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

Abstract

Background

Knee osteoarthritis (KOA) is a degenerative joint disease commonly assessed using X-ray images based on the Kellgren-Lawrence (KL) criteria. Although the KL standard exists, its ambiguity often causes patients to misunderstand their condition, leading to overtreatment or delayed treatment and challenges in guiding precise surgical decisions. Moreover, the data-driven technology has been impeded by low resolution and feature distribution inconsistency of knee X-ray images. The imbalances between positive and negative samples further degrade detection accuracy.

Objective

The objective of this study was to develop a deep learning-based model, namely Task-aligned Path Aggregation Feature Fusion For Knee Osteoarthritis Detection (TPAFFKnee), to improve KOA detection accuracy by addressing limitations in traditional methods. Its more accurate detection could help in terms of proper treatment for patients and precision in surgery by physicians.

Methods

We proposed the TPAFFKnee model based on the EfficientNetB4 network, which introduced a path aggregation network for better feature extraction and replaced Fully Convolutional Network (FCN) with task-aligned detection as the head. Additionally, the loss function was improved by replacing the original loss function with Efficient Intersection over Union Loss (EIoU Loss) to address the imbalance between positive and negative samples.

Results

The results showed that the model could accurately detect KOA categories and lesion locations based on the KL classification criteria, with a Mean Average Precision (mAP) of 93% on the Mendeley KOA dataset of 1650 knee osteoarthritis X-ray images from several hospitals. The mAP for the K2, K3, and K4 categories were 98.6%, 98.5%, and 99.6%, respectively. Compared with Faster R-CNN, SSD, RetinaNet, EfficientNetB4, and YOLOX, the proposed algorithm improved detection mAP by 14.3%, 12.4%, 15.3%, 22.7%, and 4.3%.

Conclusion

This study emphasizes the importance of the EfficientNetB4 network in KOA detection. The TPAFFKnee model provides an effective solution for improving the accuracy of KOA detection and offers a promising approach for standardized KL classification in medical applications. Future research can integrate more clinical data while improving the overall landscape of healthcare delivery through data-driven automation solutions.

© 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/10.2174/0115734056360714250612080450
2025-06-20
2025-10-29

  • From This Site

    /content/journals/cmir/10.2174/0115734056360714250612080450
    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-E15734056360714.html?itemId=/content/journals/cmir/10.2174/0115734056360714250612080450&mimeType=html&fmt=ahah

References

  1. MahmoudianA. LohmanderL.S. MobasheriA. EnglundM. LuytenF.P. Early-stage symptomatic osteoarthritis of the knee - Time for action.Nat. Rev. Rheumatol.2021171062163210.1038/s41584‑021‑00673‑434465902
     [Google Scholar]
  2. MononenM.E. LiukkonenM.K. TurunenM.J. X‐ray with finite element analysis is a viable alternative for MRI to predict knee osteoarthritis: Data from the Osteoarthritis Initiative.J. Orthop. Res.20244291964197310.1002/jor.2586138650428
     [Google Scholar]
  3. MohajerB. DolatshahiM. MoradiK. NajafzadehN. EngJ. ZikriaB. WanM. CaoX. RoemerF.W. GuermaziA. DemehriS. Role of thigh muscle changes in knee osteoarthritis outcomes: Osteoarthritis initiative data.Radiology2022305116917810.1148/radiol.21277135727152
     [Google Scholar]
  4. VanniniF. SpaldingT. AndrioloL. BerrutoM. DentiM. Espregueira-MendesJ. MenetreyJ. PerettiG.M. SeilR. FilardoG. Sport and early osteoarthritis: The role of sport in aetiology, progression and treatment of knee osteoarthritis.Knee Surg. Sports Traumatol. Arthrosc.20162461786179610.1007/s00167‑016‑4090‑527043343
     [Google Scholar]
  5. EdzieE.K.M. Dzefi-TetteyK. GorlekuP.N. IdunE.A. OseiB. CudjoeO. AsemahA.R. KusodziH. Application of information and communication technology in radiological practices: A cross-sectional study among radiologists in Ghana.J. Glob. Health Rep.20204e202004610.29392/001c.13060
     [Google Scholar]
  6. SivakumariT. VaniR. Deep learning-based automated knee joint localization in radiographic images using faster R-CNN.Curr. Med. Imaging2024201e06062423076810.2174/157340562001240606112211
     [Google Scholar]
  7. KaleR.D. KhandelwalS. A review on: deep learning and computer intelligent techniques using X-ray imaging for the early detection of knee osteoarthritis.International Conference on Machine Learning, Image Processing, Network Security and Data SciencesSpringer, Cham, 18 January 2023, pp 97–113.10.1007/978‑3‑031‑24352‑3_8
     [Google Scholar]
  8. AlyamiJ. Identification of severe grading in knee osteoarthritis from MRI using ensemble deep learning.Curr. Med. Imaging2024201e1573405628996310.2174/011573405628996324031804094538584509
     [Google Scholar]
  9. YeohP.S.Q. LaiK.W. GohS.L. HasikinK. HumY.C. TeeY.K. DhanalakshmiS. Emergence of deep learning in knee osteoarthritis diagnosis.Comput. Intell. Neurosci.202120211493143710.1155/2021/493143734804143
     [Google Scholar]
  10. TiulpinA. ThevenotJ. RahtuE. LehenkariP. SaarakkalaS. Automatic knee osteoarthritis diagnosis from plain radiographs: A deep learning- based approach.Sci. Rep.201881172710.1038/s41598‑018‑20132‑729379060
     [Google Scholar]
  11. WahyuningrumR.T. AnifahL. PurnamaI.K.E. PurnomoM.H. A new approach to classify knee osteoarthritis severity from radiographic images based on cnn-lstm method.2019 IEEE 10th International Conference on Awareness Science and Technology (iCAST)Morioka, Japan. 23-25 October 2019. pp. 1-6.10.1109/ICAwST.2019.8923284
     [Google Scholar]
  12. SchirattiJ.B. DuboisR. HerentP. CahanéD. DacharyJ. ClozelT. WainribG. Keime-GuibertF. LalandeA. PueyoM. GuillierR. GabarrocaC. MoingeonP. A deep learning method for predicting knee osteoarthritis radiographic progression from MRI.Arthritis Res. Ther.202123126210.1186/s13075‑021‑02634‑434663440
     [Google Scholar]
  13. AntonyJ. McguinnessK. O’ConnorN.E. MoranK. Quantifying radiographic knee osteoarthritis severity using deep convolutional neural networks.2016 23rd international conference on pattern recognition (ICPR)Cancun, Mexico. 04-08 December 2016. pp. 1195-1200.10.1109/ICPR.2016.7899799
     [Google Scholar]
  14. PedoiaV. NormanB. MehanyS.N. BucknorM.D. LinkT.M. MajumdarS. 3D convolutional neural networks for detection and severity staging of meniscus and PFJ cartilage morphological degenerative changes in osteoarthritis and anterior cruciate ligament subjects.J. Magn. Reson. Imaging201949240041010.1002/jmri.2624630306701
     [Google Scholar]
  15. NguyenH.H. SaarakkalaS. BlaschkoM.B. TiulpinA. Semixup: in-and out-of-manifold regularization for deep semi-supervised knee osteoarthritis severity grading from plain radiographs.IEEE Trans. Med. Imaging202039124346435610.1109/TMI.2020.301700732804644
     [Google Scholar]
  16. HuoJ. OuyangX. SiL. XuanK. WangS. YaoW. LiuY. XuJ. QianD. XueZ. WangQ. ShenD. ZhangL. Automatic grading assessments for knee mri cartilage defects via self-ensembling semi- supervised learning with dual-consistency.Med. Image Anal.20228010250810.1016/j.media.2022.10250835759870
     [Google Scholar]
  17. TariqT SuhailZ NawazZ Knee osteoarthritis detection and classification using x-rays.IEEE Access20231148294830310.1109/ACCESS.2023.3276810
     [Google Scholar]
  18. DharmaniB.C. KhatriK. Deep learning for knee osteoarthritis severity stage detection using x-ray images.2023 15th International Conference on COMmunication Systems & NETworks (COMSNETS)Bangalore, India. 03-08 January 2023, pp. 78-83.10.1109/COMSNETS56262.2023
     [Google Scholar]
  19. LiW. FengX.S. ZhaK. LiS. ZhuH.S. Summary of target detection algorithms.J. Phys. Conf. Ser.20211757101200310.1088/1742‑6596/1757/1/012003
     [Google Scholar]
  20. MainiR. AggarwalH. A comprehensive review of image enhancement techniques.Arxiv Preprint201010.48550/arXiv.1003.4053
     [Google Scholar]
  21. XuY. LiD. XieQ. WuQ. WangJ. Automatic defect detection and segmentation of tunnel surface using modified Mask R-CNN.Measurement202117810931610.1016/j.measurement.2021.109316
     [Google Scholar]
  22. MarquesG. AgarwalD. de la Torre DíezI. Automated medical diagnosis of COVID-19 through EfficientNet convolutional neural network.Appl. Soft Comput.20209610669110.1016/j.asoc.2020.10669133519327
     [Google Scholar]
  23. LuG. ZhangW. WangZ. Optimizing depthwise separable convolution operations on gpus.IEEE Trans. Parallel Distrib. Syst.2022331708710.1109/TPDS.2021.3084813
     [Google Scholar]
  24. HuJ. ShenL. SunG. Squeeze-and-excitation networks.IEEE Trans. Pattern Anal. Mach. Intell.20204282011202310.1109/TPAMI.2019.2913372
     [Google Scholar]
  25. AlamM.J. MohammadM.S. HossainM.A.F. ShowmikI.A. RaihanM.S. AhmedS. MahmudT.I. S 2 C-DeLeNet: A parameter transfer based segmentation-classification integration for detecting skin cancer lesions from dermoscopic images.Comput. Biol. Med.202215010614810.1016/j.compbiomed.2022.10614836252363
     [Google Scholar]
  26. LiZ. LangC. LiewJ.H. LiY. HouQ. FengJ. Cross- layer feature pyramid network for salient object detection.IEEE Trans. Image Process.2021304587459810.1109/TIP.2021.307281133872147
     [Google Scholar]
  27. ShaoZ. ZhangX. ZhangT. XuX. ZengT. Rbfa-net: A rotated balanced feature-aligned network for rotated sar ship detection and classification.Remote Sens.20221414334510.3390/rs14143345
     [Google Scholar]
  28. LuoH. WangP. ChenH. XuM. Object detection method based on shallow feature fusion and semantic information enhancement.IEEE Sens. J.20212119218392185110.1109/JSEN.2021.3103612
     [Google Scholar]
  29. AgarwalN. SondhiA. ChopraK. SinghG. Transfer learning: Survey and classification.Smart Innovations in Communication and Computational Sciences, Proceedings of IcsiccsSpringer, Singapore, 02 August 2020, pp 145–155.10.1007/978‑981‑15‑5345‑5_13
     [Google Scholar]
  30. WeissK. KhoshgoftaarT.M. WangD. A survey of transfer learning.J. Big Data201631910.1186/s40537‑016‑0043‑6
     [Google Scholar]
/content/journals/cmir/10.2174/0115734056360714250612080450
Loading
Effective Feature Extraction for Knee Osteoarthritis Detection on X-ray Images using Convolutional Neural Networks
Curr. Med. Imaging 21, E15734056360714 (2025); https://doi.org/10.2174/0115734056360714250612080450
/content/journals/cmir/10.2174/0115734056360714250612080450
/content/journals/cmir/10.2174/0115734056360714250612080450
Loading

Data & Media loading...


  • Article Type:     Research Article
Keyword(s): CNN; Deep learning; Knee osteoarthritis; Object detection; TPAFFKnee model; X-ray images

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