TY - JOUR

T1 - Acute ischemic stroke lesion segmentation in non-contrast CT images using 3D convolutional neural networks

AU - Dobshik, A. V.

AU - Verbitskiy, S. K.

AU - Pestunov, I. A.

AU - Sherman, K. M.

AU - Sinyavskiy, Yu N.

AU - Tulupov, A. A.

AU - Berikov, V. B.

N1 - The work was partly supported by RFBR grant No. 19-29-01175, and by the State Contract of the Sobolev Institute of Mathematics, Project No. FWNF-2022-0015. Публикация для корректировки.

PY - 2023/9

Y1 - 2023/9

N2 - In this paper, an automatic algorithm aimed at volumetric segmentation of acute ischemic stroke lesion in non-contrast computed tomography brain 3D images is proposed. Our deep-learning approach is based on the popular 3D U-Net convolutional neural network architecture, which was modified by adding the squeeze-and-excitation blocks and residual connections. Robust pre-processing methods were implemented to improve the segmentation accuracy. Moreover, a special patches sampling strategy was used to address the large size of medical images and class imbalance and to stabilize neural network training. All experiments were performed using five-fold cross-validation on the dataset containing non-contrast computed tomography volumetric brain scans of 81 patients diagnosed with acute ischemic stroke. Two radiology experts manually segmented images independently and then verified the labeling results for inconsistencies. The quantitative results of the proposed algorithm and obtained segmentation were measured by the Dice similarity coefficient, sensitivity, specificity and precision metrics. The suggested pipeline provides a Dice improvement of 12.0 %, sensitivity of 10.2 % and precision 10.0 % over the baseline and achieves an average Dice of 62.8 ± 3.3 %, sensitivity of 69.9 ± 3.9 %, specificity of 99.7 ± 0.2 % and precision of 61.9 ± 3.6 %, showing promising segmentation results.

AB - In this paper, an automatic algorithm aimed at volumetric segmentation of acute ischemic stroke lesion in non-contrast computed tomography brain 3D images is proposed. Our deep-learning approach is based on the popular 3D U-Net convolutional neural network architecture, which was modified by adding the squeeze-and-excitation blocks and residual connections. Robust pre-processing methods were implemented to improve the segmentation accuracy. Moreover, a special patches sampling strategy was used to address the large size of medical images and class imbalance and to stabilize neural network training. All experiments were performed using five-fold cross-validation on the dataset containing non-contrast computed tomography volumetric brain scans of 81 patients diagnosed with acute ischemic stroke. Two radiology experts manually segmented images independently and then verified the labeling results for inconsistencies. The quantitative results of the proposed algorithm and obtained segmentation were measured by the Dice similarity coefficient, sensitivity, specificity and precision metrics. The suggested pipeline provides a Dice improvement of 12.0 %, sensitivity of 10.2 % and precision 10.0 % over the baseline and achieves an average Dice of 62.8 ± 3.3 %, sensitivity of 69.9 ± 3.9 %, specificity of 99.7 ± 0.2 % and precision of 61.9 ± 3.6 %, showing promising segmentation results.

KW - 3D U-Net

KW - CNN

KW - brain

KW - ischemic stroke

KW - non-contrast CT

KW - segmentation

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85175071232&origin=inward&txGid=dcae91fb1f67e6f88af7c7ebc2698e12

UR - https://www.mendeley.com/catalogue/b3df77a3-c642-30b0-ab73-ff4fed56d161/

U2 - 10.18287/2412-6179-CO-1233

DO - 10.18287/2412-6179-CO-1233

M3 - Article

VL - 47

SP - 770

EP - 777

JO - Computer Optics

JF - Computer Optics

SN - 0134-2452

IS - 5

ER -