| 40 | 0 | 12 |
| 下载次数 | 被引频次 | 阅读次数 |
肺炎是一种高致死率的疾病,针对肺炎治疗与早期筛查工具的研究受到了广泛关注。然而,胸部影像的复杂性、标注的困难性,以及肺炎类间相似性、类内差异性均为研究带来了挑战。首先,采用基于孪生网络的自监督学习方法提取胸部X光影像的自监督视觉特征表示,并通过多层感知机(multilayer perceptron, MLP)检测头完成肺炎的分类任务。其次,在YOLOv10的骨干网络中引入空间通道注意力机制,利用通道内的空间信息增强肺炎病灶特征,从而提升YOLOv10网络对肺炎病灶的识别能力。最后,在COVID-19 Radiography Database和RSNA Pneumonia Detection数据集上对所提出的方法进行了评估,实验结果表明所提出的算法在基于X光影像的肺炎分类与定位任务中的有效性。
Abstract：Pneumonia is a disease characterized by high mortality, and research on its treatment and early screening tools has garnered significant attention. However, the complexity of chest images, annotation difficulty, and the inter-class similarities and intra-class variations of pneumonia all pose challenges to pneumonia classification and localization tasks based on X-ray images. Firstly, a self-supervised learning approach utilizing a siamese network was applied to extract self-supervised visual feature representation of chest X-ray images, and a multi-layer perceptron(MLP) detection head was used to perform pneumonia classification task. Secondly, a spatial-channel attention module was designed in the backbone network of YOLOv10. By utilizing the spatial information across channels to enhance pneumonia features, the ability of the YOLOv10 network to identify pneumonia lesions was enhanced. Finally, the proposed algorithm was evaluated using the COVID-19 Radiography Database and RSNA Pneumonia Detection dataset. The results demonstrated its effectiveness of the proposed algorithm in pneumonia classification and localization tasks based on X-ray images.
[1]WU L H, ZHANG J, WANG Y L, et al. Pneumonia detection based on RSNA dataset and anchor-free deep learning detector[J]. Scientific reports, 2024, 14:1929.
[2]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.
[3]RASHID N, HOSSAIN M A F, ALI M, et al. AutoCovNet:unsupervised feature learning using autoencoder and feature merging for detection of COVID-19 from chest Xray images[J]. Biocybernetics and biomedical engineering, 2021, 41(4):1685-1701.
[4]CHENXL,HEKM.Exploringsimplesiameserepre-/sentationlearning[C]∥2021IEEE CVFConferenceon ComputerVisionandPatternRecognition.Piscataway:IEEEPress,2021:15745-15753.
[5]WANG C Y, MARK LIAO H Y, WU Y H, et al. CSPNet:a new backbone that can enhance learning capability of CNN[C]∥2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Piscataway:IEEE Press, 2020:1571-1580.
[6]HOU Q B, ZHOU D Q, FENG J S. Coordinate attention for efficient mobile network design[C]∥2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Piscataway:IEEE Press, 2021:13708-13717.
[7]SIRAZITDINOV I, KHOLIAVCHENKO M, MUSTAFAEV T, et al. Deep neural network ensemble for pneumonia localization from a large-scale chest X-ray database[J]. Computers&electrical engineering, 2019, 78:388-399.
[8]VARGHESE R, M S. YOLOv8:a novel object detection algorithm with enhanced performance and robustness[C]∥2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems. Piscataway:IEEE Press, 2024:1-6.
[9]ZHOUXY,WANGDQ,KR?HENBüHLP.Objectsas/points[EB OL].(2019-04-19)[2025-01-20]. https:∥//arxiv. org pdf1904. 07850.
[10]YASEENM.WhatisYOLOv9:anin-depthexploration oftheinternalfeaturesofthenext-generationobjectdetec-/tor[EB OL].(2024-09-12)[2025-01-20]. https:∥arx-//iv. org pdf2409. 07813.
[11]KHANAMR,HUSSAINM.YOLOv11:anoverviewof/thekeyarchitecturalenhancements[EB OL].(2024-10-//23)[2025-01-20].https:∥arxiv. org pdf2410. 17725.
[12]TIANYJ,YEQX,DOERMANND.YOLOv12:atten-/tion-centricreal-timeobjectdetectors[EB OL].(2025-//01-18)[2025-01-25]. https:∥arxiv. org pdf2502. 12524.
[13]WANGA,CHENH,LIULH,etal.YOLOv10:real-/timeend-to-endobjectdetection[EB OL].(2024-05-23)//[2025-01-25].https:∥arxiv. org pdf2405. 14458.
基本信息:
DOI：10.13705/j.issn.1671-6841.2025016
中图分类号:R563.1;TP18;TP391.41
引用信息:
[1]温亚雪,李玉琴,蒋振刚,等.基于X光影像的肺炎分类与定位方法研究[J].郑州大学学报(理学版)().DOI:10.13705/j.issn.1671-6841.2025016.
基金信息:
吉林省教育厅科学技术研究项目(JJKH20240948KJ)
2026-04-24
2026-04-24
2026-04-24