A Deep Learning-Based Fully Automated Pipeline for Regurgitant Mitral Valve Anatomy Analysis From 3D Echocardiography
Three-dimensional transesophageal echocardiography (3DTEE) is the recommended imaging technique for the assessment of mitral valve (MV) morphology and lesions in case of mitral regurgitation (MR) requiring surgical or transcatheter repair. Such assessment is key to thorough intervention planning and...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2024-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10380557/ |
_version_ | 1797299831193993216 |
---|---|
author | Riccardo Munafo Simone Saitta Giacomo Ingallina Paolo Denti Francesco Maisano Eustachio Agricola Alberto Redaelli Emiliano Votta |
author_facet | Riccardo Munafo Simone Saitta Giacomo Ingallina Paolo Denti Francesco Maisano Eustachio Agricola Alberto Redaelli Emiliano Votta |
author_sort | Riccardo Munafo |
collection | DOAJ |
description | Three-dimensional transesophageal echocardiography (3DTEE) is the recommended imaging technique for the assessment of mitral valve (MV) morphology and lesions in case of mitral regurgitation (MR) requiring surgical or transcatheter repair. Such assessment is key to thorough intervention planning and to intraprocedural guidance. However, it requires segmentation from 3DTEE images, which is time-consuming, operator-dependent, and often merely qualitative. In the present work, a novel workflow to quantify the patient-specific MV geometry from 3DTEE is proposed. The developed approach relies on a 3D multi-decoder residual convolutional neural network (CNN) with a U-Net architecture for multi-class segmentation of MV annulus and leaflets. The CNN was trained and tested on a dataset comprising 55 3DTEE examinations of MR-affected patients. After training, the CNN is embedded into a fully automatic, and hence fully repeatable, pipeline that refines the predicted segmentation, detects MV anatomical landmarks and quantifies MV morphology. The trained 3D CNN achieves an average Dice score of 0.82 ± 0.06, mean surface distance of 0.43 ± 0.14 mm and 95% Hausdorff Distance (HD) of 3.57 ± 1.56 mm before segmentation refinement, outperforming a state-of-the-art baseline residual U-Net architecture, and provides an unprecedented multi-class segmentation of the annulus, anterior and posterior leaflet. The automatic 3D linear morphological measurements of the annulus and leaflets, specifically diameters and lengths, exhibit differences of less than 1.45 mm when compared to ground truth values. These measurements also demonstrate strong overall agreement with analyses conducted by semi-automated commercial software. The whole process requires minimal user interaction and requires approximately 15 seconds. |
first_indexed | 2024-03-07T22:57:20Z |
format | Article |
id | doaj.art-882d574c7f0943c1a65f322f1fe62b3c |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-07T22:57:20Z |
publishDate | 2024-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-882d574c7f0943c1a65f322f1fe62b3c2024-02-23T00:00:26ZengIEEEIEEE Access2169-35362024-01-01125295530810.1109/ACCESS.2024.334969810380557A Deep Learning-Based Fully Automated Pipeline for Regurgitant Mitral Valve Anatomy Analysis From 3D EchocardiographyRiccardo Munafo0https://orcid.org/0000-0002-5624-2776Simone Saitta1https://orcid.org/0000-0002-3974-5945Giacomo Ingallina2Paolo Denti3Francesco Maisano4Eustachio Agricola5Alberto Redaelli6Emiliano Votta7https://orcid.org/0000-0001-7115-0151Department of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, ItalyDepartment of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, ItalyUnit of Cardiovascular Imaging, IRCCS San Raffaele Hospital, Milan, ItalyCardiac Surgery Department, IRCCS San Raffaele Hospital, Milan, ItalyCardiac Surgery Department, IRCCS San Raffaele Hospital, Milan, ItalyUnit of Cardiovascular Imaging, IRCCS San Raffaele Hospital, Milan, ItalyDepartment of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, ItalyDepartment of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, ItalyThree-dimensional transesophageal echocardiography (3DTEE) is the recommended imaging technique for the assessment of mitral valve (MV) morphology and lesions in case of mitral regurgitation (MR) requiring surgical or transcatheter repair. Such assessment is key to thorough intervention planning and to intraprocedural guidance. However, it requires segmentation from 3DTEE images, which is time-consuming, operator-dependent, and often merely qualitative. In the present work, a novel workflow to quantify the patient-specific MV geometry from 3DTEE is proposed. The developed approach relies on a 3D multi-decoder residual convolutional neural network (CNN) with a U-Net architecture for multi-class segmentation of MV annulus and leaflets. The CNN was trained and tested on a dataset comprising 55 3DTEE examinations of MR-affected patients. After training, the CNN is embedded into a fully automatic, and hence fully repeatable, pipeline that refines the predicted segmentation, detects MV anatomical landmarks and quantifies MV morphology. The trained 3D CNN achieves an average Dice score of 0.82 ± 0.06, mean surface distance of 0.43 ± 0.14 mm and 95% Hausdorff Distance (HD) of 3.57 ± 1.56 mm before segmentation refinement, outperforming a state-of-the-art baseline residual U-Net architecture, and provides an unprecedented multi-class segmentation of the annulus, anterior and posterior leaflet. The automatic 3D linear morphological measurements of the annulus and leaflets, specifically diameters and lengths, exhibit differences of less than 1.45 mm when compared to ground truth values. These measurements also demonstrate strong overall agreement with analyses conducted by semi-automated commercial software. The whole process requires minimal user interaction and requires approximately 15 seconds.https://ieeexplore.ieee.org/document/10380557/3D transesophageal echocardiographymitral regurgitationautomatic segmentationconvolutional neural networkmitral valve anatomy quantificationmitral valve |
spellingShingle | Riccardo Munafo Simone Saitta Giacomo Ingallina Paolo Denti Francesco Maisano Eustachio Agricola Alberto Redaelli Emiliano Votta A Deep Learning-Based Fully Automated Pipeline for Regurgitant Mitral Valve Anatomy Analysis From 3D Echocardiography IEEE Access 3D transesophageal echocardiography mitral regurgitation automatic segmentation convolutional neural network mitral valve anatomy quantification mitral valve |
title | A Deep Learning-Based Fully Automated Pipeline for Regurgitant Mitral Valve Anatomy Analysis From 3D Echocardiography |
title_full | A Deep Learning-Based Fully Automated Pipeline for Regurgitant Mitral Valve Anatomy Analysis From 3D Echocardiography |
title_fullStr | A Deep Learning-Based Fully Automated Pipeline for Regurgitant Mitral Valve Anatomy Analysis From 3D Echocardiography |
title_full_unstemmed | A Deep Learning-Based Fully Automated Pipeline for Regurgitant Mitral Valve Anatomy Analysis From 3D Echocardiography |
title_short | A Deep Learning-Based Fully Automated Pipeline for Regurgitant Mitral Valve Anatomy Analysis From 3D Echocardiography |
title_sort | deep learning based fully automated pipeline for regurgitant mitral valve anatomy analysis from 3d echocardiography |
topic | 3D transesophageal echocardiography mitral regurgitation automatic segmentation convolutional neural network mitral valve anatomy quantification mitral valve |
url | https://ieeexplore.ieee.org/document/10380557/ |
work_keys_str_mv | AT riccardomunafo adeeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT simonesaitta adeeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT giacomoingallina adeeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT paolodenti adeeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT francescomaisano adeeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT eustachioagricola adeeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT albertoredaelli adeeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT emilianovotta adeeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT riccardomunafo deeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT simonesaitta deeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT giacomoingallina deeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT paolodenti deeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT francescomaisano deeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT eustachioagricola deeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT albertoredaelli deeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography AT emilianovotta deeplearningbasedfullyautomatedpipelineforregurgitantmitralvalveanatomyanalysisfrom3dechocardiography |