Uncertainty-Aware Adaptive Multiscale U-Net for Low-Contrast Cardiac Image Segmentation

Medical image analysis is critical for diagnosing and planning treatments, particularly in addressing heart disease, a leading cause of mortality worldwide. Precise segmentation of the left atrium, a key structure in cardiac imaging, is essential for detecting conditions such as atrial fibrillation,...

Full description

Saved in:
Bibliographic Details
Main Authors: A. S. M. Sharifuzzaman Sagar, Muhammad Zubair Islam, Jawad Tanveer, Hyung Seok Kim
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/4/2222
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850081707656478720
author A. S. M. Sharifuzzaman Sagar
Muhammad Zubair Islam
Jawad Tanveer
Hyung Seok Kim
author_facet A. S. M. Sharifuzzaman Sagar
Muhammad Zubair Islam
Jawad Tanveer
Hyung Seok Kim
author_sort A. S. M. Sharifuzzaman Sagar
collection DOAJ
description Medical image analysis is critical for diagnosing and planning treatments, particularly in addressing heart disease, a leading cause of mortality worldwide. Precise segmentation of the left atrium, a key structure in cardiac imaging, is essential for detecting conditions such as atrial fibrillation, heart failure, and stroke. However, its complex anatomy, subtle boundaries, and inter-patient variations make accurate segmentation challenging for traditional methods. Recent advancements in deep learning, especially semantic segmentation, have shown promise in addressing these limitations by enabling detailed, pixel-wise classification. This study proposes a novel segmentation framework Adaptive Multiscale U-Net (AMU-Net) combining Convolutional Neural Networks (CNNs) and transformer-based encoder–decoder architectures. The framework introduces a Contextual Dynamic Encoder (CDE) for extracting multi-scale features and capturing long-range dependencies. An Adaptive Feature Decoder Block (AFDB), leveraging an Adaptive Feature Attention Block (AFAB) improves boundary delineation. Additionally, a Spectral Synthesis Fusion Head (SFFH) synthesizes spectral and spatial features, enhancing segmentation performance in low-contrast regions. To ensure robustness, data augmentation techniques such as rotation, scaling, and flipping are applied. Laplacian approximation is employed for uncertainty estimation, enabling interpretability and identifying regions of low confidence. Our proposed model achieves a Dice score of 93.35, a Precision of 94.12, and a Recall of 92.78, outperforming existing methods.
format Article
id doaj-art-e6afebdbd44b4853a3dc779951629a19
institution DOAJ
issn 2076-3417
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-e6afebdbd44b4853a3dc779951629a192025-08-20T02:44:40ZengMDPI AGApplied Sciences2076-34172025-02-01154222210.3390/app15042222Uncertainty-Aware Adaptive Multiscale U-Net for Low-Contrast Cardiac Image SegmentationA. S. M. Sharifuzzaman Sagar0Muhammad Zubair Islam1Jawad Tanveer2Hyung Seok Kim3Department of Artificial Intelligence and Robotics, Sejong University, Seoul 05006, Republic of KoreaDepartment of Artificial Intelligence and Robotics, Sejong University, Seoul 05006, Republic of KoreaDepartment of Computer Science and Engineering, Sejong University, Seoul 05006, Republic of KoreaDepartment of Artificial Intelligence and Robotics, Sejong University, Seoul 05006, Republic of KoreaMedical image analysis is critical for diagnosing and planning treatments, particularly in addressing heart disease, a leading cause of mortality worldwide. Precise segmentation of the left atrium, a key structure in cardiac imaging, is essential for detecting conditions such as atrial fibrillation, heart failure, and stroke. However, its complex anatomy, subtle boundaries, and inter-patient variations make accurate segmentation challenging for traditional methods. Recent advancements in deep learning, especially semantic segmentation, have shown promise in addressing these limitations by enabling detailed, pixel-wise classification. This study proposes a novel segmentation framework Adaptive Multiscale U-Net (AMU-Net) combining Convolutional Neural Networks (CNNs) and transformer-based encoder–decoder architectures. The framework introduces a Contextual Dynamic Encoder (CDE) for extracting multi-scale features and capturing long-range dependencies. An Adaptive Feature Decoder Block (AFDB), leveraging an Adaptive Feature Attention Block (AFAB) improves boundary delineation. Additionally, a Spectral Synthesis Fusion Head (SFFH) synthesizes spectral and spatial features, enhancing segmentation performance in low-contrast regions. To ensure robustness, data augmentation techniques such as rotation, scaling, and flipping are applied. Laplacian approximation is employed for uncertainty estimation, enabling interpretability and identifying regions of low confidence. Our proposed model achieves a Dice score of 93.35, a Precision of 94.12, and a Recall of 92.78, outperforming existing methods.https://www.mdpi.com/2076-3417/15/4/2222deep learningmedical image analysisuncertainty estimationsegmentationradiomics
spellingShingle A. S. M. Sharifuzzaman Sagar
Muhammad Zubair Islam
Jawad Tanveer
Hyung Seok Kim
Uncertainty-Aware Adaptive Multiscale U-Net for Low-Contrast Cardiac Image Segmentation
Applied Sciences
deep learning
medical image analysis
uncertainty estimation
segmentation
radiomics
title Uncertainty-Aware Adaptive Multiscale U-Net for Low-Contrast Cardiac Image Segmentation
title_full Uncertainty-Aware Adaptive Multiscale U-Net for Low-Contrast Cardiac Image Segmentation
title_fullStr Uncertainty-Aware Adaptive Multiscale U-Net for Low-Contrast Cardiac Image Segmentation
title_full_unstemmed Uncertainty-Aware Adaptive Multiscale U-Net for Low-Contrast Cardiac Image Segmentation
title_short Uncertainty-Aware Adaptive Multiscale U-Net for Low-Contrast Cardiac Image Segmentation
title_sort uncertainty aware adaptive multiscale u net for low contrast cardiac image segmentation
topic deep learning
medical image analysis
uncertainty estimation
segmentation
radiomics
url https://www.mdpi.com/2076-3417/15/4/2222
work_keys_str_mv AT asmsharifuzzamansagar uncertaintyawareadaptivemultiscaleunetforlowcontrastcardiacimagesegmentation
AT muhammadzubairislam uncertaintyawareadaptivemultiscaleunetforlowcontrastcardiacimagesegmentation
AT jawadtanveer uncertaintyawareadaptivemultiscaleunetforlowcontrastcardiacimagesegmentation
AT hyungseokkim uncertaintyawareadaptivemultiscaleunetforlowcontrastcardiacimagesegmentation