Deep-Learning-Based Segmentation of Extraocular Muscles from Magnetic Resonance Images
In this study, we investigated the performance of four deep learning frameworks of U-Net, U-NeXt, DeepLabV3+, and ConResNet in multi-class pixel-based segmentation of the extraocular muscles (EOMs) from coronal MRI. Performances of the four models were evaluated and compared with the standard F-meas...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Bioengineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5354/10/6/699 |
| _version_ | 1797596016068788224 |
|---|---|
| author | Amad Qureshi Seongjin Lim Soh Youn Suh Bassam Mutawak Parag V. Chitnis Joseph L. Demer Qi Wei |
| author_facet | Amad Qureshi Seongjin Lim Soh Youn Suh Bassam Mutawak Parag V. Chitnis Joseph L. Demer Qi Wei |
| author_sort | Amad Qureshi |
| collection | DOAJ |
| description | In this study, we investigated the performance of four deep learning frameworks of U-Net, U-NeXt, DeepLabV3+, and ConResNet in multi-class pixel-based segmentation of the extraocular muscles (EOMs) from coronal MRI. Performances of the four models were evaluated and compared with the standard F-measure-based metrics of intersection over union (IoU) and Dice, where the U-Net achieved the highest overall IoU and Dice scores of 0.77 and 0.85, respectively. Centroid distance offset between identified and ground truth EOM centroids was measured where U-Net and DeepLabV3+ achieved low offsets (<i>p</i> > 0.05) of 0.33 mm and 0.35 mm, respectively. Our results also demonstrated that segmentation accuracy varies in spatially different image planes. This study systematically compared factors that impact the variability of segmentation and morphometric accuracy of the deep learning models when applied to segmenting EOMs from MRI. |
| first_indexed | 2024-03-11T02:45:34Z |
| format | Article |
| id | doaj.art-3dfc72640dc940a2aa22decd0c87491a |
| institution | Directory Open Access Journal |
| issn | 2306-5354 |
| language | English |
| last_indexed | 2024-03-11T02:45:34Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Bioengineering |
| spelling | doaj.art-3dfc72640dc940a2aa22decd0c87491a2023-11-18T09:21:32ZengMDPI AGBioengineering2306-53542023-06-0110669910.3390/bioengineering10060699Deep-Learning-Based Segmentation of Extraocular Muscles from Magnetic Resonance ImagesAmad Qureshi0Seongjin Lim1Soh Youn Suh2Bassam Mutawak3Parag V. Chitnis4Joseph L. Demer5Qi Wei6Department of Bioengineering, George Mason University, Fairfax, VA 22030, USADepartment of Ophthalmology, Neurology and Bioengineering, Jules Stein Eye Institute, University of California, Los Angeles, CA 90095, USADepartment of Ophthalmology, Neurology and Bioengineering, Jules Stein Eye Institute, University of California, Los Angeles, CA 90095, USADepartment of Bioengineering, George Mason University, Fairfax, VA 22030, USADepartment of Bioengineering, George Mason University, Fairfax, VA 22030, USADepartment of Ophthalmology, Neurology and Bioengineering, Jules Stein Eye Institute, University of California, Los Angeles, CA 90095, USADepartment of Bioengineering, George Mason University, Fairfax, VA 22030, USAIn this study, we investigated the performance of four deep learning frameworks of U-Net, U-NeXt, DeepLabV3+, and ConResNet in multi-class pixel-based segmentation of the extraocular muscles (EOMs) from coronal MRI. Performances of the four models were evaluated and compared with the standard F-measure-based metrics of intersection over union (IoU) and Dice, where the U-Net achieved the highest overall IoU and Dice scores of 0.77 and 0.85, respectively. Centroid distance offset between identified and ground truth EOM centroids was measured where U-Net and DeepLabV3+ achieved low offsets (<i>p</i> > 0.05) of 0.33 mm and 0.35 mm, respectively. Our results also demonstrated that segmentation accuracy varies in spatially different image planes. This study systematically compared factors that impact the variability of segmentation and morphometric accuracy of the deep learning models when applied to segmenting EOMs from MRI.https://www.mdpi.com/2306-5354/10/6/699deep learningextraocular musclesegmentationMRIstrabismusophthalmology |
| spellingShingle | Amad Qureshi Seongjin Lim Soh Youn Suh Bassam Mutawak Parag V. Chitnis Joseph L. Demer Qi Wei Deep-Learning-Based Segmentation of Extraocular Muscles from Magnetic Resonance Images Bioengineering deep learning extraocular muscle segmentation MRI strabismus ophthalmology |
| title | Deep-Learning-Based Segmentation of Extraocular Muscles from Magnetic Resonance Images |
| title_full | Deep-Learning-Based Segmentation of Extraocular Muscles from Magnetic Resonance Images |
| title_fullStr | Deep-Learning-Based Segmentation of Extraocular Muscles from Magnetic Resonance Images |
| title_full_unstemmed | Deep-Learning-Based Segmentation of Extraocular Muscles from Magnetic Resonance Images |
| title_short | Deep-Learning-Based Segmentation of Extraocular Muscles from Magnetic Resonance Images |
| title_sort | deep learning based segmentation of extraocular muscles from magnetic resonance images |
| topic | deep learning extraocular muscle segmentation MRI strabismus ophthalmology |
| url | https://www.mdpi.com/2306-5354/10/6/699 |
| work_keys_str_mv | AT amadqureshi deeplearningbasedsegmentationofextraocularmusclesfrommagneticresonanceimages AT seongjinlim deeplearningbasedsegmentationofextraocularmusclesfrommagneticresonanceimages AT sohyounsuh deeplearningbasedsegmentationofextraocularmusclesfrommagneticresonanceimages AT bassammutawak deeplearningbasedsegmentationofextraocularmusclesfrommagneticresonanceimages AT paragvchitnis deeplearningbasedsegmentationofextraocularmusclesfrommagneticresonanceimages AT josephldemer deeplearningbasedsegmentationofextraocularmusclesfrommagneticresonanceimages AT qiwei deeplearningbasedsegmentationofextraocularmusclesfrommagneticresonanceimages |