A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Automatic 4D mitral valve segmentation from transesophageal echocardiography: a semi-supervised learning approach
https://orcid.org/0000-0002-5624-2776
https://orcid.org/0000-0002-3974-5945
https://orcid.org/0009-0009-8038-5476
https://orcid.org/0000-0002-4102-6405
https://orcid.org/0000-0002-3075-0203
https://orcid.org/0000-0002-3691-1709
https://orcid.org/0000-0002-4834-2187
https://orcid.org/0000-0001-7115-0151
Abstract Performing automatic and standardized 4D TEE segmentation and mitral valve analysis is challenging due to the limitations of echocardiography and the scarcity of manually annotated 4D images. This work proposes a semi-supervised training strategy using pseudo labelling for MV segmentation in 4D TEE; it employs a Teacher-Student framework to ensure reliable pseudo-label generation. 120 4D TEE recordings from 60 candidates for MV repair are used. The Teacher model, an ensemble of three convolutional neural networks, is trained on end-systole and end-diastole frames and is used to generate MV pseudo-segmentations on intermediate frames of the cardiac cycle. The pseudo-annotated frames augment the Student model’s training set, improving segmentation accuracy and temporal consistency. The Student outperforms individual Teachers, achieving a Dice score of 0.82, an average surface distance of 0.37 mm, and a 95% Hausdorff distance of 1.72 mm for MV leaflets. The Student model demonstrates reliable frame-by-frame MV segmentation, accurately capturing leaflet morphology and dynamics throughout the cardiac cycle, with a significant reduction in inference time compared to the ensemble. This approach greatly reduces manual annotation workload and ensures reliable, repeatable, and time-efficient MV analysis. Our method holds strong potential to enhance the precision and efficiency of MV diagnostics and treatment planning in clinical settings. Graphical abstract
Automatic 4D mitral valve segmentation from transesophageal echocardiography: a semi-supervised learning approach
https://orcid.org/0000-0002-5624-2776
https://orcid.org/0000-0002-3974-5945
https://orcid.org/0009-0009-8038-5476
https://orcid.org/0000-0002-4102-6405
https://orcid.org/0000-0002-3075-0203
https://orcid.org/0000-0002-3691-1709
https://orcid.org/0000-0002-4834-2187
https://orcid.org/0000-0001-7115-0151
Abstract Performing automatic and standardized 4D TEE segmentation and mitral valve analysis is challenging due to the limitations of echocardiography and the scarcity of manually annotated 4D images. This work proposes a semi-supervised training strategy using pseudo labelling for MV segmentation in 4D TEE; it employs a Teacher-Student framework to ensure reliable pseudo-label generation. 120 4D TEE recordings from 60 candidates for MV repair are used. The Teacher model, an ensemble of three convolutional neural networks, is trained on end-systole and end-diastole frames and is used to generate MV pseudo-segmentations on intermediate frames of the cardiac cycle. The pseudo-annotated frames augment the Student model’s training set, improving segmentation accuracy and temporal consistency. The Student outperforms individual Teachers, achieving a Dice score of 0.82, an average surface distance of 0.37 mm, and a 95% Hausdorff distance of 1.72 mm for MV leaflets. The Student model demonstrates reliable frame-by-frame MV segmentation, accurately capturing leaflet morphology and dynamics throughout the cardiac cycle, with a significant reduction in inference time compared to the ensemble. This approach greatly reduces manual annotation workload and ensures reliable, repeatable, and time-efficient MV analysis. Our method holds strong potential to enhance the precision and efficiency of MV diagnostics and treatment planning in clinical settings. Graphical abstract
Automatic 4D mitral valve segmentation from transesophageal echocardiography: a semi-supervised learning approach
https://orcid.org/0000-0002-5624-2776
https://orcid.org/0000-0002-3974-5945
https://orcid.org/0009-0009-8038-5476
https://orcid.org/0000-0002-4102-6405
https://orcid.org/0000-0002-3075-0203
https://orcid.org/0000-0002-3691-1709
https://orcid.org/0000-0002-4834-2187
https://orcid.org/0000-0001-7115-0151
Abstract Performing automatic and standardized 4D TEE segmentation and mitral valve analysis is challenging due to the limitations of echocardiography and the scarcity of manually annotated 4D images. This work proposes a semi-supervised training strategy using pseudo labelling for MV segmentation in 4D TEE; it employs a Teacher-Student framework to ensure reliable pseudo-label generation. 120 4D TEE recordings from 60 candidates for MV repair are used. The Teacher model, an ensemble of three convolutional neural networks, is trained on end-systole and end-diastole frames and is used to generate MV pseudo-segmentations on intermediate frames of the cardiac cycle. The pseudo-annotated frames augment the Student model’s training set, improving segmentation accuracy and temporal consistency. The Student outperforms individual Teachers, achieving a Dice score of 0.82, an average surface distance of 0.37 mm, and a 95% Hausdorff distance of 1.72 mm for MV leaflets. The Student model demonstrates reliable frame-by-frame MV segmentation, accurately capturing leaflet morphology and dynamics throughout the cardiac cycle, with a significant reduction in inference time compared to the ensemble. This approach greatly reduces manual annotation workload and ensures reliable, repeatable, and time-efficient MV analysis. Our method holds strong potential to enhance the precision and efficiency of MV diagnostics and treatment planning in clinical settings. Graphical abstract
Automatic 4D mitral valve segmentation from transesophageal echocardiography: a semi-supervised learning approach
Med Biol Eng Comput
Munafò, Riccardo (author) / Saitta, Simone (author) / Tondi, Davide (author) / Ingallina, Giacomo (author) / Denti, Paolo (author) / Maisano, Francesco (author) / Agricola, Eustachio (author) / Votta, Emiliano (author)
2025-01-11
Article (Journal)
Electronic Resource
English
Semi-Automatic Segmentation of the Mitral Valve Leaflets on 4D Ultrasound Images
| BASE | 2015
Instance segmentation using semi-supervised learning for fire recognition
| Elsevier | 2022
Semi-supervised learning network for deep-sea nodule mineral image segmentation
| Elsevier | 2025