Cargando…

Cardiac Magnetic Resonance Image Segmentation Method Based on Multi-Scale Feature Fusion and Sequence Relationship Learning

Accurate segmentation of the left atrial structure using magnetic resonance images provides an important basis for the diagnosis of atrial fibrillation (AF) and its treatment using robotic surgery. In this study, an image segmentation method based on sequence relationship learning and multi-scale fe...

Descripción completa

Detalles Bibliográficos
Autores principales: Qi, Yushi, Hu, Chunhu, Zuo, Liling, Yang, Bo, Lv, Youlong
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9865693/
https://www.ncbi.nlm.nih.gov/pubmed/36679487
http://dx.doi.org/10.3390/s23020690
_version_ 1784875901911040000
author Qi, Yushi
Hu, Chunhu
Zuo, Liling
Yang, Bo
Lv, Youlong
author_facet Qi, Yushi
Hu, Chunhu
Zuo, Liling
Yang, Bo
Lv, Youlong
author_sort Qi, Yushi
collection PubMed
description Accurate segmentation of the left atrial structure using magnetic resonance images provides an important basis for the diagnosis of atrial fibrillation (AF) and its treatment using robotic surgery. In this study, an image segmentation method based on sequence relationship learning and multi-scale feature fusion is proposed for 3D to 2D sequence conversion in cardiac magnetic resonance images and the varying scales of left atrial structures within different slices. Firstly, a convolutional neural network layer with an attention module was designed to extract and fuse contextual information at different scales in the image, to strengthen the target features using the correlation between features in different regions within the image, and to improve the network’s ability to distinguish the left atrial structure. Secondly, a recurrent neural network layer oriented to two-dimensional images was designed to capture the correlation of left atrial structures in adjacent slices by simulating the continuous relationship between sequential image slices. Finally, a combined loss function was constructed to reduce the effect of positive and negative sample imbalance and improve model stability. The Dice, IoU, and Hausdorff distance values reached 90.73%, 89.37%, and 4.803 mm, respectively, based on the LASC2013 (left atrial segmentation challenge in 2013) dataset; the corresponding values reached 92.05%, 89.41% and 9.056 mm, respectively, based on the ASC2018 (atrial segmentation challenge at 2018) dataset.
format Online
Article
Text
id pubmed-9865693
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-98656932023-01-22 Cardiac Magnetic Resonance Image Segmentation Method Based on Multi-Scale Feature Fusion and Sequence Relationship Learning Qi, Yushi Hu, Chunhu Zuo, Liling Yang, Bo Lv, Youlong Sensors (Basel) Article Accurate segmentation of the left atrial structure using magnetic resonance images provides an important basis for the diagnosis of atrial fibrillation (AF) and its treatment using robotic surgery. In this study, an image segmentation method based on sequence relationship learning and multi-scale feature fusion is proposed for 3D to 2D sequence conversion in cardiac magnetic resonance images and the varying scales of left atrial structures within different slices. Firstly, a convolutional neural network layer with an attention module was designed to extract and fuse contextual information at different scales in the image, to strengthen the target features using the correlation between features in different regions within the image, and to improve the network’s ability to distinguish the left atrial structure. Secondly, a recurrent neural network layer oriented to two-dimensional images was designed to capture the correlation of left atrial structures in adjacent slices by simulating the continuous relationship between sequential image slices. Finally, a combined loss function was constructed to reduce the effect of positive and negative sample imbalance and improve model stability. The Dice, IoU, and Hausdorff distance values reached 90.73%, 89.37%, and 4.803 mm, respectively, based on the LASC2013 (left atrial segmentation challenge in 2013) dataset; the corresponding values reached 92.05%, 89.41% and 9.056 mm, respectively, based on the ASC2018 (atrial segmentation challenge at 2018) dataset. MDPI 2023-01-07 /pmc/articles/PMC9865693/ /pubmed/36679487 http://dx.doi.org/10.3390/s23020690 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Qi, Yushi
Hu, Chunhu
Zuo, Liling
Yang, Bo
Lv, Youlong
Cardiac Magnetic Resonance Image Segmentation Method Based on Multi-Scale Feature Fusion and Sequence Relationship Learning
title Cardiac Magnetic Resonance Image Segmentation Method Based on Multi-Scale Feature Fusion and Sequence Relationship Learning
title_full Cardiac Magnetic Resonance Image Segmentation Method Based on Multi-Scale Feature Fusion and Sequence Relationship Learning
title_fullStr Cardiac Magnetic Resonance Image Segmentation Method Based on Multi-Scale Feature Fusion and Sequence Relationship Learning
title_full_unstemmed Cardiac Magnetic Resonance Image Segmentation Method Based on Multi-Scale Feature Fusion and Sequence Relationship Learning
title_short Cardiac Magnetic Resonance Image Segmentation Method Based on Multi-Scale Feature Fusion and Sequence Relationship Learning
title_sort cardiac magnetic resonance image segmentation method based on multi-scale feature fusion and sequence relationship learning
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9865693/
https://www.ncbi.nlm.nih.gov/pubmed/36679487
http://dx.doi.org/10.3390/s23020690
work_keys_str_mv AT qiyushi cardiacmagneticresonanceimagesegmentationmethodbasedonmultiscalefeaturefusionandsequencerelationshiplearning
AT huchunhu cardiacmagneticresonanceimagesegmentationmethodbasedonmultiscalefeaturefusionandsequencerelationshiplearning
AT zuoliling cardiacmagneticresonanceimagesegmentationmethodbasedonmultiscalefeaturefusionandsequencerelationshiplearning
AT yangbo cardiacmagneticresonanceimagesegmentationmethodbasedonmultiscalefeaturefusionandsequencerelationshiplearning
AT lvyoulong cardiacmagneticresonanceimagesegmentationmethodbasedonmultiscalefeaturefusionandsequencerelationshiplearning