Automated Segmentation of Acute Ischemic Stroke Using Attention U-net with Patch Mechanism
This paper addresses ischemic stroke detection using deep learning techniques to interpret medical images like MRI and CT scans, with a focus on segmentation. Ischemic stroke occurs when a blockage in brain arteries disrupts blood flow, impairing brain functions. The study aims to develop a model...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Stefan cel Mare University of Suceava
2025-02-01
|
| Series: | Advances in Electrical and Computer Engineering |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.4316/AECE.2025.01004 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849762635761844224 |
|---|---|
| author | CINAR, N. UCAN, M. KAYA, B. KAYA, M. |
| author_facet | CINAR, N. UCAN, M. KAYA, B. KAYA, M. |
| author_sort | CINAR, N. |
| collection | DOAJ |
| description | This paper addresses ischemic stroke detection using deep learning techniques to interpret medical images
like MRI and CT scans, with a focus on segmentation. Ischemic stroke occurs when a blockage in brain
arteries disrupts blood flow, impairing brain functions. The study aims to develop a model for automatic
segmentation of ischemic stroke areas, facilitating efficient diagnosis in medical settings. An enhanced
Attention U-Net model with a patch-based approach using MRI data is proposed for this purpose. The model
was validated on the ISLES’22 public ischemic stroke dataset. The segmentation process consisted of three
stages. First, the standard Attention U-Net model achieved a Dice Similarity Coefficient (DSC) of 88.9%.
In the second stage, the MRI images were divided into 32x32 patches and reanalyzed, increasing the DSC
to 93%. In the final stage, different attention mechanism methods were added to the U-Net architecture
and the effect of attention mechanism on segmentation success was observed. As a result of the experiments,
the U-Net architecture using spatial attention achieved 94.86%, the U-Net architecture using SE attention
achieved 95.40%, and the U-Net architecture using CBAM attention achieved 96.47% DCS success. The study
concludes that the enhanced model outperforms existing methods, demonstrating that the proposed approach
is effective for segmenting ischemic strokes and yielding significant results compared to similar studies
in the literature. |
| format | Article |
| id | doaj-art-348fc745ecfb432da97009bd5bb9cc75 |
| institution | DOAJ |
| issn | 1582-7445 1844-7600 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Stefan cel Mare University of Suceava |
| record_format | Article |
| series | Advances in Electrical and Computer Engineering |
| spelling | doaj-art-348fc745ecfb432da97009bd5bb9cc752025-08-20T03:05:42ZengStefan cel Mare University of SuceavaAdvances in Electrical and Computer Engineering1582-74451844-76002025-02-01251294210.4316/AECE.2025.01004Automated Segmentation of Acute Ischemic Stroke Using Attention U-net with Patch MechanismCINAR, N.UCAN, M.KAYA, B.KAYA, M.This paper addresses ischemic stroke detection using deep learning techniques to interpret medical images like MRI and CT scans, with a focus on segmentation. Ischemic stroke occurs when a blockage in brain arteries disrupts blood flow, impairing brain functions. The study aims to develop a model for automatic segmentation of ischemic stroke areas, facilitating efficient diagnosis in medical settings. An enhanced Attention U-Net model with a patch-based approach using MRI data is proposed for this purpose. The model was validated on the ISLES’22 public ischemic stroke dataset. The segmentation process consisted of three stages. First, the standard Attention U-Net model achieved a Dice Similarity Coefficient (DSC) of 88.9%. In the second stage, the MRI images were divided into 32x32 patches and reanalyzed, increasing the DSC to 93%. In the final stage, different attention mechanism methods were added to the U-Net architecture and the effect of attention mechanism on segmentation success was observed. As a result of the experiments, the U-Net architecture using spatial attention achieved 94.86%, the U-Net architecture using SE attention achieved 95.40%, and the U-Net architecture using CBAM attention achieved 96.47% DCS success. The study concludes that the enhanced model outperforms existing methods, demonstrating that the proposed approach is effective for segmenting ischemic strokes and yielding significant results compared to similar studies in the literature.http://dx.doi.org/10.4316/AECE.2025.01004attention u-netbrain stroke segmentationdeep learningischemic strokeu-net |
| spellingShingle | CINAR, N. UCAN, M. KAYA, B. KAYA, M. Automated Segmentation of Acute Ischemic Stroke Using Attention U-net with Patch Mechanism Advances in Electrical and Computer Engineering attention u-net brain stroke segmentation deep learning ischemic stroke u-net |
| title | Automated Segmentation of Acute Ischemic Stroke Using Attention U-net with Patch Mechanism |
| title_full | Automated Segmentation of Acute Ischemic Stroke Using Attention U-net with Patch Mechanism |
| title_fullStr | Automated Segmentation of Acute Ischemic Stroke Using Attention U-net with Patch Mechanism |
| title_full_unstemmed | Automated Segmentation of Acute Ischemic Stroke Using Attention U-net with Patch Mechanism |
| title_short | Automated Segmentation of Acute Ischemic Stroke Using Attention U-net with Patch Mechanism |
| title_sort | automated segmentation of acute ischemic stroke using attention u net with patch mechanism |
| topic | attention u-net brain stroke segmentation deep learning ischemic stroke u-net |
| url | http://dx.doi.org/10.4316/AECE.2025.01004 |
| work_keys_str_mv | AT cinarn automatedsegmentationofacuteischemicstrokeusingattentionunetwithpatchmechanism AT ucanm automatedsegmentationofacuteischemicstrokeusingattentionunetwithpatchmechanism AT kayab automatedsegmentationofacuteischemicstrokeusingattentionunetwithpatchmechanism AT kayam automatedsegmentationofacuteischemicstrokeusingattentionunetwithpatchmechanism |