelligent Systems,2026,21(4):963-978.[doi:10.11992/tis.202510034]
Copy

Classification model for pediatric pneumonia built by integrating spatial and multi-axis frequency domain analysis

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 21 Number of periods: 2026 4 Page number: 963-978 Column: 学术论文—机器感知与模式识别 Public date: 2026-07-05
Title:
Classification model for pediatric pneumonia built by integrating spatial and multi-axis frequency domain analysis
Author(s):
Lü Jia1; 2;  GONG Xiao1;  HE Ling3
1. College of Computer and Information Sciences, Chongqing Normal University, Chongqing 401331, China;
2. National Center for Applied Mathematics in Chongqing, Chongqing Normal University, Chongqing 401331, China;
3. Ministry of Education Key Laboratory of Child Development and Disorders, National Clinical Research Center for Child Health and Disorders, Children’s Hospital of Chongqing Medical University, Chongqing 400014, China
Keywords:
pediatric pneumonia; image classification; deep learning; X-Ray; spatial-frequency domain collaboration; mutli-axis frequency domain analysis; multi-scale; dynamic gating
CLC:
TP391.4
DOI:
10.11992/tis.202510034
Abstract:
Pediatric pneumonia poses a serious threat to children’s health worldwide, and its precise classification is crucial for effective treatment. Chest X-ray images used for pneumonia classification exhibit subtle pathological textures and large-scale diffuse lesions, making it difficult for a single spatial domain model to efficiently and effectively represent both characteristics. To address this issue, this paper proposes a classification model that fuses spatial domain and multi-axis frequency domain analysis. In the first two stages of the model, a shift-aware multi-scale unit is constructed to capture fine-grained pathological textures. In the last two stages, a multi-axis Fourier unit is constructed to model long-range dependencies in the frequency domain and characterize large-scale lesion features. Additionally, a multi-kernel convolutional gated linear unit is developed to enhance feature representation at each stage. Experimental results showed that on the ChestXRay2017 dataset, the model achieved improvements of 2.72% and 2.75% in accuracy and F1-score, respectively, compared to the state-of-the-art model. It also achieved competitive performance on the VinDr-PCXR and Curated X-Ray datasets, validating the model’s effectiveness and generalization. Thus, the proposed model provides a new solution for the auxiliary diagnosis of pediatric pneumonia.
References:
[1] ZHENG Jingyuan, LIU Dingming, YANG Zhifu, et al. SEACC: self-evolving and adaptive contrastive learning for classification of pediatric pneumonia and anteroposterior/posteroanterior chest radiographs[J]. Displays, 2024, 81: 102607
[2] 吕佳, 于明楷 , 陈欣, 等. 差异感知引导的边界Transformer儿童肺炎CT图像分割网络[J]. 光学学报, 2025, 45(15): 112-136 L? Jia, YU Mingkai, CHEN Xin, et al. Difference aware guided boundary Transformer network for childhoodPneumonia CT image segmentation[J]. Acta optica sinca, 2025, 45(15): 112-136
[3] KAYA M. Feature fusion-based ensemble CNN learning optimization for automated detection of pediatric pneumonia[J]. Biomedical signal processing and control, 2024, 87: 105472
[4] 贠恺, 贾荣浩, 魏国辉, 等. 基于CNN与Transformer混合模型的肺炎辅助诊断[J]. 计算机系统应用, 2025, 34(2): 216-224 YUN Kai, JIA Ronghao, WEI Guowei, et al. Pneumonia assisted diagnosis based on hybrid model of CNN and Transformer[J]. Computer systems and applications, 2025, 34(2): 216-224
[5] RICKARD D, KABIR M A, HOMAIRA N. Machine learning-based approaches for distinguishing viral and bacterial pneumonia in paediatrics: a scoping review[J]. Computer methods and programs in biomedicine, 2025, 268: 108802
[6] 程文娟, 于国庆. 基于XDense-RC-NET的CXR图像分类算法[J]. 计算机应用研究, 2022, 39(12): 3803-3807 CHENG Wenjun, YU Guoqing. Improved CXR image classification algorithm based on XDense-RC-NET[J]. Application research of computers, 2022, 39(12): 3803-3807
[7] KHAN W, ZAKI N, ALI L. Intelligent pneumonia identification from chest X-rays: a systematic literature review[J]. IEEE access, 2021, 9: 51747-51771
[8] LITJENS G, KOOI T, BEJNORDI B E, et al. A survey on deep learning in medical image analysis[J]. Medical image analysis, 2017, 42: 60-88
[9] RAJPURKAR P, IRVIN J, ZHU K, et al. Chexnet: radiologist-level pneumonia detection on chest X-rays with deep learning[EB/OL]. (2017-12-25)[2025-08-21]. https://arxiv.org/abs/1711.05225.
[10] 潘伟, 马志庆, 魏国辉, 等. 卷积神经网络在肺炎图像分类中的研究进展与展望[J]. 中国医疗设备, 2025, 40(7): 147-154 PAN Wei, MA Zhiqing, WEI Guohui, et al. Research progress and prospect of convolutional neural networks in pneumoniaimage classification[J]. China medical devices, 2025, 40(7): 147-154
[11] HAMMOUDI K, BENHABILES H, MELKEMI M, et al. Deep learning on chest X-ray images to detect and evaluate pneumonia cases at the era of COVID-19[J]. Journal of medical systems, 2021, 45(7): 75
[12] DU Jie, GUAN Kai, ZHOU Yanhong, et al. Parameter-free similarity-aware attention module for medical image classification and segmentation[J]. IEEE transactions on emerging topics in computational intelligence, 2023, 7(3): 845-857
[13] 周涛, 叶鑫宇, 陆惠玲, 等. 基于多分辨率注意密集网络的肺炎分类识别方法[J]. 模式识别与人工智能, 2023, 36(6): 544-555 ZHOU Tao, YE Xinyu, LU Huiling, et al. Pneumonia classification and recognition method based on multi-resolution attention dense network[J]. Pattern recognition and artificial intelligence, 2023, 36(6): 544-555
[14] SUN Junding, XUE Jianxiang, XU Zhaozhao, et al. SAGCN: self-adaptive graph convolutional network for pneumonia detection[J]. Biomedical signal processing and control, 2025, 106: 107634
[15] DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16x16 words: transformers forimage recognition at scale[C]//International Conference on Learning Representations. Online: ICLR Foundation, 2021: 1-22.
[16] OKOLO G I, KATSIGIANNIS S, RAMZAN N. IEViT: an enhanced vision transformer architecture for chest X-ray image classification[J]. Computer methods and programs in biomedicine, 2022, 226: 107141
[17] UKWUOMA C C, QIN Zhiguang, BELAL BIN HEYAT M, et al. A hybrid explainable ensemble transformer encoder for pneumonia identification from chest X-ray images[J]. Journal of advanced research, 2023, 48: 191-211
[18] MANZARI O N, AHMADABADI H, KASHIANI H, et al. MedViT: a robust vision transformer for generalized medical image classification[J]. Computers in biology and medicine, 2023, 157: 106791
[19] LASKER A, GHOSH M, SK M O, et al. LungConVT-Net: a visual transformer network with blended features for Pneumonia detection[J]. Pattern recognition, 2026, 171: 112150
[20] 王梦溪, 雷涛, 姜由涛, 等. 基于空频协同的CNN-Transformer多器官分割网络[J]. 智能系统学报, 2025, 20(5): 1266-1280 WANG Mengxi, LEI Tao, JIANG Youtao, et al. CNN-transformer multiorgan segmentation network based on space-frequency collaboration[J]. CAAI transactions on intelligent systems, 2025, 20(5): 1266-1280
[21] 葛斌, 卞文强, 夏晨星, 等. 基于空间频域交互感知的医学图像分割[J/OL]. 光电子·激光(2025-09-26) [2025-11-26]. https://link.cnki.net/urlid/12.1182.O4.20250926.1622.022. GE Bin, BIAN Wenqiang, XIA Chenxing, et al. Medical image segmentation based on spatial frequency domain interactive perception[J/OL]. Journal of optoelectronics·laser(2025-09-26) [2025-11-26]. https://link.cnki.net/urlid/12.1182.O4.20250926.1622.022.
[22] RUAN Jiacheng, GAO Jingsheng, XIE Mingye, et al. Learning multi-axis representation in frequency domain for medical image segmentation[J]. Machine learning, 2025, 114(1): 10
[23] WANG Wenxuan, WANG Jing, CHEN Chen, et al. FreMIM: fourier transform meets masked image modeling for medical image segmentation[C]//2024 IEEE/CVF Winter Conference on Applications of Computer Vision. Waikoloa: IEEE, 2024.
[24] GAO Tian, ZHANG Yu, ZHANG Zhiyuan, et al. BHViT: binarized hybrid vision transformer[C]//2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville: IEEE, 2025.
[25] SZEGEDY C, VANHOUCKE V, IOFFE S, et al. Rethinking the inception architecture for computer vision[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016.
[26] SHI Dai. TransNeXt: robust foveal visual perception for vision transformers[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2024.
[27] KERMANY D S, GOLDBAUM M, CAI Wenjia, et al. Identifying medical diagnoses and treatable diseases by image-based deep learning[J]. Cell, 2018, 172(5): 1122-1131. e9.
[28] PHAM H H, NGUYEN N H, TRAN T T, et al. PediCXR: an open, large-scale chest radiograph dataset for interpretation of common thoracic diseases in children[J]. Scientific data, 2023, 10: 240
[29] SAIT U, LAL K, PRAJAPATI S, et al. Curated dataset for covid-19 posterior-anterior chest radiography images(X-rays)[EB/OL]. (2020-09-15)[2025-08-21]. https://data.mendeley.com/datasets/9xkhgts2s6/1.
[30] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016.
[31] HUANG Gao, LIU Zhuang, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017.
[32] TAN M, LE Q. Efficientnetv2: smaller models and faster training[C]//Proceedings of the 38th International Conference on Machine Learning. Online: IMLS, 2021.
[33] HOWARD A, SANDLER M, CHEN Bo, et al. Searching for MobileNetV3[C]//2019 IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019.
[34] KUMAR V, GUPTA A, HAZARIKA B B, et al. Automatic diagnosis of COVID-19 from chest X-ray images using transfer learning-based deep features and machine learning models[J]. China communications, 2025, 22(7): 274-289
[35] KILI?ARSLAN S, K?ZKURT C, BA? S, et al. Detection and classification of pneumonia using novel superior exponential(SupEx) activation function in convolutional neural networks[J]. Expert systems with applications, 2023, 217: 119503
[36] LIU Zhuang, MAO Hanzi, WU Chaoyuan, et al. A ConvNet for the 2020s[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022.
[37] WANG Ao, CHEN Hui, LIN Zijia, et al. LSNet: see large, focus small[C]//2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville: IEEE, 2025.
[38] OUYANG Haibin, LIU Dongmei, LI S, et al. Two-stage deep feature selection method using voting differential evolution algorithm for pneumonia detection from chest X-ray images[J]. IEEE transactions on emerging topics in computational intelligence, 2025, 9(1): 918-932
[39] LIU Ze, LIN Yutong, CAO Yue, et al. Swin transformer: hierarchical vision transformer using shifted windows[C]//2021 IEEE/CVF International Conference on Computer Vision. Montreal: IEEE, 2021.
[40] MEHTA S, RASTEGARI M. MobileViT: light-weight, general-purpose, and mobile-friendly uivision transformer[EB/OL]. (2021-10-05)[2025-10-17]. https://arxiv.org/abs/2110.02178.
[41] YU Weihao, LUO Mi, ZHOU Pan, et al. MetaFormer is actually what you need for vision[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. New Orleans: IEEE, 2022.
[42] LOU Meng, ZHANG Shu, ZHOU Hongyu, et al. TransXNet: learning both global and local dynamics with a dual dynamic token mixer for visual recognition[J]. IEEE transactions on neural networks and learning systems, 2025, 36(6): 11534-11547
Similar References:

Memo

-

Last Update: 1900-01-01

Copyright © CAAI Transactions on Intelligent Systems