Loading…
Automatic Left Ventricle Segmentation from Short-Axis MRI Images Using U-Net with Study of the Papillary Muscles’ Removal Effect
Purpose In clinical routines, the evaluation of cardiac wall motion is based on manual segmentation of ventricular contours. This task is time-consuming and leads to inter-observer variability. In this context, the aim of this paper is to propose a fully automatic method based on U-net architecture...
Saved in:
| Published in: | Journal of medical and biological engineering 2023-06, Vol.43 (3), p.278-290 |
|---|---|
| Main Authors: | , , , , |
| Format: | Article |
| Language: | English |
| Subjects: | |
| Citations: | Items that this one cites |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| cited_by | |
|---|---|
| cites | cdi_FETCH-LOGICAL-c304t-99284a9fb1a0f02ccdd83eeaac36e4a797a79c70169189f6101f7abac2cd7fe73 |
| container_end_page | 290 |
| container_issue | 3 |
| container_start_page | 278 |
| container_title | Journal of medical and biological engineering |
| container_volume | 43 |
| creator | Baccouch, Wafa Oueslati, Sameh Solaiman, Basel Lahidheb, Dhaker Labidi, Salam |
| description | Purpose
In clinical routines, the evaluation of cardiac wall motion is based on manual segmentation of ventricular contours. This task is time-consuming and leads to inter-observer variability. In this context, the aim of this paper is to propose a fully automatic method based on U-net architecture for left ventricle (LV) segmentation while studying the impact of papillary muscles presence and elimination on the segmentation accuracy.
Methods
In this work, we developed and evaluated an automatic approach based on U-Net architecture for LV segmentation. We started with a preprocessing pipeline which consists in cropping original images using convolutional neural network (CNN) and eliminating pillars using morphological operators. Regarding segmentation, our neural network was trained and validated using ACDC dataset composed of 150 patients. The performance of the proposed method was evaluated on an internal database composed of 100 patients (more than 2500 frames) using technical metrics including Hausdorff distance (HD), Jaccard coefficient (IoU), and Dice Similarity Coefficient (DSC).
Results
A comparative study demonstrated that the proposed architecture outperformed the original U-Net. Quantitative analysis of the obtained results confirmed the strength of our method that reveals the superlative segmentation performance as evaluated using the following indices including
HD
= 6.541 ± 1.6 mm,
IoU
= 94.85 ± 2%, and
DSC
= 93.27 ± 5% with
p
value |
| doi_str_mv | 10.1007/s40846-023-00794-z |
| format | article |
| fullrecord | <record><control><sourceid>proquest_hal_p</sourceid><recordid>TN_cdi_hal_primary_oai_HAL_hal_04543126v1</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2831807325</sourcerecordid><originalsourceid>FETCH-LOGICAL-c304t-99284a9fb1a0f02ccdd83eeaac36e4a797a79c70169189f6101f7abac2cd7fe73</originalsourceid><addsrcrecordid>eNp9kc1qGzEUhUVoISbJC3Ql6KoLNfrzaLQ0IT8Gpylx3K1Q5Ct7wszIlTRpk1XIW_T1-iRVMqXdRXAR997vHC4chD4w-plRqo6TpLWsCOWClFZL8riHJpxpTaSaqndowiqqCdX1dB8dpXRHyxO6qlg9Qc-zIYfO5sbhBfiMv0GfY-NawEvYdKUpq9BjH0OHl9sQM5n9bBK-vJ7jeWc3kPAqNf0Gr8gXyPhHk7d4mYf1Aw4e5y3gr3bXtK2ND_hySMU2_X76ha-hC_e2xafeg8uH6L23bYKjv_8BWp2d3pxckMXV-fxktiBOUJmJ1ryWVvtbZqmn3Ln1uhYA1jpRgbRKq1JOUVZpVmtfMcq8srfWcbdWHpQ4QJ9G361tzS42XbnKBNuYi9nCvMyonErBeHXPCvtxZHcxfB8gZXMXhtiX8wyvBaupEnxaKD5SLoaUIvh_toyal2jMGI0p0ZjXaMxjEYlRlArcbyD-t35D9Qc5u5MB</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2831807325</pqid></control><display><type>article</type><title>Automatic Left Ventricle Segmentation from Short-Axis MRI Images Using U-Net with Study of the Papillary Muscles’ Removal Effect</title><source>Springer Nature</source><creator>Baccouch, Wafa ; Oueslati, Sameh ; Solaiman, Basel ; Lahidheb, Dhaker ; Labidi, Salam</creator><creatorcontrib>Baccouch, Wafa ; Oueslati, Sameh ; Solaiman, Basel ; Lahidheb, Dhaker ; Labidi, Salam</creatorcontrib><description>Purpose
In clinical routines, the evaluation of cardiac wall motion is based on manual segmentation of ventricular contours. This task is time-consuming and leads to inter-observer variability. In this context, the aim of this paper is to propose a fully automatic method based on U-net architecture for left ventricle (LV) segmentation while studying the impact of papillary muscles presence and elimination on the segmentation accuracy.
Methods
In this work, we developed and evaluated an automatic approach based on U-Net architecture for LV segmentation. We started with a preprocessing pipeline which consists in cropping original images using convolutional neural network (CNN) and eliminating pillars using morphological operators. Regarding segmentation, our neural network was trained and validated using ACDC dataset composed of 150 patients. The performance of the proposed method was evaluated on an internal database composed of 100 patients (more than 2500 frames) using technical metrics including Hausdorff distance (HD), Jaccard coefficient (IoU), and Dice Similarity Coefficient (DSC).
Results
A comparative study demonstrated that the proposed architecture outperformed the original U-Net. Quantitative analysis of the obtained results confirmed the strength of our method that reveals the superlative segmentation performance as evaluated using the following indices including
HD
= 6.541 ± 1.6 mm,
IoU
= 94.85 ± 2%, and
DSC
= 93.27 ± 5% with
p
value < 0.0032. After the preprocessing application, the segmentation accuracy was improved. Thus, new mean HD, IoU, and DSC were 5.034 ± 2 mm, 98.83 ± 3.4%, and 98.04 ± 4%, respectively, with
p
value < 0.0018. Clinically, pillars’ exclusion facilitated middle and apical sections’ interpretation and helped in pathologies localization and clinical parameters’ estimation.
Conclusion
Experimental results demonstrate that the proposed approach offers a promising tool for LV segmentation and verifies its potential clinical applicability. In addition, pillars’ elimination using morphological operations proves its usefulness in improving segmentation accuracy.</description><identifier>ISSN: 1609-0985</identifier><identifier>EISSN: 2199-4757</identifier><identifier>DOI: 10.1007/s40846-023-00794-z</identifier><language>eng</language><publisher>Berlin/Heidelberg: Springer Berlin Heidelberg</publisher><subject>Accuracy ; Artificial neural networks ; Biological Techniques ; Biomedical and Life Sciences ; Biomedical Engineering and Bioengineering ; Biomedical Engineering/Biotechnology ; Biomedicine ; Comparative studies ; Engineering Sciences ; Image processing ; Image segmentation ; Localization ; Medical imaging ; Metric space ; Morphology ; Muscles ; Neural networks ; Original Article ; Parameter estimation ; Performance evaluation ; Preprocessing ; Regenerative Medicine/Tissue Engineering ; Ventricle</subject><ispartof>Journal of medical and biological engineering, 2023-06, Vol.43 (3), p.278-290</ispartof><rights>Taiwanese Society of Biomedical Engineering 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.</rights><rights>Distributed under a Creative Commons Attribution 4.0 International License</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c304t-99284a9fb1a0f02ccdd83eeaac36e4a797a79c70169189f6101f7abac2cd7fe73</cites><orcidid>0000-0001-5775-4864 ; 0000-0002-5977-230X</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>230,314,777,781,882,27905,27906</link.rule.ids><backlink>$$Uhttps://imt-atlantique.hal.science/hal-04543126$$DView record in HAL$$Hfree_for_read</backlink></links><search><creatorcontrib>Baccouch, Wafa</creatorcontrib><creatorcontrib>Oueslati, Sameh</creatorcontrib><creatorcontrib>Solaiman, Basel</creatorcontrib><creatorcontrib>Lahidheb, Dhaker</creatorcontrib><creatorcontrib>Labidi, Salam</creatorcontrib><title>Automatic Left Ventricle Segmentation from Short-Axis MRI Images Using U-Net with Study of the Papillary Muscles’ Removal Effect</title><title>Journal of medical and biological engineering</title><addtitle>J. Med. Biol. Eng</addtitle><description>Purpose
In clinical routines, the evaluation of cardiac wall motion is based on manual segmentation of ventricular contours. This task is time-consuming and leads to inter-observer variability. In this context, the aim of this paper is to propose a fully automatic method based on U-net architecture for left ventricle (LV) segmentation while studying the impact of papillary muscles presence and elimination on the segmentation accuracy.
Methods
In this work, we developed and evaluated an automatic approach based on U-Net architecture for LV segmentation. We started with a preprocessing pipeline which consists in cropping original images using convolutional neural network (CNN) and eliminating pillars using morphological operators. Regarding segmentation, our neural network was trained and validated using ACDC dataset composed of 150 patients. The performance of the proposed method was evaluated on an internal database composed of 100 patients (more than 2500 frames) using technical metrics including Hausdorff distance (HD), Jaccard coefficient (IoU), and Dice Similarity Coefficient (DSC).
Results
A comparative study demonstrated that the proposed architecture outperformed the original U-Net. Quantitative analysis of the obtained results confirmed the strength of our method that reveals the superlative segmentation performance as evaluated using the following indices including
HD
= 6.541 ± 1.6 mm,
IoU
= 94.85 ± 2%, and
DSC
= 93.27 ± 5% with
p
value < 0.0032. After the preprocessing application, the segmentation accuracy was improved. Thus, new mean HD, IoU, and DSC were 5.034 ± 2 mm, 98.83 ± 3.4%, and 98.04 ± 4%, respectively, with
p
value < 0.0018. Clinically, pillars’ exclusion facilitated middle and apical sections’ interpretation and helped in pathologies localization and clinical parameters’ estimation.
Conclusion
Experimental results demonstrate that the proposed approach offers a promising tool for LV segmentation and verifies its potential clinical applicability. In addition, pillars’ elimination using morphological operations proves its usefulness in improving segmentation accuracy.</description><subject>Accuracy</subject><subject>Artificial neural networks</subject><subject>Biological Techniques</subject><subject>Biomedical and Life Sciences</subject><subject>Biomedical Engineering and Bioengineering</subject><subject>Biomedical Engineering/Biotechnology</subject><subject>Biomedicine</subject><subject>Comparative studies</subject><subject>Engineering Sciences</subject><subject>Image processing</subject><subject>Image segmentation</subject><subject>Localization</subject><subject>Medical imaging</subject><subject>Metric space</subject><subject>Morphology</subject><subject>Muscles</subject><subject>Neural networks</subject><subject>Original Article</subject><subject>Parameter estimation</subject><subject>Performance evaluation</subject><subject>Preprocessing</subject><subject>Regenerative Medicine/Tissue Engineering</subject><subject>Ventricle</subject><issn>1609-0985</issn><issn>2199-4757</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><recordid>eNp9kc1qGzEUhUVoISbJC3Ql6KoLNfrzaLQ0IT8Gpylx3K1Q5Ct7wszIlTRpk1XIW_T1-iRVMqXdRXAR997vHC4chD4w-plRqo6TpLWsCOWClFZL8riHJpxpTaSaqndowiqqCdX1dB8dpXRHyxO6qlg9Qc-zIYfO5sbhBfiMv0GfY-NawEvYdKUpq9BjH0OHl9sQM5n9bBK-vJ7jeWc3kPAqNf0Gr8gXyPhHk7d4mYf1Aw4e5y3gr3bXtK2ND_hySMU2_X76ha-hC_e2xafeg8uH6L23bYKjv_8BWp2d3pxckMXV-fxktiBOUJmJ1ryWVvtbZqmn3Ln1uhYA1jpRgbRKq1JOUVZpVmtfMcq8srfWcbdWHpQ4QJ9G361tzS42XbnKBNuYi9nCvMyonErBeHXPCvtxZHcxfB8gZXMXhtiX8wyvBaupEnxaKD5SLoaUIvh_toyal2jMGI0p0ZjXaMxjEYlRlArcbyD-t35D9Qc5u5MB</recordid><startdate>20230601</startdate><enddate>20230601</enddate><creator>Baccouch, Wafa</creator><creator>Oueslati, Sameh</creator><creator>Solaiman, Basel</creator><creator>Lahidheb, Dhaker</creator><creator>Labidi, Salam</creator><general>Springer Berlin Heidelberg</general><general>Springer Nature B.V</general><general>Springer Verlag</general><scope>AAYXX</scope><scope>CITATION</scope><scope>K9.</scope><scope>1XC</scope><scope>VOOES</scope><orcidid>https://orcid.org/0000-0001-5775-4864</orcidid><orcidid>https://orcid.org/0000-0002-5977-230X</orcidid></search><sort><creationdate>20230601</creationdate><title>Automatic Left Ventricle Segmentation from Short-Axis MRI Images Using U-Net with Study of the Papillary Muscles’ Removal Effect</title><author>Baccouch, Wafa ; Oueslati, Sameh ; Solaiman, Basel ; Lahidheb, Dhaker ; Labidi, Salam</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c304t-99284a9fb1a0f02ccdd83eeaac36e4a797a79c70169189f6101f7abac2cd7fe73</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Accuracy</topic><topic>Artificial neural networks</topic><topic>Biological Techniques</topic><topic>Biomedical and Life Sciences</topic><topic>Biomedical Engineering and Bioengineering</topic><topic>Biomedical Engineering/Biotechnology</topic><topic>Biomedicine</topic><topic>Comparative studies</topic><topic>Engineering Sciences</topic><topic>Image processing</topic><topic>Image segmentation</topic><topic>Localization</topic><topic>Medical imaging</topic><topic>Metric space</topic><topic>Morphology</topic><topic>Muscles</topic><topic>Neural networks</topic><topic>Original Article</topic><topic>Parameter estimation</topic><topic>Performance evaluation</topic><topic>Preprocessing</topic><topic>Regenerative Medicine/Tissue Engineering</topic><topic>Ventricle</topic><toplevel>online_resources</toplevel><creatorcontrib>Baccouch, Wafa</creatorcontrib><creatorcontrib>Oueslati, Sameh</creatorcontrib><creatorcontrib>Solaiman, Basel</creatorcontrib><creatorcontrib>Lahidheb, Dhaker</creatorcontrib><creatorcontrib>Labidi, Salam</creatorcontrib><collection>CrossRef</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>Hyper Article en Ligne (HAL)</collection><collection>Hyper Article en Ligne (HAL) (Open Access)</collection><jtitle>Journal of medical and biological engineering</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Baccouch, Wafa</au><au>Oueslati, Sameh</au><au>Solaiman, Basel</au><au>Lahidheb, Dhaker</au><au>Labidi, Salam</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Automatic Left Ventricle Segmentation from Short-Axis MRI Images Using U-Net with Study of the Papillary Muscles’ Removal Effect</atitle><jtitle>Journal of medical and biological engineering</jtitle><stitle>J. Med. Biol. Eng</stitle><date>2023-06-01</date><risdate>2023</risdate><volume>43</volume><issue>3</issue><spage>278</spage><epage>290</epage><pages>278-290</pages><issn>1609-0985</issn><eissn>2199-4757</eissn><abstract>Purpose
In clinical routines, the evaluation of cardiac wall motion is based on manual segmentation of ventricular contours. This task is time-consuming and leads to inter-observer variability. In this context, the aim of this paper is to propose a fully automatic method based on U-net architecture for left ventricle (LV) segmentation while studying the impact of papillary muscles presence and elimination on the segmentation accuracy.
Methods
In this work, we developed and evaluated an automatic approach based on U-Net architecture for LV segmentation. We started with a preprocessing pipeline which consists in cropping original images using convolutional neural network (CNN) and eliminating pillars using morphological operators. Regarding segmentation, our neural network was trained and validated using ACDC dataset composed of 150 patients. The performance of the proposed method was evaluated on an internal database composed of 100 patients (more than 2500 frames) using technical metrics including Hausdorff distance (HD), Jaccard coefficient (IoU), and Dice Similarity Coefficient (DSC).
Results
A comparative study demonstrated that the proposed architecture outperformed the original U-Net. Quantitative analysis of the obtained results confirmed the strength of our method that reveals the superlative segmentation performance as evaluated using the following indices including
HD
= 6.541 ± 1.6 mm,
IoU
= 94.85 ± 2%, and
DSC
= 93.27 ± 5% with
p
value < 0.0032. After the preprocessing application, the segmentation accuracy was improved. Thus, new mean HD, IoU, and DSC were 5.034 ± 2 mm, 98.83 ± 3.4%, and 98.04 ± 4%, respectively, with
p
value < 0.0018. Clinically, pillars’ exclusion facilitated middle and apical sections’ interpretation and helped in pathologies localization and clinical parameters’ estimation.
Conclusion
Experimental results demonstrate that the proposed approach offers a promising tool for LV segmentation and verifies its potential clinical applicability. In addition, pillars’ elimination using morphological operations proves its usefulness in improving segmentation accuracy.</abstract><cop>Berlin/Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><doi>10.1007/s40846-023-00794-z</doi><tpages>13</tpages><orcidid>https://orcid.org/0000-0001-5775-4864</orcidid><orcidid>https://orcid.org/0000-0002-5977-230X</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1609-0985 |
| ispartof | Journal of medical and biological engineering, 2023-06, Vol.43 (3), p.278-290 |
| issn | 1609-0985 2199-4757 |
| language | eng |
| recordid | cdi_hal_primary_oai_HAL_hal_04543126v1 |
| source | Springer Nature |
| subjects | Accuracy Artificial neural networks Biological Techniques Biomedical and Life Sciences Biomedical Engineering and Bioengineering Biomedical Engineering/Biotechnology Biomedicine Comparative studies Engineering Sciences Image processing Image segmentation Localization Medical imaging Metric space Morphology Muscles Neural networks Original Article Parameter estimation Performance evaluation Preprocessing Regenerative Medicine/Tissue Engineering Ventricle |
| title | Automatic Left Ventricle Segmentation from Short-Axis MRI Images Using U-Net with Study of the Papillary Muscles’ Removal Effect |
| url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-19T01%3A28%3A57IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_hal_p&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Automatic%20Left%20Ventricle%20Segmentation%20from%20Short-Axis%20MRI%20Images%20Using%20U-Net%20with%20Study%20of%20the%20Papillary%20Muscles%E2%80%99%20Removal%20Effect&rft.jtitle=Journal%20of%20medical%20and%20biological%20engineering&rft.au=Baccouch,%20Wafa&rft.date=2023-06-01&rft.volume=43&rft.issue=3&rft.spage=278&rft.epage=290&rft.pages=278-290&rft.issn=1609-0985&rft.eissn=2199-4757&rft_id=info:doi/10.1007/s40846-023-00794-z&rft_dat=%3Cproquest_hal_p%3E2831807325%3C/proquest_hal_p%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c304t-99284a9fb1a0f02ccdd83eeaac36e4a797a79c70169189f6101f7abac2cd7fe73%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2831807325&rft_id=info:pmid/&rfr_iscdi=true |