Standard

Multi-Class Brain Tumor Segmentation via 3d and 2d Neural Networks. / Pnev, Sergey; Groza, Vladimir; Tuchinov, Bair et al.

ISBI 2022 - Proceedings: 2022 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2022. (Proceedings - International Symposium on Biomedical Imaging; Vol. 2022-March).

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Harvard

Pnev, S, Groza, V, Tuchinov, B, Amelina, E, Pavlovskiy, E, Tolstokulakov, N, Amelin, M, Golushko, S & Letyagin, A 2022, Multi-Class Brain Tumor Segmentation via 3d and 2d Neural Networks. in ISBI 2022 - Proceedings: 2022 IEEE International Symposium on Biomedical Imaging. Proceedings - International Symposium on Biomedical Imaging, vol. 2022-March, IEEE Computer Society, 19th IEEE International Symposium on Biomedical Imaging, ISBI 2022, Kolkata, India, 28.03.2022. https://doi.org/10.1109/ISBI52829.2022.9761538

APA

Pnev, S., Groza, V., Tuchinov, B., Amelina, E., Pavlovskiy, E., Tolstokulakov, N., Amelin, M., Golushko, S., & Letyagin, A. (2022). Multi-Class Brain Tumor Segmentation via 3d and 2d Neural Networks. In ISBI 2022 - Proceedings: 2022 IEEE International Symposium on Biomedical Imaging (Proceedings - International Symposium on Biomedical Imaging; Vol. 2022-March). IEEE Computer Society. https://doi.org/10.1109/ISBI52829.2022.9761538

Vancouver

Pnev S, Groza V, Tuchinov B, Amelina E, Pavlovskiy E, Tolstokulakov N et al. Multi-Class Brain Tumor Segmentation via 3d and 2d Neural Networks. In ISBI 2022 - Proceedings: 2022 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society. 2022. (Proceedings - International Symposium on Biomedical Imaging). doi: 10.1109/ISBI52829.2022.9761538

Author

Pnev, Sergey ; Groza, Vladimir ; Tuchinov, Bair et al. / Multi-Class Brain Tumor Segmentation via 3d and 2d Neural Networks. ISBI 2022 - Proceedings: 2022 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2022. (Proceedings - International Symposium on Biomedical Imaging).

BibTeX

@inproceedings{7505299554a746c895eaf0a6ad6a2d24,
title = "Multi-Class Brain Tumor Segmentation via 3d and 2d Neural Networks",
abstract = "Brain tumor segmentation is an important and time-consuming part of the usual clinical diagnosis process. Multi-class segmentation of different tumor types is a challenging task, due to the differences in shape, size, location and scanner parameters. Many 2D and 3D convolution neural network architectures have been proposed to address this problem achieving a significant success. It is well known that 2D approach is generally faster and more popular in the most of such problems. However, the usage of 3D models allows us to simultaneously improve the quality of segmentation. Accounting the context along the sagittal plane leads to the learning of 3-dimensional features that we used for computationally expensive 3D operations what in its turn increases the learning time as well as decreases the speed of operation.In this paper, we compare the 2D and 3D approaches on 2 datasets with MRI images: the one from the BraTS 2020 competition and a private Siberian Brain tumor dataset. In each dataset, any single scan is represented by 4 sequences T1, T1C, T2 and T2-Flair, annotated by two certified neuro-radiologist specialists. The datasets differ from each other in the dimension, grade set and tumor type. Numerical comparison was performed based on the Dice score index. We provide the case by case analysis for the samples that caused most difficulties for the models. The results obtained in our work demonstrate the significant over performing of 3D methods keeping robustness in a regard of data source and type that allow us to get a little closer to AI-assisted diagnosis.",
keywords = "Brain, Deep Learning, Medical Imaging, MRI, Neural Network, Segmentation",
author = "Sergey Pnev and Vladimir Groza and Bair Tuchinov and Evgeniya Amelina and Evgeniy Pavlovskiy and Nikolay Tolstokulakov and Mihail Amelin and Sergey Golushko and Andrey Letyagin",
note = "Funding Information: The reported study was funded by RFBR according to the research project No 19-29-01103. Publisher Copyright: {\textcopyright} 2022 IEEE.; 19th IEEE International Symposium on Biomedical Imaging, ISBI 2022 ; Conference date: 28-03-2022 Through 31-03-2022",
year = "2022",
doi = "10.1109/ISBI52829.2022.9761538",
language = "English",
isbn = "9781665429238",
series = "Proceedings - International Symposium on Biomedical Imaging",
publisher = "IEEE Computer Society",
booktitle = "ISBI 2022 - Proceedings",
address = "United States",

}

RIS

TY - GEN

T1 - Multi-Class Brain Tumor Segmentation via 3d and 2d Neural Networks

AU - Pnev, Sergey

AU - Groza, Vladimir

AU - Tuchinov, Bair

AU - Amelina, Evgeniya

AU - Pavlovskiy, Evgeniy

AU - Tolstokulakov, Nikolay

AU - Amelin, Mihail

AU - Golushko, Sergey

AU - Letyagin, Andrey

N1 - Funding Information: The reported study was funded by RFBR according to the research project No 19-29-01103. Publisher Copyright: © 2022 IEEE.

PY - 2022

Y1 - 2022

N2 - Brain tumor segmentation is an important and time-consuming part of the usual clinical diagnosis process. Multi-class segmentation of different tumor types is a challenging task, due to the differences in shape, size, location and scanner parameters. Many 2D and 3D convolution neural network architectures have been proposed to address this problem achieving a significant success. It is well known that 2D approach is generally faster and more popular in the most of such problems. However, the usage of 3D models allows us to simultaneously improve the quality of segmentation. Accounting the context along the sagittal plane leads to the learning of 3-dimensional features that we used for computationally expensive 3D operations what in its turn increases the learning time as well as decreases the speed of operation.In this paper, we compare the 2D and 3D approaches on 2 datasets with MRI images: the one from the BraTS 2020 competition and a private Siberian Brain tumor dataset. In each dataset, any single scan is represented by 4 sequences T1, T1C, T2 and T2-Flair, annotated by two certified neuro-radiologist specialists. The datasets differ from each other in the dimension, grade set and tumor type. Numerical comparison was performed based on the Dice score index. We provide the case by case analysis for the samples that caused most difficulties for the models. The results obtained in our work demonstrate the significant over performing of 3D methods keeping robustness in a regard of data source and type that allow us to get a little closer to AI-assisted diagnosis.

AB - Brain tumor segmentation is an important and time-consuming part of the usual clinical diagnosis process. Multi-class segmentation of different tumor types is a challenging task, due to the differences in shape, size, location and scanner parameters. Many 2D and 3D convolution neural network architectures have been proposed to address this problem achieving a significant success. It is well known that 2D approach is generally faster and more popular in the most of such problems. However, the usage of 3D models allows us to simultaneously improve the quality of segmentation. Accounting the context along the sagittal plane leads to the learning of 3-dimensional features that we used for computationally expensive 3D operations what in its turn increases the learning time as well as decreases the speed of operation.In this paper, we compare the 2D and 3D approaches on 2 datasets with MRI images: the one from the BraTS 2020 competition and a private Siberian Brain tumor dataset. In each dataset, any single scan is represented by 4 sequences T1, T1C, T2 and T2-Flair, annotated by two certified neuro-radiologist specialists. The datasets differ from each other in the dimension, grade set and tumor type. Numerical comparison was performed based on the Dice score index. We provide the case by case analysis for the samples that caused most difficulties for the models. The results obtained in our work demonstrate the significant over performing of 3D methods keeping robustness in a regard of data source and type that allow us to get a little closer to AI-assisted diagnosis.

KW - Brain

KW - Deep Learning

KW - Medical Imaging

KW - MRI

KW - Neural Network

KW - Segmentation

UR - http://www.scopus.com/inward/record.url?scp=85129597274&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/aff9af67-94a5-36ac-b418-726fca5fd8ff/

U2 - 10.1109/ISBI52829.2022.9761538

DO - 10.1109/ISBI52829.2022.9761538

M3 - Conference contribution

AN - SCOPUS:85129597274

SN - 9781665429238

T3 - Proceedings - International Symposium on Biomedical Imaging

BT - ISBI 2022 - Proceedings

PB - IEEE Computer Society

T2 - 19th IEEE International Symposium on Biomedical Imaging, ISBI 2022

Y2 - 28 March 2022 through 31 March 2022

ER -

ID: 36075783