Sparsity-precise iterative hard thresholding for medical image compressive sensing

Abstract

This paper firstly analyzes the characteristics of medical images, and then proposes a specialized compressive sensing algorithm called sparsity-precise iterative hard thresholding (SIHT), which is specifically designed to address their specific features such as low sparsity and low frequency. SIHT adaptively measures sparsity and step length which becomes more precise during the iteration process to achieve a certain quality improvement in medical image reconstruction. Experimental results demonstrate that as compared to other image compressive sensing (ICS) reconstruction algorithms across three different types of medical image datasets, SIHT can achieve the best subjective recovery quality particularly in terms of mitigating blocky artifacts and noise, where a notable improvement is obtained in terms of peak signal-to-noise ratio (PSNR) and structural similarity index measurement (SSIM) of medical ICS reconstruction.

Keywords

A

References

[1] Lu L, Xu W, Wang Yet al.. Recovery conditions of sparse signals using orthogonal least squares-type algorithms. IEEE transactions on signal processing, 2022, 70: 4727-4741 J]

[2] Song Y C, Wu F Y, Peng R. A neighborhood-based multiple orthogonal least square method for sparse signal recovery. Signal processing, 2023, 209: 109044 J]

[3] Lyu S, Jiang M, Su Cet al.. Novel phase difference extraction method of FPP system based on DWT and OMP algorithm. Optoelectronics letters, 2020, 16(2): 131-136 J]

[4] Xu J, Yang J, Kimishima Fet al.. Compressive sensing based image codec with partial pre-calculation. IEEE transactions on multimedia, 2023, 26: 4871-4883 J]

[5] Sadik S, Et-Tolba M, Nsiri B. An efficient OMP for multidimensional data denoising and recovery in compressed sensing. 47th International Conference on Telecommunications and Signal Processing (TSP), July 10–12, 2024, Prague, Czech Republic, 2024, New York, IEEE195200[C]

[6] Chen Y, Gao H, Wu Zet al.. An adaptive image sparse reconstruction method combined with nonlocal similarity and cosparsity for mixed Gaussian-Poisson noise removal. Optoelectronics letters, 2018, 14: 57-60 J]

[7] Dong N, Zhang L, Zhou Het al.. Two-stage fast matching pursuit algorithm for multi-target localization. IEEE access, 2023, 11: 66318-66326 J]

[8] Daubechies I, Defrise M, De Mol C. An iterative thresholding algorithm for linear inverse problems with a sparsity constraint. Communications on pure and applied mathematics, 2004, 57(11): 1413-1457 J]

[9] Blumensath T, Davies M E. Iterative hard thresholding for compressed sensing. Applied and computational harmonic analysis, 2009, 27(3): 265-274 J]

[10] Blanchard J D, Tanner J, Wei K. Conjugate gradient iterative hard thresholding: observed noise stability for compressed sensing. IEEE transactions on signal processing, 2014, 63(2): 528-537 J]

[11] Meng N, Zhao Y B. Newton-step-based hard thresholding algorithms for sparse signal recovery. IEEE transactions on signal processing, 2020, 68: 6594-6606 J]

[12] Zhou S, Pan L, Xiu N. Newton method for ℓ₀-regularized optimization. Numerical algorithms, 2021, 88: 1541-1570 J]

[13] Zhou S, Xiu N, Qi H D. Global and quadratic convergence of Newton hard-thresholding pursuit. Journal of machine learning research, 2021, 22(12): 1-45[J]

[14] Zhou S. Gradient projection Newton pursuit for sparsity constrained optimization. Applied and computational harmonic analysis, 2022, 61: 75-100 J]

[15] He Z, Shu Q, Wang Yet al.. A ReLU-based hardthresholding algorithm for non-negative sparse signal recovery. Signal processing, 2024, 215: 109260 J]

[16] Pan J, Yu H, Gao Zet al.. Iterative residual optimization network for limited-angle tomographic reconstruction. IEEE transactions on image processing, 2024, 33: 910-925 J]

[17] Wang X, Wang H, Zhang Y. Research on ultrasound tomography based on DWT-IRLS algorithm. Journal of optoelectronics·laser, 2023, 34(4): 441-448[J]

[18] Chowdhury M E H, Rahman T, Khandakar Aet al.. Can AI help in screening viral and COVID-19 pneumonia?. IEEE access, 2020, 8: 132665-132676 J]

[19] Van Den Heuvel T L, De Bruijn D, De Korte C Let al.. Automated measurement of fetal head circumference using 2D ultrasound images. PloS one, 2018, 13(8): e0200412 J]

[20] Menze B H, Jakab A, Bauer Set al.. The multimodal brain tumor image segmentation benchmark (BRATS). IEEE transactions on medical imaging, 2014, 34101993-2024 J]

RIGHTS & PERMISSIONS

Tianjin University of Technology