Adjusting U-Net for the aortic abdominal aneurysm CT segmentation case
In this paper, we address the issue of developing of a convolutional neural network for the problem of aneurysm segmentation into three classes and of exploring ways for improving the quality of final segmentation masks. As a result of our study, macro dice score for classes of interest reaches 83.1...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Samara National Research University
2024-06-01
|
| Series: | Компьютерная оптика |
| Subjects: | |
| Online Access: | https://www.computeroptics.ru/KO/Annot/KO48-3/480312.html |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832086544266035200 |
|---|---|
| author | R.U. Epifanov N.A. Nikitin A.A. Rabtsun L.N. Kurdyukov A.A. Karpenko R.I. Mullyadzhanov |
| author_facet | R.U. Epifanov N.A. Nikitin A.A. Rabtsun L.N. Kurdyukov A.A. Karpenko R.I. Mullyadzhanov |
| author_sort | R.U. Epifanov |
| collection | DOAJ |
| description | In this paper, we address the issue of developing of a convolutional neural network for the problem of aneurysm segmentation into three classes and of exploring ways for improving the quality of final segmentation masks. As a result of our study, macro dice score for classes of interest reaches 83.12% ± 4.27%. We explored different augmentation styles and showed the importance of applying intensity augmentation style to improve segmentation algorithm robustness in conditions of clinical data diversity. Augmentation with spatial and insensitive styles increase macro dice score up to 3%. The comparison of various inference mode indicate that combination of overlapping inference and segmentation window enlargement ameliorate macro dice up to 1.4%. Overall improvement of the quality of segmentation masks by macro dice score amounted up to 6% using combination of data-based augmentation style and advanced inference technique. |
| format | Article |
| id | doaj-art-de95137568cd4de5ae4a8373ba026123 |
| institution | Kabale University |
| issn | 0134-2452 2412-6179 |
| language | English |
| publishDate | 2024-06-01 |
| publisher | Samara National Research University |
| record_format | Article |
| series | Компьютерная оптика |
| spelling | doaj-art-de95137568cd4de5ae4a8373ba0261232025-02-06T12:21:37ZengSamara National Research UniversityКомпьютерная оптика0134-24522412-61792024-06-0148341842410.18287/2412-6179-CO-1338Adjusting U-Net for the aortic abdominal aneurysm CT segmentation caseR.U. Epifanov0N.A. Nikitin1A.A. Rabtsun2L.N. Kurdyukov3A.A. Karpenko4R.I. Mullyadzhanov5Novosibirsk State UniversityMeshalkin National Medical Research CenterNovosibirsk State Medical UniversityNovosibirsk State Medical UniversityMeshalkin National Medical Research CenterNovosibirsk State University; Institute of Thermophysics SB RASIn this paper, we address the issue of developing of a convolutional neural network for the problem of aneurysm segmentation into three classes and of exploring ways for improving the quality of final segmentation masks. As a result of our study, macro dice score for classes of interest reaches 83.12% ± 4.27%. We explored different augmentation styles and showed the importance of applying intensity augmentation style to improve segmentation algorithm robustness in conditions of clinical data diversity. Augmentation with spatial and insensitive styles increase macro dice score up to 3%. The comparison of various inference mode indicate that combination of overlapping inference and segmentation window enlargement ameliorate macro dice up to 1.4%. Overall improvement of the quality of segmentation masks by macro dice score amounted up to 6% using combination of data-based augmentation style and advanced inference technique.https://www.computeroptics.ru/KO/Annot/KO48-3/480312.htmlabdominal aortic aneurysmneural networksemantic segmentationcalcifications |
| spellingShingle | R.U. Epifanov N.A. Nikitin A.A. Rabtsun L.N. Kurdyukov A.A. Karpenko R.I. Mullyadzhanov Adjusting U-Net for the aortic abdominal aneurysm CT segmentation case Компьютерная оптика abdominal aortic aneurysm neural network semantic segmentation calcifications |
| title | Adjusting U-Net for the aortic abdominal aneurysm CT segmentation case |
| title_full | Adjusting U-Net for the aortic abdominal aneurysm CT segmentation case |
| title_fullStr | Adjusting U-Net for the aortic abdominal aneurysm CT segmentation case |
| title_full_unstemmed | Adjusting U-Net for the aortic abdominal aneurysm CT segmentation case |
| title_short | Adjusting U-Net for the aortic abdominal aneurysm CT segmentation case |
| title_sort | adjusting u net for the aortic abdominal aneurysm ct segmentation case |
| topic | abdominal aortic aneurysm neural network semantic segmentation calcifications |
| url | https://www.computeroptics.ru/KO/Annot/KO48-3/480312.html |
| work_keys_str_mv | AT ruepifanov adjustingunetfortheaorticabdominalaneurysmctsegmentationcase AT nanikitin adjustingunetfortheaorticabdominalaneurysmctsegmentationcase AT aarabtsun adjustingunetfortheaorticabdominalaneurysmctsegmentationcase AT lnkurdyukov adjustingunetfortheaorticabdominalaneurysmctsegmentationcase AT aakarpenko adjustingunetfortheaorticabdominalaneurysmctsegmentationcase AT rimullyadzhanov adjustingunetfortheaorticabdominalaneurysmctsegmentationcase |