Segmentation and severity classification of scar tissues in LGE-CMR images using HDResC-Net with Flamingo gannet search optimization.

B Abinaya, M Malleswaran
Author Information
  1. B Abinaya: Department of ECE, Easwari Engineering College, Ramapuram, Chennai, Tamilnadu 603201 India.
  2. M Malleswaran: Department of ECE, University College of Engineering Kancheepuram, Ponnerikkarai, Tamilnadu 631552 India.

Abstract

Late gadolinium enhanced-cardiac magnetic resonance (LGE-CMR) images play a critical role in evaluating cardiac pathology, where scar tissue serves as a vital indicator impacting prognosis and treatment decisions. However, accurately segmenting scar tissues and assessing their severity present challenges due to complex tissue composition and imaging artifacts. Existing methods often lack precision and robustness, limiting their clinical applicability. This work proposes a novel methodology that integrates the optimal segmentation algorithm (OSA) for segmentation and Flamingo Gannet search optimization-enabled hybrid deep residual convolutional network (FGSO-HDResC-Net) for severity classification of scar tissues in LGE-CMR images. Initially, the input image is pre-processed by using the adaptive Gabor Kuwahara filter. Then, the approach combines myocardium segmentation via region-based convolutional neural network and scar segmentation using OSA. Subsequently, FGSO-HDResC-Net integrates feature extraction and classification while optimizing hyperparameters through Flamingo Gannet search optimization. The feature extraction stage introduces two sets of techniques: localization features with texture analysis and spatial/temporal features using a deep residual network, complemented by feature fusion using the fractional concept. These features are inputted into a customized 1D convolutional neural network model for severity classification. Through comprehensive evaluation, the effectiveness of FGSO-HDResC-Net in accurately classifying scar tissue severity is demonstrated, offering improved disease assessment and treatment planning for cardiac patients. Moreover, the proposed FGSO-HDResC-Net model demonstrated superior performance, achieving an accuracy of 96.45%, a true positive rate of 95.42%, a true negative rate of 96.48%, a positive predictive value of 94.20%, and a negative predictive value of 94.18%. The accuracy of the devised model is 14.50%, 12.99%, 10.74%, 9.75%, 12.79%, and 11.26% improved than the traditional models.

Keywords

References

  1. BMC Res Notes. 2023 Aug 24;16(1):185 [PMID: 37620937]
  2. Curr Cardiol Rep. 2021 Jul 1;23(8):100 [PMID: 34196815]
  3. IEEE Trans Med Imaging. 2014 Jan;33(1):159-72 [PMID: 24107924]
  4. Artif Intell Med. 2015 Jul;64(3):205-15 [PMID: 26239472]
  5. Diagnostics (Basel). 2021 Jun 18;11(6): [PMID: 34207123]
  6. Heart Vessels. 2023 Jul;38(7):984-991 [PMID: 36786857]
  7. Biol Chem. 2020 Oct 02;402(8):911-923 [PMID: 33006947]
  8. IEEE Trans Med Imaging. 2016 May;35(5):1240-1251 [PMID: 26960222]
  9. Front Cardiovasc Med. 2021 Oct 28;8:724271 [PMID: 34778395]
  10. J Cardiol. 2020 Dec;76(6):601-609 [PMID: 32675026]
  11. Proc Inst Mech Eng H. 2020 Sep;234(9):1029-1035 [PMID: 32650699]
  12. Med Phys. 2020 Apr;47(4):1645-1655 [PMID: 31955415]
  13. Heliyon. 2024 May 23;10(11):e31629 [PMID: 38845929]
  14. Compr Physiol. 2015 Sep 20;5(4):1877-909 [PMID: 26426470]
  15. Circulation. 2006 Jun 13;113(23):2733-43 [PMID: 16754804]
  16. Proc IEEE Int Symp Biomed Imaging. 2023 Apr;2023: [PMID: 38523738]
  17. Neural Comput Appl. 2022;34(22):19751-19790 [PMID: 36060097]
  18. Med Image Anal. 2022 Apr;77:102362 [PMID: 35091277]
  19. Curr Med Sci. 2021 Apr;41(2):398-404 [PMID: 33877559]
  20. Med Image Anal. 2020 Feb;60:101595 [PMID: 31811981]
  21. Med Image Anal. 2020 May;62:101668 [PMID: 32276185]

Word Cloud

Created with Highcharts 10.0.0scarseveritysegmentationFGSO-HDResC-NetclassificationusingLGE-CMRnetworkimagestissuetissuesFlamingosearchconvolutionalfeaturefeaturesmodelcardiactreatmentaccuratelyintegratesOSAGannetdeepresidualGaborKuwaharafilterneuralextractionoptimizationdemonstratedimprovedaccuracy96truepositiveratenegativepredictivevalue9412ScarLategadoliniumenhanced-cardiacmagneticresonanceplaycriticalroleevaluatingpathologyservesvitalindicatorimpactingprognosisdecisionsHoweversegmentingassessingpresentchallengesduecomplexcompositionimagingartifactsExistingmethodsoftenlackprecisionrobustnesslimitingclinicalapplicabilityworkproposesnovelmethodologyoptimalalgorithmoptimization-enabledhybridInitiallyinputimagepre-processedadaptiveapproachcombinesmyocardiumviaregion-basedSubsequentlyoptimizinghyperparametersstageintroducestwosetstechniques:localizationtextureanalysisspatial/temporalcomplementedfusionfractionalconceptinputtedcustomized1DcomprehensiveevaluationeffectivenessclassifyingofferingdiseaseassessmentplanningpatientsMoreoverproposedsuperiorperformanceachieving45%9542%48%20%18%devised1450%99%1074%975%79%1126%traditionalmodelsSegmentationHDResC-NetgannetAdaptiveCNNDRNMyocardiumR-CNN

Similar Articles

Cited By