Effect of Attention Mechanisms in a Fuzzified Nested U-Net for Medical Image Segmentation.

Saved in:

Bibliographic Details
Title:	Effect of Attention Mechanisms in a Fuzzified Nested U-Net for Medical Image Segmentation.
Authors:	Basheer, Noor M.¹ noor.23enp115@student.uomosul.edu.iq, Al-Saegh, Ali¹ ali.alsaegh@uomosul.edu.iq
Source:	Al-Rafadain Engineering Journal. Jun2026, Vol. 31 Issue 2, p23-37. 15p.
Subjects:	Image segmentation, Deep learning, Convolutional neural networks, Computer-assisted image analysis (Medicine), Cardiovascular disease diagnosis
Abstract (English):	Accurate segmentation of cardiac structures (right ventricle (RV), left ventricle (LV), and myocardium (Myo)) from cardiac MRI images plays an important role in the diagnosis and treatment of cardiovascular disease. However, despite all this, segmentation of these structures at the micro level remains a major challenge due to the complex anatomical diversity and noise inherent in MRI data. This paper examines the performance of various cardiac MRI segmentation techniques, with a primary focus on the overlapping U-Net structure, attention mechanisms, and fuzzy pooling strategies. This study comprehensively evaluates these methods, both independently and in combination, to determine their effectiveness in improving segmentation quality for the RV, LV, and myocardial regions. Additionally, the effect of thresholding strategies on segmentation accuracy is examined. The experimental results on the Automated Cardiac Diagnosis Challenge (ACDC) dataset show that the proposed model (combining nested U-Net, attention mechanisms, and fuzzy pooling) achieved a dice score of 98.20%, an accuracy of 96.83%, and a recall of 96.83%, superior to other basic methods. In comparison, the best-performing core model, ANU-Net, achieved a Dice score of 94.31%, accuracy of 95.19%, and recall of 93.44%. These findings underscore the superior performance of the hybrid model in terms of segmentation and boundary delineation accuracy. These results confirm the potential of hybrid deep learning models in developing cardiac image analysis. Future work will focus on improving these configurations across diverse datasets and also exploring real-time deployment strategies in clinical settings. [ABSTRACT FROM AUTHOR]
Abstract (Arabic):	تركز المقالة على تقييم تأثير آليات الانتباه المدمجة في بنية U-Net المتداخلة المُغَمَّشة (fuzzified nested U-Net) لتقسيم صور الرنين المغناطيسي القلبي. باستخدام مجموعة بيانات تحدي التشخيص القلبي الآلي (Automated Cardiac Diagnosis Challenge - ACDC)، تقترح الدراسة نموذجًا هجينًا يجمع بين U-Net المتداخلة وآليات الانتباه القنوي (channel attention)، المكاني (spatial attention)، والذاتي (self-attention)، إلى جانب التجميع الضبابي (fuzzy pooling)، لتحسين تقسيم البطين الأيمن (right ventricle - RV)، البطين الأيسر (left ventricle - LV)، وعضلة القلب (myocardium - MYO). تُظهر النتائج التجريبية أن تركيبات الانتباه المزدوجة، وخاصة الانتباه الذاتي مع الانتباه المكاني، عند عتبة مثلى تتراوح بين 0.3 و0.5، تحقق دقة تقسيم متفوقة، حيث تصل إلى متوسط درجة دايس (Dice score) بنسبة 98.2% ودقة تصنيف تبلغ 96% عبر خمس حالات قلبية. كما يُظهر النموذج قدرة جيدة على التعميم عند تطبيقه على مجموعة بيانات القلب الخارجية Sunnybrook، محافظًا على دقة عالية، ويتفوق على عدة طرق حديثة في التعلم العميق، مما يبرز إمكانياته في تحليل وتشخيص صور القلب سريريًا. [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

Description
Abstract:	Accurate segmentation of cardiac structures (right ventricle (RV), left ventricle (LV), and myocardium (Myo)) from cardiac MRI images plays an important role in the diagnosis and treatment of cardiovascular disease. However, despite all this, segmentation of these structures at the micro level remains a major challenge due to the complex anatomical diversity and noise inherent in MRI data. This paper examines the performance of various cardiac MRI segmentation techniques, with a primary focus on the overlapping U-Net structure, attention mechanisms, and fuzzy pooling strategies. This study comprehensively evaluates these methods, both independently and in combination, to determine their effectiveness in improving segmentation quality for the RV, LV, and myocardial regions. Additionally, the effect of thresholding strategies on segmentation accuracy is examined. The experimental results on the Automated Cardiac Diagnosis Challenge (ACDC) dataset show that the proposed model (combining nested U-Net, attention mechanisms, and fuzzy pooling) achieved a dice score of 98.20%, an accuracy of 96.83%, and a recall of 96.83%, superior to other basic methods. In comparison, the best-performing core model, ANU-Net, achieved a Dice score of 94.31%, accuracy of 95.19%, and recall of 93.44%. These findings underscore the superior performance of the hybrid model in terms of segmentation and boundary delineation accuracy. These results confirm the potential of hybrid deep learning models in developing cardiac image analysis. Future work will focus on improving these configurations across diverse datasets and also exploring real-time deployment strategies in clinical settings. [ABSTRACT FROM AUTHOR]
ISSN:	18130526