SEGMENTATION OF SUBARACHNOID HEMORRHAGE ON BRAIN CT IMAGES USING U-NET AND ATTENTION U-NET: A COMPARATIVE ANALYSIS

Ilham Tristadika Saputra, Afu Ichsan Pradana, Dwi Hartanti

Abstract


Subarachnoid Hemorrhage (SAH) is a critical medical condition involving bleeding into the subarachnoid space, often caused by aneurysmal rupture or trauma. Prompt and accurate diagnosis is essential to avoid severe neurological damage. This study compares two deep learning architectures—U-Net and Attention U-Net—for automated segmentation of SAH on brain CT scan images. Both models were trained and tested using a publicly available CT dataset, and evaluated using Dice Score, Intersection over Union (IoU), Precision, Recall, and F1 Score. The Attention U-Net achieved a Dice Score of 0.896 and IoU of 0.877, outperforming the standard U-Net which scored 0.867 and 0.848 respectively. While U-Net yielded higher precision (0.726), it showed more conservative segmentation behavior and lower recall. Visual inspection of segmented masks confirmed that the Attention U-Net was better at capturing small or low-contrast hemorrhagic regions. These findings demonstrate the effectiveness of attention mechanisms in improving spatial sensitivity and segmentation quality, indicating its potential in real-time neuroimaging diagnostic systems.

References


C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J Big Data, vol. 6, no. 1, p. 60, Dec. 2019, doi: 10.1186/s40537-019-0197-0.

N. M. Dubosh and J. A. Edlow, “Diagnosis and Initial Emergency Department Management of Subarachnoid Hemorrhage,” Emerg Med Clin North Am, vol. 39, no. 1, pp. 87–99, Feb. 2021, doi: 10.1016/j.emc.2020.09.005.

A. Arab et al., “A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT,” Sci Rep, vol. 10, no. 1, p. 19389, Nov. 2020, doi: 10.1038/s41598-020-76459-7.

Y. Yu, M. Li, L. Liu, Y. Li, and J. Wang, “Clinical big data and deep learning: Applications, challenges, and future outlooks,” Big Data Mining and Analytics, vol. 2, no. 4, pp. 288–305, Dec. 2019, doi: 10.26599/BDMA.2019.9020007.

S. Umapathy, M. Murugappan, D. Bharathi, and M. Thakur, “Automated Computer-Aided Detection and Classification of Intracranial Hemorrhage Using Ensemble Deep Learning Techniques,” Diagnostics, vol. 13, no. 18, p. 2987, Sep. 2023, doi: 10.3390/diagnostics13182987.

M. D. Hssayeni, M. S. Croock, A. D. Salman, H. F. Al-khafaji, Z. A. Yahya, and B. Ghoraani, “Intracranial Hemorrhage Segmentation Using a Deep Convolutional Model,” Data (Basel), vol. 5, no. 1, p. 14, Feb. 2020, doi: 10.3390/data5010014.

G. Litjens et al., “A survey on deep learning in medical image analysis,” Med Image Anal, vol. 42, pp. 60–88, Dec. 2017, doi: 10.1016/j.media.2017.07.005.

O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” May 2015.

M. M. Khan et al., “A Deep Learning-Based Automatic Segmentation and 3D Visualization Technique for Intracranial Hemorrhage Detection Using Computed Tomography Images,” Diagnostics, vol. 13, no. 15, p. 2537, Jul. 2023, doi: 10.3390/diagnostics13152537.

L. Dai, Md Gapar Md Johar, and Mohammed Hazim Alkawaz, “Review of Semi-Supervised Medical Image Segmentation based on the U-Net,” Academic Journal of Science and Technology, vol. 11, no. 1, pp. 147–154, May 2024, doi: 10.54097/gmhkht38.

A. Vaswani et al., “Attention is All you Need,” in Advances in Neural Information Processing Systems, I. Guyon, U. Von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett, Eds., Curran Associates, Inc., 2017. [Online]. Available: https://proceedings.neurips.cc/paper_files/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf

O. Oktay et al., “Attention U-Net: Learning Where to Look for the Pancreas,” Apr. 2018.

F. El Abassi, A. Darouichi, and A. Ouaarab, “Comparative Analysis of U-Net with Transfer Learning and Attention Mechanism for Enhanced Medical Image Segmentation,” 2024, pp. 551–560. doi: 10.1007/978-3-031-68653-5_52.

X. L. Yang and Z. Y. Jin, “Att GGO-Net:A Semantic Segmentation Method of Lung CT Images with Self and Cross Attention Mechanism,” J Phys Conf Ser, vol. 2504, no. 1, p. 012017, May 2023, doi: 10.1088/1742-6596/2504/1/012017.

S. Tenny, J. Das, and W. Thorell, “Intracranial Hemorrhage - StatPearls - NCBI Bookshelf.” Accessed: May 26, 2025. [Online]. Available: https://www.ncbi.nlm.nih.gov/books/NBK470242/




DOI: https://doi.org/10.33387/jiko.v8i2.9958

Refbacks

  • There are currently no refbacks.