Deep learning has shown outstanding efficiency in medical image segmentation. Segmentation of knee tissues is an important task for early diagnosis of rheumatoid arthritis (RA) with selecting variant features. Automated segmentation and feature extraction of knee tissues are desirable for faster and reliable analysis of large datasets and further diagnosis. In this paper a novel architecture called as Discrete-MultiResUNet, which is a combination of discrete wavelet transform (DWT) with MultiResUNet architecture is applied for feature extraction and segmentation, respectively. This hybrid architecture captures more prominent features from the knee magnetic resonance image efficiently with segmenting vital knee tissues. The hybrid model is evaluated on the knee MR dataset demonstrating outperforming performance compared with baseline models. The model achieves excellent segmentation performance accuracy of 96.77% with a dice coefficient of 98%.
BenhamouC.L., et al., Fractal Analysis of Radiographic Trabecular Bone Texture and Bone Mineral Density: Two Complementary Parameters Related to Osteoporotic Fractures, Journal of Bone and Mineral Research16(4) (2001), 697–704. doi: 10.1359/jbmr.2001.16.4.697.
2.
MacKayJ.W., et al., Association of subchondral bone texture on magnetic resonance imaging with radiographic knee osteoarthritis progression: data from the Osteoarthritis Initiative Bone Ancillary Study,”, European Radiology28(11) (2018), 4687–4695. doi: 10.1007/s00330-018-5444-9
3.
WiseB.L., et al., Bone shape mediates the relationship between sex and incident knee osteoarthritis, BMC Musculoskeletal Disorders19(1) (2018), 331–10.1186/s12891-018-2251-z
4.
ÖztürkC.N. and AlbayrakS., Automatic segmentation of cartilage in high-field magnetic resonance images of the knee joint with an improved voxel-classification-driven region-growing algorithm using vicinity-correlated subsampling, Computers in Biology and Medicine72 (2016), 90–107. doi: 10.1016/j.compbiomed.2016.03.011
5.
GraichenH., Eisenhart-RotheR.V., VoglT., EnglmeierK.H. and EcksteinF., Quantitative Assessment of Cartilage Status in Osteoarthritis by Quantitative Magnetic Resonance Imaging: Technical Validation for Use in Analysis of Cartilage Volume and Further Morphologic Parameters, Arthritis and Rheumatism50(3) (2004), 811–816. doi: 10.1002/art.20191
6.
GrässelS. and AszódiA., Cartilage volume 2: Pathophysiology, vol. 2. Cham: Springer International Publishing, (2017).
7.
BaldwinM.A., LangenderferJ.E., RullkoetterP.J. and LazP.J., Development of subject-specific and statistical shape models of the knee using an efficient segmentation and mesh-morphing approach, Computer Methods and Programs in Biomedicine97(3) (2010), 232–240. doi: 10.1016/j.cmpb.2009.07.005
8.
VincentG., WolstenholmeC., ScottI. and BowesM., Fully Automatic Segmentation of the Knee Joint using Active Appearance Models, Medical Image Analysis for the Clinic: A Grand Challenge (2010), 224–230.
9.
GajS., YangM., NakamuraK. and LiX., Automated cartilage and meniscus segmentation of knee MRI with conditional generative adversarial networks, Magnetic Resonance in Medicine84(1) (2020), 437–449. doi: 10.1002/mrm.28111
10.
AmbellanF., TackA., EhlkeM. and ZachowS., Automated segmentation of knee bone and cartilage combining statistical shape knowledge and convolutional neural networks: Data from the Osteoarthritis Initiative, Medical Image Analysis52 (2019), 109–118. doi: 10.1016/j.media.2018.11.009
11.
DenizC.M., et al., Segmentation of the Proximal Femur from MR Images using Deep Convolutional Neural Networks, Scientific Reports8(1) (2018), 1–14. doi: 10.1038/s41598-018-34817-6
12.
LanH.C., ChangT.R., LiaoW.C., ChungY.N. and ChungP.C., Knee MR image segmentation combining contextual constrained neural network and level set evolution, 2009 IEEE Symposium on Computational Intelligence in Bioinformatics and Computational Biology, CIBCB 2009 - Proceedings (2009), 271–277, doi:10.1109/CIBCB.2009.4925738
13.
LiuF., ZhouZ., JangH., SamsonovA., ZhaoG. and KijowskiR., Deep convolutional neural network and 3D deformable approach for tissue segmentation in musculoskeletal magnetic resonance imaging, Magnetic Resonance in Medicine79(4) (2018), 2379–2391. doi: 10.1002/mrm.26841
14.
LiuF., et al., Deep learning approach for evaluating knee MR images: Achieving high diagnostic performance for cartilage lesion detection, Radiology289(1) (2018), 160–169. doi: 10.1148/radiol.2018172986
15.
RinkyB.P., MondalP., ManikantanK. and RamachandranS., DWT based Feature Extraction using Edge Tracked Scale Normalization for Enhanced Face Recognition, Procedia Technology6 (2012), 344–353. doi: 10.1016/j.protcy.2012.10.041
16.
GokulalakshmiA., KarthikS., KarthikeyanN. and KavithaM.S., ICM-BTD: improved classification model for brain tumor diagnosis using discrete wavelet transform-based feature extraction and SVM classifier, Soft Computing24(24) (2020), 18599–18609. doi: 10.1007/s00500-020-05096-z
17.
GhazaliK.H., MansorM.F., MustafaM.M. and HussainA., Feature Extraction Technique using Discrete Wavelet Transform for Image Classification, no. January, (20080, doi:10.1109/SCORED.2007.4451366
18.
LaimecheL., FaridaH. and SmaineM., A new feature extraction scheme in wavelet transform for stego image classification, Evolving Systems0(0) (2017), 0. doi: 10.1007/s12530-017-9174-z
19.
KesslerD.A., et al., The optimisation of deep neural networks for segmenting multiple knee joint tissues from MRIs, Computerized Medical Imaging and Graphics86 (2020), 101793. doi: 10.1016/j.compmedimag.2020.101793
20.
LiD., ChenC., LiJ. and WangL., Dense gate network for biomedical image segmentation, International Journal of Computer Assisted Radiology and Surgery15(8) (2020), 1247–1255. doi: 10.1007/s11548-020-02138-7
21.
JhaD., RieglerM.A., JohansenD., HalvorsenP. and JohansenH.D., DoubleU-Net: A deep convolutional neural network for medical image segmentation, Proceedings - IEEE Symposium on Computer-Based Medical Systems2020 (2020), 558–564. doi: 10.1109/CBMS49503.2020.00111
22.
MoreS., SinglaJ., AbugabahA. and AlzubiA.A., Machine Learning Techniques for Quantification of Knee Segmentation from MRI, Complexity2020 (2020), doi: 10.1155/2020/6613191
23.
RamamurthyM., KrishnamurthiI., VimalS. and RobinsonY.H., Deep learning based genome analysis and NGS-RNA LL identification with a novel hybrid model, Bio Systems197 (2020), 104211. doi: 10.1016/j.biosystems.2020.104211
24.
PatelF.K. and SinghM., Segmentation of cartilage from knee MRI images using the watershed algorithm, International Journal of Advance Research, Ideas and Innovations in Technology4(2) (2018), 1727–1730.
25.
ZhouZ., ZhaoG., KijowskiR. and LiuF., Deep convolutional neural network for segmentation of knee joint anatomy, Magnetic Resonance in Medicine80(6) (2018), 2759–2770. doi: 10.1002/mrm.27229
26.
ZhaoX., ZhangT., LiuH., ZhuG. and ZouX., Automatic Windowing for MRI with Convolutional Neural Network, IEEE Access7 (2019), 68594–68606. doi: 10.1109/ACCESS.2019.2918814
27.
ZhangK., LuW. and MarzilianoP., Automatic knee cartilage segmentation from multi-contrast MR images using support vector machine classification with spatial dependencies, Magnetic Resonance Imaging31(10) (2013), 1731–1743. doi: 10.1016/j.mri.2013.06.005
28.
PathakD., KrahenbuhlP., DonahueJ., DarrellT. and EfrosA.A., Context Encoders: Feature Learning by Inpainting, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2016), 2536–2544. doi:10.1109/CVPR.2016.278
29.
WengW. and ZhuX., INet: Convolutional Networks for Biomedical Image Segmentation,”, IEEE Access9 (2021), 16591–16603. doi: 10.1109/ACCESS.2021.3053408
30.
MoreS., et al., Security Assured CNN-Based Model for Reconstruction of Medical Images on the Internet of Healthcare Things, IEEE Access8 (2020), 126333–126346. doi: 10.1109/ACCESS.2020.3006346
31.
AmbellanF., TackA., EhlkeM. and ZachowS., Automated segmentation of knee bone and cartilage combining statistical shape knowledge and convolutional neural networks: Data from the Osteoarthritis Initiative, Medical Image Analysis52 (2019), 109–118. doi: 10.1016/j.media.2018.11.009
32.
QamarS., JinH., ZhengR., AhmadP. and UsamaM., A variant form of 3D-UNet for infant brain segmentation, Future Generation Computer Systems108 (2020), 613–623. doi: 10.1016/j.future.2019.11.021
