Segmentation Improvement of Cardiac Regions in MRI using Hybrid Early-Late Fusion U-Net (HELFU-Net).

Saved in:
Bibliographic Details
Title: Segmentation Improvement of Cardiac Regions in MRI using Hybrid Early-Late Fusion U-Net (HELFU-Net).
Authors: Salim, Ula T.1 ula.tariq@uomosul.edu.iq, Dawwd, Shefa A.1 shefa.dawwd@uomosul.edu.iq, Ali, Fakhrulddin H.1 fhazaa@uomosul.edu.iq
Source: Al-Rafadain Engineering Journal. Jun2026, Vol. 31 Issue 2, p38-45. 8p.
Subjects: Image segmentation, Image fusion, Deep learning, Computer-assisted image analysis (Medicine)
Abstract (English): Image fusion and the U-Net architecture have been successfully applied to many real applications, in particular, cardiac segmentation. This paper suggests a new version named Hybrid Early-Late Fusion U-Net(HELFU-Net) to segment the cardiac structure into regions. The design has been built by extending the U-Net with five encoder branches and one decoder branch. The encoder branches take advantage of the adjacency property within the cardiac slice-images stack to boost the accuracy of the target image. The first branch of the encoder in the U-Net is to merge adjacent images using the concept of early-stage fusion. The next three branches apply late-stage fusion to the features of adjacent slices that are processed separately. The last branch is for the target slice of the image, while the decoder branch retrieves the data. This design boosts spatial information collection using only 2D image slices. HELFU-Net is evaluated using a public dataset of the ACDC challenge. The experimental results gave mean dice coefficients of 0.942, 0.856, and 0.893 for left ventricular cavity, right ventricular cavity, and left ventricular myocardium, respectively, on the test dataset. Additionally, the suggested HELFU-Net gives 94.9% comparable predicted accuracy on the test dataset over a test time of 1.065 sec. [ABSTRACT FROM AUTHOR]
Abstract (Arabic): تركز المقالة على تطوير وتقييم بنية تعلم عميق جديدة تُسمى الشبكة الهجينة للدمج المبكر والمتأخر U-Net (HELFU-Net) لتحسين تجزئة مناطق القلب في التصوير بالرنين المغناطيسي (MRI). تقوم HELFU-Net بتوسيع شبكة U-Net التقليدية من خلال دمج خمسة فروع مشفرة (encoder branches) تدمج المعلومات المكانية من شرائح صور القلب المجاورة باستخدام تقنيات الدمج في المراحل المبكرة والمتأخرة، مما يعزز دقة التجزئة مع الحفاظ على كفاءة الحوسبة. تم تقييم HELFU-Net على مجموعة بيانات تحدي التشخيص القلبي الآلي (Automated Cardiac Diagnosis Challenge - ACDC) العامة، حيث حققت متوسط معاملات التشابه Dice بقيم 0.942 لتجويف البطين الأيسر، و0.856 لتجويف البطين الأيمن، و0.893 لعضلة البطين الأيسر (myocardium)، مع زمن اختبار يقارب 1.065 ثانية لكل صورة. تسلط الدراسة الضوء على أن HELFU-Net تقدم دقة تنافسية مع عدد أقل من المعاملات وتكلفة حسابية منخفضة مقارنة بالطرق الحالية، مع تفوق خاص في تجزئة تجويف البطين الأيسر، رغم الحاجة إلى تحسينات إضافية في تجزئة عضلة القلب والبطين الأيمن. يقترح العمل المستقبلي استكشاف استراتيجيات الدمج على مستويات كل من المشفر (encoder) وفك التشفير (decoder) وتطبيق النهج على مهام تصوير طبي أخرى. [Extracted from the article]
Copyright of Al-Rafadain Engineering Journal is the property of Republic of Iraq Ministry of Higher Education & Scientific Research (MOHESR) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Links:
  – Type: pdflink
Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 194879651
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Segmentation Improvement of Cardiac Regions in MRI using Hybrid Early-Late Fusion U-Net (HELFU-Net).
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Salim%2C+Ula+T%2E%22">Salim, Ula T.</searchLink><relatesTo>1</relatesTo><i> ula.tariq@uomosul.edu.iq</i><br /><searchLink fieldCode="AR" term="%22Dawwd%2C+Shefa+A%2E%22">Dawwd, Shefa A.</searchLink><relatesTo>1</relatesTo><i> shefa.dawwd@uomosul.edu.iq</i><br /><searchLink fieldCode="AR" term="%22Ali%2C+Fakhrulddin+H%2E%22">Ali, Fakhrulddin H.</searchLink><relatesTo>1</relatesTo><i> fhazaa@uomosul.edu.iq</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Al-Rafadain+Engineering+Journal%22">Al-Rafadain Engineering Journal</searchLink>. Jun2026, Vol. 31 Issue 2, p38-45. 8p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Image+segmentation%22">Image segmentation</searchLink><br /><searchLink fieldCode="DE" term="%22Image+fusion%22">Image fusion</searchLink><br /><searchLink fieldCode="DE" term="%22Deep+learning%22">Deep learning</searchLink><br /><searchLink fieldCode="DE" term="%22Computer-assisted+image+analysis+%28Medicine%29%22">Computer-assisted image analysis (Medicine)</searchLink>
– Name: Abstract
  Label: Abstract (English)
  Group: Ab
  Data: Image fusion and the U-Net architecture have been successfully applied to many real applications, in particular, cardiac segmentation. This paper suggests a new version named Hybrid Early-Late Fusion U-Net(HELFU-Net) to segment the cardiac structure into regions. The design has been built by extending the U-Net with five encoder branches and one decoder branch. The encoder branches take advantage of the adjacency property within the cardiac slice-images stack to boost the accuracy of the target image. The first branch of the encoder in the U-Net is to merge adjacent images using the concept of early-stage fusion. The next three branches apply late-stage fusion to the features of adjacent slices that are processed separately. The last branch is for the target slice of the image, while the decoder branch retrieves the data. This design boosts spatial information collection using only 2D image slices. HELFU-Net is evaluated using a public dataset of the ACDC challenge. The experimental results gave mean dice coefficients of 0.942, 0.856, and 0.893 for left ventricular cavity, right ventricular cavity, and left ventricular myocardium, respectively, on the test dataset. Additionally, the suggested HELFU-Net gives 94.9% comparable predicted accuracy on the test dataset over a test time of 1.065 sec. [ABSTRACT FROM AUTHOR]
– Name: Abstract
  Label: Abstract (Arabic)
  Group: Ab
  Data: تركز المقالة على تطوير وتقييم بنية تعلم عميق جديدة تُسمى الشبكة الهجينة للدمج المبكر والمتأخر U-Net (HELFU-Net) لتحسين تجزئة مناطق القلب في التصوير بالرنين المغناطيسي (MRI). تقوم HELFU-Net بتوسيع شبكة U-Net التقليدية من خلال دمج خمسة فروع مشفرة (encoder branches) تدمج المعلومات المكانية من شرائح صور القلب المجاورة باستخدام تقنيات الدمج في المراحل المبكرة والمتأخرة، مما يعزز دقة التجزئة مع الحفاظ على كفاءة الحوسبة. تم تقييم HELFU-Net على مجموعة بيانات تحدي التشخيص القلبي الآلي (Automated Cardiac Diagnosis Challenge - ACDC) العامة، حيث حققت متوسط معاملات التشابه Dice بقيم 0.942 لتجويف البطين الأيسر، و0.856 لتجويف البطين الأيمن، و0.893 لعضلة البطين الأيسر (myocardium)، مع زمن اختبار يقارب 1.065 ثانية لكل صورة. تسلط الدراسة الضوء على أن HELFU-Net تقدم دقة تنافسية مع عدد أقل من المعاملات وتكلفة حسابية منخفضة مقارنة بالطرق الحالية، مع تفوق خاص في تجزئة تجويف البطين الأيسر، رغم الحاجة إلى تحسينات إضافية في تجزئة عضلة القلب والبطين الأيمن. يقترح العمل المستقبلي استكشاف استراتيجيات الدمج على مستويات كل من المشفر (encoder) وفك التشفير (decoder) وتطبيق النهج على مهام تصوير طبي أخرى. [Extracted from the article]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Al-Rafadain Engineering Journal is the property of Republic of Iraq Ministry of Higher Education & Scientific Research (MOHESR) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=194879651
RecordInfo BibRecord:
  BibEntity:
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 8
        StartPage: 38
    Subjects:
      – SubjectFull: Image segmentation
        Type: general
      – SubjectFull: Image fusion
        Type: general
      – SubjectFull: Deep learning
        Type: general
      – SubjectFull: Computer-assisted image analysis (Medicine)
        Type: general
    Titles:
      – TitleFull: Segmentation Improvement of Cardiac Regions in MRI using Hybrid Early-Late Fusion U-Net (HELFU-Net).
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Salim, Ula T.
      – PersonEntity:
          Name:
            NameFull: Dawwd, Shefa A.
      – PersonEntity:
          Name:
            NameFull: Ali, Fakhrulddin H.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 06
              Text: Jun2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 18130526
          Numbering:
            – Type: volume
              Value: 31
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: Al-Rafadain Engineering Journal
              Type: main
ResultId 1