Temporal dependency modeling for improved medical image segmentation: The R-UNet perspective

In this study, we propose a modified version of the widely used UNet architecture, enhanced by the integration of recurrent blocks at each step of the encoder (down-sampling) and decoder (up-sampling) stages. The proposed Recurrent UNet (R-UNet) architecture aims to improve the performance of semant...

Full description

Saved in:
Bibliographic Details
Main Authors: Ahmed Alweshah, Roohollah Barzamini, Farshid Hajati, Shoorangiz Shams Shamsabad Farahani, Mohammad Arabian, Behnaz Sohani
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:Franklin Open
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2773186324001129
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850054334516035584
author Ahmed Alweshah
Roohollah Barzamini
Farshid Hajati
Shoorangiz Shams Shamsabad Farahani
Mohammad Arabian
Behnaz Sohani
author_facet Ahmed Alweshah
Roohollah Barzamini
Farshid Hajati
Shoorangiz Shams Shamsabad Farahani
Mohammad Arabian
Behnaz Sohani
author_sort Ahmed Alweshah
collection DOAJ
description In this study, we propose a modified version of the widely used UNet architecture, enhanced by the integration of recurrent blocks at each step of the encoder (down-sampling) and decoder (up-sampling) stages. The proposed Recurrent UNet (R-UNet) architecture aims to improve the performance of semantic segmentation tasks by allowing the model to capture temporal dependencies and long-range contextual information. The R-UNet architecture consists of two main components: a recurrent encoder and a recurrent decoder. The recurrent encoder is composed of a series of convolutional and recurrent blocks, which extract features from the input image and propagate them across time. The recurrent decoder consists of a similar series of convolutional and recurrent blocks, which use the extracted features to generate the final segmentation mask. An attention mechanism is employed to enhance feature extraction at the bottleneck of the model. The proposed R-UNet architecture is evaluated on multiple benchmark datasets, including those for liver segmentation, brain tumor detection, mitochondria segmentation, lung imaging, a proprietary lung CT COVID-19 dataset, as well as various multi-organ imaging datasets. The experimental results demonstrate that the proposed R-UNet architecture outperforms the standard UNet architecture and several other state-of-the-art semantic segmentation models in terms of accuracy score, achieving an overall accuracy of 97.2 % on the Mitochondria dataset, 97.83 % on the Liver dataset, 89.17 % on the Tumor dataset and 97.22 % Lung dataset.
format Article
id doaj-art-9ffae33fb5a846adb609521cf52982dd
institution DOAJ
issn 2773-1863
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series Franklin Open
spelling doaj-art-9ffae33fb5a846adb609521cf52982dd2025-08-20T02:52:17ZengElsevierFranklin Open2773-18632024-12-01910018210.1016/j.fraope.2024.100182Temporal dependency modeling for improved medical image segmentation: The R-UNet perspectiveAhmed Alweshah0Roohollah Barzamini1Farshid Hajati2Shoorangiz Shams Shamsabad Farahani3Mohammad Arabian4Behnaz Sohani5Department of Electrical Engineering, Islamic Azad University Science and Research Branch, Tehran, IranDepartment of Electrical Engineering, Central Tehran Branch, Islamic Azad University, Tehran 1148963537, IranSchool of Science and Technology, Faculty of Science, Agriculture, Business and Law, University of New England, Armidale, NSW 2350, AustraliaDepartment of Electrical Engineering, Islamshahr branch, Islamic Azad University, Islamshahr, IranDepartment of Electrical Engineering, Central Tehran Branch, Islamic Azad University, Tehran 1148963537, IranWolfson School of Mechanical, Electrical & Manufacturing Engineering, Loughborough University, Loughborough, Leicestershire, LE11 3TU, UK; Corresponding author.In this study, we propose a modified version of the widely used UNet architecture, enhanced by the integration of recurrent blocks at each step of the encoder (down-sampling) and decoder (up-sampling) stages. The proposed Recurrent UNet (R-UNet) architecture aims to improve the performance of semantic segmentation tasks by allowing the model to capture temporal dependencies and long-range contextual information. The R-UNet architecture consists of two main components: a recurrent encoder and a recurrent decoder. The recurrent encoder is composed of a series of convolutional and recurrent blocks, which extract features from the input image and propagate them across time. The recurrent decoder consists of a similar series of convolutional and recurrent blocks, which use the extracted features to generate the final segmentation mask. An attention mechanism is employed to enhance feature extraction at the bottleneck of the model. The proposed R-UNet architecture is evaluated on multiple benchmark datasets, including those for liver segmentation, brain tumor detection, mitochondria segmentation, lung imaging, a proprietary lung CT COVID-19 dataset, as well as various multi-organ imaging datasets. The experimental results demonstrate that the proposed R-UNet architecture outperforms the standard UNet architecture and several other state-of-the-art semantic segmentation models in terms of accuracy score, achieving an overall accuracy of 97.2 % on the Mitochondria dataset, 97.83 % on the Liver dataset, 89.17 % on the Tumor dataset and 97.22 % Lung dataset.http://www.sciencedirect.com/science/article/pii/S2773186324001129Machine learningMedical image processingR-UNetDeep learningRecurrent neural networkCNN
spellingShingle Ahmed Alweshah
Roohollah Barzamini
Farshid Hajati
Shoorangiz Shams Shamsabad Farahani
Mohammad Arabian
Behnaz Sohani
Temporal dependency modeling for improved medical image segmentation: The R-UNet perspective
Franklin Open
Machine learning
Medical image processing
R-UNet
Deep learning
Recurrent neural network
CNN
title Temporal dependency modeling for improved medical image segmentation: The R-UNet perspective
title_full Temporal dependency modeling for improved medical image segmentation: The R-UNet perspective
title_fullStr Temporal dependency modeling for improved medical image segmentation: The R-UNet perspective
title_full_unstemmed Temporal dependency modeling for improved medical image segmentation: The R-UNet perspective
title_short Temporal dependency modeling for improved medical image segmentation: The R-UNet perspective
title_sort temporal dependency modeling for improved medical image segmentation the r unet perspective
topic Machine learning
Medical image processing
R-UNet
Deep learning
Recurrent neural network
CNN
url http://www.sciencedirect.com/science/article/pii/S2773186324001129
work_keys_str_mv AT ahmedalweshah temporaldependencymodelingforimprovedmedicalimagesegmentationtherunetperspective
AT roohollahbarzamini temporaldependencymodelingforimprovedmedicalimagesegmentationtherunetperspective
AT farshidhajati temporaldependencymodelingforimprovedmedicalimagesegmentationtherunetperspective
AT shoorangizshamsshamsabadfarahani temporaldependencymodelingforimprovedmedicalimagesegmentationtherunetperspective
AT mohammadarabian temporaldependencymodelingforimprovedmedicalimagesegmentationtherunetperspective
AT behnazsohani temporaldependencymodelingforimprovedmedicalimagesegmentationtherunetperspective