Brain Tumor Segmentation Meets Efficiency: Res-UNet Improved by Attention Mechanisms and Quantization

Aprianto, Kasiful; Riana, Dwiza

doi:10.21512/commit.v19i2.13177

Open access Journal SINTA 1

Brain Tumor Segmentation Meets Efficiency: Res-UNet Improved by Attention Mechanisms and Quantization

CommIT (Communication and Information Technology) Journal · Vol. 19 No. 2 (2025) · pp. 293–314

DOI: 10.21512/commit.v19i2.13177
ISSN: 2460-7010, 1979-2484
Published: 2025-10-13
Publisher: Bina Nusantara University
Language: eng
License: creativecommons.org/licenses/by-sa/4.0

Abstract

Brain tumor segmentation from Magnetic Resonance Imaging (MRI) images is a crucial step in medical diagnosis and treatment planning, which directly impacts clinical decision-making and patient outcomes, particularly in resource-constrained medical environments. However, achieving high segmentation accuracy while maintaining computational efficiency remains a challenge, particularly for complex tumor types. Therefore, the research aims to use the brain tumor segmentation dataset and the brain tumor MRI dataset from Kaggle to evaluate segmentation performance. The analysis also investigates the trade-off between model accuracy and efficiency by optimizing the Res-UNet architecture with attention mechanisms, including the Attention Gate (AG), Squeeze-and-Excitation (SE) Block, and the Convolutional Block Attention Module (CBAM). As the result, attention mechanisms improve feature representation and segmentation precision. Then, these procedures also add computational cost. To address this challenge, Dynamic Range Quantization (DRQ) compresses the model from 127 MB to 32 MB (75% reduction) and speeds up inference by 37% (0.3143 s to 0.1973 s). During the process, the best model, Res-UNet with AG, achieves a mean Intersection over Union (IoU) of 0.845 and drops only by less than 0.0004 after quantization. Unlike previous studies that explored attention or quantization in isolation, the researchers combine both to achieve accurate, efficient, and deployable brain tumor segmentation for resource-constrained settings.

Keywords

Collaboration network of the authors

15 authors · 29 co-authorship links · 2 across different organisations

Hover or tap a name to highlight its co-authors. Double-click to open the author's page.

Universitas Nusa Mandiri
Universiti Malaysia Pahang Al-Sultan Abdullah
University of Indonesia
Affiliation unknown

Different organisations
Same organisation
Affiliation unknown
Thicker line = more joint articles · bigger circle = more articles

Show as table

Co-author	Organisation	Joint articles
Muji Ernawati	Affiliation unknown	2
Mila Desi Anasanti	Universitas Nusa Mandiri	1
Wan Ahmad Gazali Kodri	Universitas Nusa Mandiri	1
Sri Hadianti	Universitas Nusa Mandiri	1
Ferda Ernawan	Universiti Malaysia Pahang Al-Sultan Abdullah	1
Muhammad Romadhona Kusuma	Affiliation unknown	1
Yudhiarma	Affiliation unknown	1
Muhamad Ropi	Affiliation unknown	1
Muhammad Romadhona Kusuma	Affiliation unknown	1
Achmad Nizar Hidayanto	University of Indonesia	1
Yudi Ramdhani	Affiliation unknown	1
Rizki Tri Prasetio	Affiliation unknown	1
Fahmi B. Marasabessy	Affiliation unknown	1

Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.

Metadata source: OAI-PMH (metadata jurnal) https://journal.binus.ac.id/index.php/commit/oai, first harvested 8 Nov 2026, last seen 8 Nov 2026.

Original metadata (oai_dc)

<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">Brain Tumor Segmentation Meets Efficiency: Res-UNet Improved by Attention Mechanisms and Quantization</dc:title>
	<dc:creator>Aprianto, Kasiful</dc:creator>
	<dc:creator>Riana, Dwiza</dc:creator>
	<dc:subject xml:lang="en-US">Brain Tumor Segmentation</dc:subject>
	<dc:subject xml:lang="en-US">Res-UNet</dc:subject>
	<dc:subject xml:lang="en-US">Attention Mechanisms</dc:subject>
	<dc:subject xml:lang="en-US">Quantization</dc:subject>
	<dc:description xml:lang="en-US">Brain tumor segmentation from Magnetic Resonance Imaging (MRI) images is a crucial step in medical diagnosis and treatment planning, which directly impacts clinical decision-making and patient outcomes, particularly in resource-constrained medical environments. However, achieving high segmentation accuracy while maintaining computational efficiency remains a challenge, particularly for complex tumor types. Therefore, the research aims to use the brain tumor segmentation dataset and the brain tumor MRI dataset from Kaggle to evaluate segmentation performance. The analysis also investigates the trade-off between model accuracy and efficiency by optimizing the Res-UNet architecture with attention mechanisms, including the Attention Gate (AG), Squeeze-and-Excitation (SE) Block, and the Convolutional Block Attention Module (CBAM). As the result, attention mechanisms improve feature representation and segmentation precision. Then, these procedures also add computational cost. To address this challenge, Dynamic Range Quantization (DRQ) compresses the model from 127 MB to 32 MB (75% reduction) and speeds up inference by 37% (0.3143 s to 0.1973 s). During the process, the best model, Res-UNet with AG, achieves a mean Intersection over Union (IoU) of 0.845 and drops only by less than 0.0004 after quantization. Unlike previous studies that explored attention or quantization in isolation, the researchers combine both to achieve accurate, efficient, and deployable brain tumor segmentation for resource-constrained settings.</dc:description>
	<dc:publisher xml:lang="en-US">Bina Nusantara University</dc:publisher>
	<dc:date>2025-10-13</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:type xml:lang="en-US">Peer-reviewed Article</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://journal.binus.ac.id/index.php/commit/article/view/13177</dc:identifier>
	<dc:identifier>10.21512/commit.v19i2.13177</dc:identifier>
	<dc:source xml:lang="en-US">CommIT (Communication and Information Technology) Journal; Vol. 19 No. 2 (2025): CommIT Journal ; 293-314</dc:source>
	<dc:source>2460-7010</dc:source>
	<dc:source>1979-2484</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://journal.binus.ac.id/index.php/commit/article/view/13177/5543</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2025 Kasiful Aprianto, Dwiza Riana</dc:rights>
	<dc:rights xml:lang="en-US">https://creativecommons.org/licenses/by-sa/4.0</dc:rights>
</oai_dc:dc>