Digitala Vetenskapliga Arkivet

Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Poincare guided geometric UNet for left atrial epicardial adipose tissue segmentation in Dixon MRI images
Linköping University, Department of Health, Medicine and Caring Sciences, Division of Diagnostics and Specialist Medicine. Linköping University, Faculty of Medicine and Health Sciences.
Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.
Linköping University, Department of Health, Medicine and Caring Sciences, Division of Diagnostics and Specialist Medicine. Linköping University, Faculty of Medicine and Health Sciences.ORCID iD: 0000-0001-6142-3005
Linköping University, Department of Health, Medicine and Caring Sciences, Division of Diagnostics and Specialist Medicine. Linköping University, Faculty of Medicine and Health Sciences. Region Östergötland, Heart Center, Department of Clinical Physiology in Linköping. Linköping University, Center for Medical Image Science and Visualization (CMIV).ORCID iD: 0000-0003-2198-9690
2025 (English)In: Scientific Reports, E-ISSN 2045-2322, Vol. 15, no 1, article id 25549Article in journal (Refereed) Published
Abstract [en]

Epicardial Adipose Tissue (EAT) is a recognized risk factor for cardiovascular diseases and plays a pivotal role in the pathophysiology of Atrial Fibrillation (AF). Accurate automatic segmentation of the EAT around the Left Atrium (LA) from Magnetic Resonance Imaging (MRI) data remains challenging. While Convolutional Neural Networks excel at multi-scale feature extraction using stacked convolutions, they struggle to capture long-range self-similarity and hierarchical relationships, which are essential in medical image segmentation. In this study, we present and validate PoinUNet, a deep learning model that integrates a Poincar & eacute; embedding layer into a 3D UNet to enhance LA wall and fat segmentation from Dixon MRI data. By using hyperbolic space learning, PoinUNet captures complex LA and EAT relationships and addresses class imbalance and fat geometry challenges using a new loss function. Sixty-six participants, including forty-eight AF patients, were scanned at 1.5T. The first network identified fat regions, while the second utilized Poincar & eacute; embeddings and convolutional layers for precise segmentation, enhanced by fat fraction maps. PoinUNet achieved a Dice Similarity Coefficient of 0.87 and a Hausdorff distance of 9.42 on the test set. This performance surpasses state-of-the-art methods, providing accurate quantification of the LA wall and LA EAT.

Place, publisher, year, edition, pages
NATURE PORTFOLIO , 2025. Vol. 15, no 1, article id 25549
Keywords [en]
Epicadial adipose tissue; Left atrium; Convolutional neural networks; Segmentation; Poincar & eacute; layer; Deep learning
National Category
Medical Imaging
Identifiers
URN: urn:nbn:se:liu:diva-216559DOI: 10.1038/s41598-025-10110-1ISI: 001530067200029PubMedID: 40664735Scopus ID: 2-s2.0-105010617414OAI: oai:DiVA.org:liu-216559DiVA, id: diva2:1990948
Note

Funding Agencies|Linkping University

Available from: 2025-08-21 Created: 2025-08-21 Last updated: 2025-09-25

Open Access in DiVA

fulltext(2682 kB)103 downloads
File information
File name FULLTEXT01.pdfFile size 2682 kBChecksum SHA-512
0c01ca5324639b7e568994f0ffdc0296cceebf24a59039ea4122c10a36298ae04f71cd7f11ffebaaebecbda153ddeff83419c76dee45b392bf2e82c79a4ca543
Type fulltextMimetype application/pdf

Other links

Publisher's full textPubMedScopus

Search in DiVA

By author/editor
Firouznia, MarjanYlipää, ErikHenningsson, MarkusCarlhäll, Carljohan
By organisation
Division of Diagnostics and Specialist MedicineFaculty of Medicine and Health SciencesMedia and Information TechnologyFaculty of Science & EngineeringDepartment of Clinical Physiology in LinköpingCenter for Medical Image Science and Visualization (CMIV)
In the same journal
Scientific Reports
Medical Imaging

Search outside of DiVA

GoogleGoogle Scholar
Total: 103 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
pubmed
urn-nbn

Altmetric score

doi
pubmed
urn-nbn
Total: 105 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf