Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRI
<p dir="ltr">Ischemic stroke, caused by cerebral vessel occlusion, presents substantial challenges in medical imaging due to the variability and subtlety of stroke lesions. <u>Magnetic Resonance Imaging</u> (MRI) plays a crucial role in diagnosing and managing <u>is...
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| مؤلفون آخرون: | , , , , , |
| منشور في: |
2025
|
| الموضوعات: | |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| _version_ | 1864513532039528448 |
|---|---|
| author | Ashiqur Rahman (3969347) |
| author2 | Muhammad E.H. Chowdhury (17151154) Md Sharjis Ibne Wadud (22803611) Rusab Sarmun (17632269) Adam Mushtak (22048013) Sohaib Bassam Zoghoul (22803506) Israa Al-Hashimi (18131374) |
| author2_role | author author author author author author |
| author_facet | Ashiqur Rahman (3969347) Muhammad E.H. Chowdhury (17151154) Md Sharjis Ibne Wadud (22803611) Rusab Sarmun (17632269) Adam Mushtak (22048013) Sohaib Bassam Zoghoul (22803506) Israa Al-Hashimi (18131374) |
| author_role | author |
| dc.creator.none.fl_str_mv | Ashiqur Rahman (3969347) Muhammad E.H. Chowdhury (17151154) Md Sharjis Ibne Wadud (22803611) Rusab Sarmun (17632269) Adam Mushtak (22048013) Sohaib Bassam Zoghoul (22803506) Israa Al-Hashimi (18131374) |
| dc.date.none.fl_str_mv | 2025-02-15T12:00:00Z |
| dc.identifier.none.fl_str_mv | 10.1016/j.bspc.2025.107676 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/journal_contribution/Deep_learning-driven_segmentation_of_ischemic_stroke_lesions_using_multi-channel_MRI/30819599 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Biomedical and clinical sciences Neurosciences Engineering Biomedical engineering Health sciences Health services and systems Information and computing sciences Artificial intelligence Machine learning Ischemic stroke segmentation MRI modalities Deep learning SelfONN DWI |
| dc.title.none.fl_str_mv | Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRI |
| dc.type.none.fl_str_mv | Text Journal contribution info:eu-repo/semantics/publishedVersion text contribution to journal |
| description | <p dir="ltr">Ischemic stroke, caused by cerebral vessel occlusion, presents substantial challenges in medical imaging due to the variability and subtlety of stroke lesions. <u>Magnetic Resonance Imaging</u> (MRI) plays a crucial role in diagnosing and managing <u>ischemic stroke</u>, yet existing segmentation techniques often fail to accurately delineate lesions. This study introduces a novel deep learning-based method for segmenting ischemic stroke lesions using multi-channel MRI modalities, including Diffusion Weighted Imaging (DWI), Apparent Diffusion Coefficient (ADC), and enhanced Diffusion Weighted Imaging (eDWI). The proposed architecture integrates DenseNet121 as the encoder with Self-Organized Operational Neural Networks (SelfONN) in the decoder, enhanced by Channel and Space Compound Attention (CSCA) and Double Squeeze-and-Excitation (DSE) blocks. Additionally, a custom loss function combining Dice Loss and Jaccard Loss with weighted averages is introduced to improve model performance. Trained and evaluated on the ISLES 2022 dataset, the model achieved Dice Similarity Coefficients (DSC) of 83.88 % using <u>DWI </u>alone, 85.86 % with DWI and<u> ADC</u>, and 87.49 % with the integration of DWI, <u>ADC</u>, and eDWI. This approach not only outperforms existing methods but also addresses key limitations in current segmentation practices. These advancements significantly enhance diagnostic precision and treatment planning for ischemic stroke, providing valuable support for clinical decision-making.</p><h2>Other Information</h2><p dir="ltr">Published in: Biomedical Signal Processing and Control<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.bspc.2025.107676" target="_blank">https://dx.doi.org/10.1016/j.bspc.2025.107676</a></p> |
| eu_rights_str_mv | openAccess |
| id | Manara2_e91eb831640e0acb087a26083312a520 |
| identifier_str_mv | 10.1016/j.bspc.2025.107676 |
| network_acronym_str | Manara2 |
| network_name_str | Manara2 |
| oai_identifier_str | oai:figshare.com:article/30819599 |
| publishDate | 2025 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRIAshiqur Rahman (3969347)Muhammad E.H. Chowdhury (17151154)Md Sharjis Ibne Wadud (22803611)Rusab Sarmun (17632269)Adam Mushtak (22048013)Sohaib Bassam Zoghoul (22803506)Israa Al-Hashimi (18131374)Biomedical and clinical sciencesNeurosciencesEngineeringBiomedical engineeringHealth sciencesHealth services and systemsInformation and computing sciencesArtificial intelligenceMachine learningIschemic stroke segmentationMRI modalitiesDeep learningSelfONNDWI<p dir="ltr">Ischemic stroke, caused by cerebral vessel occlusion, presents substantial challenges in medical imaging due to the variability and subtlety of stroke lesions. <u>Magnetic Resonance Imaging</u> (MRI) plays a crucial role in diagnosing and managing <u>ischemic stroke</u>, yet existing segmentation techniques often fail to accurately delineate lesions. This study introduces a novel deep learning-based method for segmenting ischemic stroke lesions using multi-channel MRI modalities, including Diffusion Weighted Imaging (DWI), Apparent Diffusion Coefficient (ADC), and enhanced Diffusion Weighted Imaging (eDWI). The proposed architecture integrates DenseNet121 as the encoder with Self-Organized Operational Neural Networks (SelfONN) in the decoder, enhanced by Channel and Space Compound Attention (CSCA) and Double Squeeze-and-Excitation (DSE) blocks. Additionally, a custom loss function combining Dice Loss and Jaccard Loss with weighted averages is introduced to improve model performance. Trained and evaluated on the ISLES 2022 dataset, the model achieved Dice Similarity Coefficients (DSC) of 83.88 % using <u>DWI </u>alone, 85.86 % with DWI and<u> ADC</u>, and 87.49 % with the integration of DWI, <u>ADC</u>, and eDWI. This approach not only outperforms existing methods but also addresses key limitations in current segmentation practices. These advancements significantly enhance diagnostic precision and treatment planning for ischemic stroke, providing valuable support for clinical decision-making.</p><h2>Other Information</h2><p dir="ltr">Published in: Biomedical Signal Processing and Control<br>License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.bspc.2025.107676" target="_blank">https://dx.doi.org/10.1016/j.bspc.2025.107676</a></p>2025-02-15T12:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1016/j.bspc.2025.107676https://figshare.com/articles/journal_contribution/Deep_learning-driven_segmentation_of_ischemic_stroke_lesions_using_multi-channel_MRI/30819599CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/308195992025-02-15T12:00:00Z |
| spellingShingle | Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRI Ashiqur Rahman (3969347) Biomedical and clinical sciences Neurosciences Engineering Biomedical engineering Health sciences Health services and systems Information and computing sciences Artificial intelligence Machine learning Ischemic stroke segmentation MRI modalities Deep learning SelfONN DWI |
| status_str | publishedVersion |
| title | Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRI |
| title_full | Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRI |
| title_fullStr | Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRI |
| title_full_unstemmed | Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRI |
| title_short | Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRI |
| title_sort | Deep learning-driven segmentation of ischemic stroke lesions using multi-channel MRI |
| topic | Biomedical and clinical sciences Neurosciences Engineering Biomedical engineering Health sciences Health services and systems Information and computing sciences Artificial intelligence Machine learning Ischemic stroke segmentation MRI modalities Deep learning SelfONN DWI |