Skip to content
2000
Volume 21, Issue 1
  • ISSN: 1573-4056
  • E-ISSN: 1875-6603
file format pdf download PDF
file format text download EPUB
side by side viewer icon HTML

Abstract

Background

Early and timely detection of pulmonary nodules and initiation treatment can substantially improve the survival rate of lung carcinoma. However, current detection methods based on convolutional neural networks (CNNs) cannot easily detect pulmonary nodules owing to low detection accuracy and the difficulty in detecting small-sized pulmonary nodules; meanwhile, more accurate CNN-based models are slow and require high hardware specifications.

Objective

The aim of this study is to develop a detection model that achieves both high accuracy and real-time performance, ensuring effective and timely results.

Methods

In this study, based on YOLOv5s, a concentrated-comprehensive convolution (C3_ODC) module with multidimensional attention is designed in the convolutional layer of the original backbone network for enhancing the feature-extraction capabilities of the model. Moreover, lightweight convolution is combined with weighted bidirectional feature pyramid networks (BiFPNs) to form a GS-BiFPN structure that enhances the fusion of multiscale features while reducing the number of model parameters. Finally, Focal Loss is combined with the normalized Wasserstein distance (NWD) to optimize the loss function. Focal loss focuses on carcinoma-positive samples to mitigate class imbalance, whereas the NWD enhances the detection performance of small lung nodules.

Results

In comparison experiments against the YOLOv5s, the proposed model improved the average precision by 8.7% and reduced the number of parameters and floating-point operations by 5.4% and 8.2%, respectively, while achieving 116.7 frames per second.

Conclusion

The proposed model balances high detection accuracy against real-time requirements.

© 2025 The Author(s). Published by Bentham Science Publishers.
This is an open access article published under CC BY 4.0 https://creativecommons.org/licenses/by/4.0/legalcode
Loading

Article metrics loading...

/content/journals/cmir/10.2174/0115734056310722241210055412
2025-01-02
2025-11-02

  • From This Site

    /content/journals/cmir/10.2174/0115734056310722241210055412
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

/deliver/fulltext/cmir/21/1/CMIR-21-E15734056310722.html?itemId=/content/journals/cmir/10.2174/0115734056310722241210055412&mimeType=html&fmt=ahah

References

  1. SungH. FerlayJ. SiegelR.L. LaversanneM. SoerjomataramI. JemalA. BrayF. Global cancer statistics 2020: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries.CA Cancer J. Clin.202171320924910.3322/caac.2166033538338
     [Google Scholar]
  2. ZhangW. WangX. LiX. ChenJ. 3D skeletonization feature based computer-aided detection system for pulmonary nodules in CT datasets.Comput. Biol. Med.201892647210.1016/j.compbiomed.2017.11.00829154123
     [Google Scholar]
  3. KayaA. CanA.B. A weighted rule based method for predicting malignancy of pulmonary nodules by nodule characteristics.J. Biomed. Inform.201556697910.1016/j.jbi.2015.05.01126008877
     [Google Scholar]
  4. ChenH. XiongW. WuJ. ZhuangQ. YuG. Decision-making model based on ensemble method in auxiliary medical system for non-small cell lung cancer.IEEE Access2020817190317191110.1109/ACCESS.2020.3024840
     [Google Scholar]
  5. WuJ. TanY. ChenZ. ZhaoM. Decision based on big data research for non-small cell lung cancer in medical artificial system in developing country.Comput. Methods Prog. Biomed.20181598710110.1016/j.cmpb.2018.03.00429650322
     [Google Scholar]
  6. RenS. HeK. GirshickR. SunJ. Faster R-CNN: Towards real-time object detection with region proposal networks.IEEE Trans. Pattern Anal. Mach. Intell.20173961137114910.1109/TPAMI.2016.257703127295650
     [Google Scholar]
  7. JiangP. ErguD. LiuF. CaiY. MaB. A review of Yolo algorithm developments.Procedia Comput. Sci.20221991066107310.1016/j.procs.2022.01.135
     [Google Scholar]
  8. LiuW. AnguelovD. ErhanD. SzegedyC. ReedS. FuC.Y. BergA.C. SSD: Single Shot MultiBox DetectorComputer Vision–ECCV 2016. Lecture Notes in Computer ScienceChamSpringer20162137
     [Google Scholar]
  9. JiZ. WuY. ZengX. AnY. ZhaoL. WangZ. GanchevI. Lung nodule detection in medical images based on improved Yolov5s.IEEE Access202311763717638710.1109/ACCESS.2023.3296530
     [Google Scholar]
  10. HongwuQ. XinyuW. FeiX. Pulmonary nodule detection algorithm based on lightweight improved YOLOv8 network model.Bull. Pacific State Uni.20243743546
     [Google Scholar]
  11. ShariatyF. DavydovV.V. YushkovaV.V. GlinushkinA.P. RudV.Y. Automated pulmonary nodule detection system in computed tomography images based on Active-contour and SVM classification algorithm.J. Phys. Conf. Ser.20191410101207510.1088/1742‑6596/1410/1/012075
     [Google Scholar]
  12. PaingM.P. HamamotoK. TungjitkusolmunS. VisitsattapongseS. PintaviroojC. Automatic detection of pulmonary nodules using three-dimensional chain coding and optimized random forest.Appl. Sci.2020107234610.3390/app10072346
     [Google Scholar]
  13. HeK. GkioxariG. DollárP. GirshickR. Mask r-cnn.arXiv:1703.06870201710.48550/arXiv.1703.06870
     [Google Scholar]
  14. NguyenC.C. TranG.S. NguyenV.T. BurieJ.C. NghiemT.P. Pulmonary nodule detection based on faster R-CNN with adaptive anchor box.IEEE Access2021915474015475110.1109/ACCESS.2021.3128942
     [Google Scholar]
  15. LiaoJ. LiW. DuL.Y. Detection of pulmonary nodule lesions based on improved Mask R-CNN algorithm.International Conference on Artificial Intelligence and Intelligent Information Processing (AIIIP 2022)30 Nov, 2022, Qingdao, China, pp. 1245609.10.1117/12.2659647
     [Google Scholar]
  16. HanL. LiF. YuH. XiaK. XinQ. ZouX. BiRPN-YOLOvX: A weighted bidirectional recursive feature pyramid algorithm for lung nodule detection.J. XRay Sci. Technol.202331230131710.3233/XST‑22131036617767
     [Google Scholar]
  17. MaJ. LiX. LiH. MenzeB.H. LiangS. ZhangR. ZhengW.S. group-attention single-shot detector (GA-SSD): Finding pulmonary nodules in large-scale CT images.Proceed. Mach. Learn. Res.2019102358369
     [Google Scholar]
  18. RedmonJ. DivvalaS. GirshickR. FarhadiA. You only look once: Unified, real-time object detection.Proceedings of the IEEE conference on computer vision and pattern recognition201677978810.1109/CVPR.2016.91
     [Google Scholar]
  19. YangL.I. ShiqiG.A.O. Lung nodule detection system based on data augmentation and attention mechanism.J. Beijing Uni. Posts. Telecommun.202245425
     [Google Scholar]
  20. LiuY. AoY. Deformable attention mechanism-based YOLOv7 structure for lung nodule detection.J. Supercomput.20248017254502546910.1007/s11227‑024‑06381‑6
     [Google Scholar]
  21. MeiS. JiangH.Q. MaL. YOLO-lung: A practical detector based on imporved YOLOv4 for pulmonary nodule detection.2021 14th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)202110.1109/CISP‑BMEI53629.2021.9624373
     [Google Scholar]
  22. GuoN. BaiZ. Multi-scale pulmonary nodule detection by fusion of cascade R-CNN and FPN.2021 International Conference on Computer Communication and Artificial Intelligence (CCAI)2021151910.1109/CCAI50917.2021.9447531
     [Google Scholar]
  23. YangB. BenderG. LeQ.V. NgiamJ. Condconv: Conditionally parameterized convolutions for efficient inference.Proceedings of the 33rd International Conference on Neural Information Processing Systems201913071318
     [Google Scholar]
  24. ChenY. DaiX. LiuM. ChenD. YuanL. LiuZ. Dynamic convolution: Attention over convolution kernels.Proceedings of the IEEE/CVF conference on computer vision and pattern recognition.2020Available from:https://ieeexplore.ieee.org/document/9157588 (accessed on 18-11-2024).
     [Google Scholar]
  25. LiC. ZhouA. YaoA. Omni-dimensional dynamic convolutio.arXiv:2209.07947202210.48550/arXiv.2209.07947
     [Google Scholar]
  26. LinT.Y. DollárP. GirshickR. HeK. HariharanB. BelongieS. Feature pyramid networks for object detection.Proceedings of the IEEE conference on computer vision and pattern recognition201721172125
     [Google Scholar]
  27. LiuS. QiL. QinH. ShiJ. JiaJ. Path aggregation network for instance segmentation.Proceedings of the IEEE conference on computer vision and pattern recognition201887598768
     [Google Scholar]
  28. TanM. PangR. LeQ.V. Efficient Det: Scalable and efficient object detection.ArXiv:1911.09070, 2020.202010.48550/arXiv.1911.09070
     [Google Scholar]
  29. LiH. LiJ. WeiH. LiuZ. ZhanZ. RenQ. Slim-neck by GSConv: A better design paradigm of detector architectures for autonomous vehicles.2022Available from:https://arxiv.org/vc/arxiv/papers/2206/ 2206.02424v1.pdf (accessed on 18-11-2024).
     [Google Scholar]
  30. ZhengZ. WangP. LiuW. LiJ. YeR. RenD. Distance-IOU loss: Faster and better learning for bounding box regression.Proc. Conf. AAAI Artif. Intell.2020347129931300010.1609/aaai.v34i07.6999
     [Google Scholar]
  31. ZhangY.F. RenW. ZhangZ. JiaZ. WangL. TanT. Focal and efficient IOU loss for accurate bounding box regression.Neurocomputing202250614615710.1016/j.neucom.2022.07.042
     [Google Scholar]
  32. WangJ. XuC. YangW. YuL. A normalized Gaussian Wasserstein distance for tiny object detection.arXiv:2110.13389202110.48550/arXiv.2110.13389
     [Google Scholar]
/content/journals/cmir/10.2174/0115734056310722241210055412
Loading
Lightweight Lung-nodule Detection Model Combined with Multidimensional Attention Convolution
Curr. Med. Imaging 21, E15734056310722 (2025); https://doi.org/10.2174/0115734056310722241210055412
/content/journals/cmir/10.2174/0115734056310722241210055412
/content/journals/cmir/10.2174/0115734056310722241210055412
Loading

Data & Media loading...


  • Article Type:     Research Article
Keyword(s): Convolutional neural networks (CNNs); Lightweight; Lung-nodule detection; Omni-dimensional dynamic convolution; Wasserstein distance; YOLOv5s

Most Cited Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error
Please enter a valid_number test