Reduction of Computational Complexity in Segmentation of Medical Images

Subject Areas :

Reza Ghafouri ¹ , ALIREZA ALESAADI ² , Babak Gholami ³

1 - Department of Biomedical Engineering, Kazerun Branch, Islamic Azad University, Kazerun, Iran
2 - 2Department of Electrical Engineering, Kazerun Branch, Islamic Azad University, Kazerun, Iran
3 - Department of Electrical Engineering, Kazerun Branch, Islamic Azad University, Kazerun, Iran

Received: 2024-11-19 Accepted : 2025-02-12 Published : 2025-02-13

Keywords: Segmentation, Medical Images, Level Sets, Binary Variance, Computational Complexity, Image Derivatives,

Abstract :

With the development of AI-based medical image segmentation methods, there is and increasing demand for using these methods in segmentation of tissues and organs in medical images [1]. Medical image segmentation plays a very important role in automatically examining the structure of organs, obtaining their dimensions and in some cases, examining their functionality. AI-based methods, along with advantages such as high accuracy and the ability to learn and improve their performance, can slow down the overall speed of image processing due to enormous number of calculations and the need for large and diverse data sets in the training process. This point is more noticeable in the simulation of parallel processing methods with digital computers because in best cases, only one artificial neuron can be updated in each processor clock pulse. In this article, we will use an analytical method for medical image segmentation and finally, we will increase the speed of image segmentation by using a novel method to decrease computational complexity of calculation of image derivatives. Comparing the speed of the algorithm with the proposed method in this study with the conventional method of segmentation shows an acceptable increase in processing speed and reaching the final answer without losing segmentation accuracy

References:

[1] A. Eltanboly, M. Ghazal, H. Hajjdiab, A. Shalaby, A. Switala, A. Mahmoud, P. Sahoo, M. El-Azab, A. El-Baz, “A level sets-based image segmentation approach using statistical shape priors”, Applied Mathematics and Computation, vol. 340, pp. 164–179, 2019.
[2] C. Huang, L. Zeng, “Level set evolution model for image segmentation based on variable exponent p-Laplace equation”, Applied Mathematical Modelling, vol. 40, pp. 7739–7750, 2016.
[3] A. Joshi, M. S. Khan, K. N. Choi, “Medical image segmentation using combined level set and saliency analysis”, IEEE Access, vol. 12, pp. 102016–102026, 2024.
[4] Z. Li, S. Tang, Y. Zeng, S. Chai, W. Ye, F. Yang, K. Huang, “A level-set method with a multiplicative–additive constraint model for image segmentation and bias correction”, Knowledge-Based Systems, vol. 297, 2024.
[5] S. Wali, C. Li, M. Imran, A. Shakoor, A. Basit, “Level-set evolution for medical image segmentation with alternating direction method of multipliers”, Signal Processing, vol. 211, 2023.
[6] S. Osher, J. A. Sethian, “Fronts propagating with curvature-dependent speed: Algorithms based on Hamilton-Jacobi formulations”, International Journal of Computational Physics, vol. 79, no. 1, pp. 12–49, 1988.
[7] A. Tsai, A. Yezzi, W. Wells, C. Tempany, A. Fan, W. A. Grimson, A. Wilsky, “A shape-based approach to the segmentation of medical imagery using level sets”, IEEE Transactions on Medical Imaging, vol. 22, no. 2, pp. 137–154, Feb. 2003.
[8] M. Leventon, E. Grimson, O. Faugeras, “Statistical shape influence in geodesic active contours”, in Proc. IEEE Computer Vision and Pattern Recognition, Hilton Head, SC, USA, pp. 316–322, Jun. 2000.
[9] T. Chan, L. Vese, “Active contours without edges”, IEEE Transactions on Image Processing, vol. 10, no. 2, pp. 266–277, Feb. 2001.
[10] A. Yezzi, A. Tsai, A. Willsky, “A statistical approach to snakes for bimodal and trimodal imagery”, in Proc. Int. Conf. Computer Vision, Corfu, Greece, vol. 2, pp. 898–903, 1999.