Standard

Modified U-net with Different Attention Mechanisms for Acute Ischemic Stroke Segmentation using Non-Contrast CT. / Pnev, Sergey; Tulupov, Andrey; Berikov, Vladimir.

Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 133-136 9454963 (Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review

Harvard

Pnev, S, Tulupov, A & Berikov, V 2021, Modified U-net with Different Attention Mechanisms for Acute Ischemic Stroke Segmentation using Non-Contrast CT. in Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021., 9454963, Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021, Institute of Electrical and Electronics Engineers Inc., pp. 133-136, 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021, Yekaterinburg, Russian Federation, 13.05.2021. https://doi.org/10.1109/USBEREIT51232.2021.9454963

APA

Pnev, S., Tulupov, A., & Berikov, V. (2021). Modified U-net with Different Attention Mechanisms for Acute Ischemic Stroke Segmentation using Non-Contrast CT. In Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021 (pp. 133-136). [9454963] (Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/USBEREIT51232.2021.9454963

Vancouver

Pnev S, Tulupov A, Berikov V. Modified U-net with Different Attention Mechanisms for Acute Ischemic Stroke Segmentation using Non-Contrast CT. In Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 133-136. 9454963. (Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021). doi: 10.1109/USBEREIT51232.2021.9454963

Author

Pnev, Sergey ; Tulupov, Andrey ; Berikov, Vladimir. / Modified U-net with Different Attention Mechanisms for Acute Ischemic Stroke Segmentation using Non-Contrast CT. Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 133-136 (Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021).

BibTeX

@inproceedings{2c53effa990443809f9100444f1f4956,
title = "Modified U-net with Different Attention Mechanisms for Acute Ischemic Stroke Segmentation using Non-Contrast CT",
abstract = "The instant diagnosis of acute ischemic stroke using non-contrast computed tomography brain scans is important for right decision upon a treatment. Artificial intelligence and deep learning tools can assist a radiology specialist in analysis and interpretation of CT images. This work aims at improving U-net model and testing it on real non-contrast CT images of acute ischemic stroke. In this work, we use the following attention modules to learn a better feature representation and for more accurate segmentation: Convolutional Block Attention Module on skip-connection stage, double attention gates on decoding stage, and Feature Pyramid Attention as bottleneck. Experiments were conducted using a combination of the Binary Cross-Entropy Loss and Dice Loss as the loss function, and separately with the Focal Tversky Loss. An anonymized sample of 500 patients with ischemic stroke was obtained from International Tomography Center SB RAS. After verification, 25 patients were used in our study. The application of the considered architecture in 2D ischemic stroke segmentation was quite successful.",
keywords = "acute stroke, attention, CBAM, deep neural network, texture segmentation, U-net",
author = "Sergey Pnev and Andrey Tulupov and Vladimir Berikov",
note = "Funding Information: The work was partly supported by RFBR grant 19-29-01175. The study was carried out within the framework of the state contract of Sobolev Institute of Mathematics (project no 0314-2019-0015). Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021 ; Conference date: 13-05-2021 Through 14-05-2021",
year = "2021",
month = may,
day = "13",
doi = "10.1109/USBEREIT51232.2021.9454963",
language = "English",
series = "Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "133--136",
booktitle = "Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021",
address = "United States",

}

RIS

TY - GEN

T1 - Modified U-net with Different Attention Mechanisms for Acute Ischemic Stroke Segmentation using Non-Contrast CT

AU - Pnev, Sergey

AU - Tulupov, Andrey

AU - Berikov, Vladimir

N1 - Funding Information: The work was partly supported by RFBR grant 19-29-01175. The study was carried out within the framework of the state contract of Sobolev Institute of Mathematics (project no 0314-2019-0015). Publisher Copyright: © 2021 IEEE.

PY - 2021/5/13

Y1 - 2021/5/13

N2 - The instant diagnosis of acute ischemic stroke using non-contrast computed tomography brain scans is important for right decision upon a treatment. Artificial intelligence and deep learning tools can assist a radiology specialist in analysis and interpretation of CT images. This work aims at improving U-net model and testing it on real non-contrast CT images of acute ischemic stroke. In this work, we use the following attention modules to learn a better feature representation and for more accurate segmentation: Convolutional Block Attention Module on skip-connection stage, double attention gates on decoding stage, and Feature Pyramid Attention as bottleneck. Experiments were conducted using a combination of the Binary Cross-Entropy Loss and Dice Loss as the loss function, and separately with the Focal Tversky Loss. An anonymized sample of 500 patients with ischemic stroke was obtained from International Tomography Center SB RAS. After verification, 25 patients were used in our study. The application of the considered architecture in 2D ischemic stroke segmentation was quite successful.

AB - The instant diagnosis of acute ischemic stroke using non-contrast computed tomography brain scans is important for right decision upon a treatment. Artificial intelligence and deep learning tools can assist a radiology specialist in analysis and interpretation of CT images. This work aims at improving U-net model and testing it on real non-contrast CT images of acute ischemic stroke. In this work, we use the following attention modules to learn a better feature representation and for more accurate segmentation: Convolutional Block Attention Module on skip-connection stage, double attention gates on decoding stage, and Feature Pyramid Attention as bottleneck. Experiments were conducted using a combination of the Binary Cross-Entropy Loss and Dice Loss as the loss function, and separately with the Focal Tversky Loss. An anonymized sample of 500 patients with ischemic stroke was obtained from International Tomography Center SB RAS. After verification, 25 patients were used in our study. The application of the considered architecture in 2D ischemic stroke segmentation was quite successful.

KW - acute stroke

KW - attention

KW - CBAM

KW - deep neural network

KW - texture segmentation

KW - U-net

UR - http://www.scopus.com/inward/record.url?scp=85113768061&partnerID=8YFLogxK

U2 - 10.1109/USBEREIT51232.2021.9454963

DO - 10.1109/USBEREIT51232.2021.9454963

M3 - Conference contribution

AN - SCOPUS:85113768061

T3 - Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021

SP - 133

EP - 136

BT - Proceedings - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2021 Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology, USBEREIT 2021

Y2 - 13 May 2021 through 14 May 2021

ER -

ID: 34125379