Compressive sensing MRI using dual tree complex wavelet transform with wavelet tree sparsity

Ragab, Mohamed; Omer, Osama A.; Hussien, Hany S.

doi:10.1109/nrsc.2017.7893519

Cited by 4 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The images reconstructed with this approach have the maximum level of sharpness and edge preservation with highest artifacts removal being emphasized but the improvements are achieved at the expense of the central processing unit (CPU) time. Dual tree complex wavelet transform using wavelet tree sparsity was explored in [14]. The approach showed better enhancement when the two algorithms were examined separately but even at 20% sampling ratio, the achieved SNR of 18dB is still low and with insignificant visual enhancement.…”

Section: ░ 1 Introductionmentioning

confidence: 99%

Improved Magnetic Resonance Image Reconstruction using Compressed Sensing and Adaptive Multi Extreme Particle Swarm Optimization Algorithm

Nalumansi,

Mwangi,

Kamucha

2024

IJEER

View full text Add to dashboard Cite

One powerful technique that can offer a thorough examination of the body's internal structure is magnetic resonance imaging (MRI). MRI's lengthy acquisition times, however, may restrict its clinical usefulness, particularly in situations where time is of the essence. Compressed sensing (CS) has emerged as a potentially useful method for cutting down on MRI acquisition times; nevertheless, the effectiveness of CS-MRI is dependent on the selection of the sparsity-promoting algorithm and sampling scheme. This research paper presents a novel method based on adaptive multi-extreme particle swarm optimization (AMEPSO) and dual tree complex wavelet transform (DTCWT) for fast image acquisition in magnetic resonance. The method uses AMEPSO in order to maximize the sampling pattern and minimize reconstruction error, while also exploiting the sparsity of MR images in the DTCWT domain to improve directional selectivity and shift invariance. MATLAB software was used for simulation of the proposed method. In comparison with the particle swarm optimized-DTCWT (PSODTCWT) and DTCWT algorithms, respectively, the results demonstrated an improvement in the peak signal-to-noise ratio of 8.92% and 15.92% and a higher structural similarity index measure of 3.69% and 7.5%. Based on these improvements, the proposed method could potentially make high-quality, real-time MRI imaging possible, which might improve detection and treatment of medical conditions and increase the throughput of MRI machines.

show abstract

Section: ░ 1 Introductionmentioning

confidence: 99%

Improved Magnetic Resonance Image Reconstruction using Compressed Sensing and Adaptive Multi Extreme Particle Swarm Optimization Algorithm

Nalumansi,

Mwangi,

Kamucha

2024

IJEER

View full text Add to dashboard Cite

show abstract

“…Three recovery strategies orthogonal matching pursuit (OMP), MP, and basis pursuit are investigated in [11] for CS‐based MRI image reconstruction. A compression algorithm that uses CS with a dual‐tree complex WT for MRI images is proposed in [12]. A 3D multichannel data approach to reconstruct MRI image using CS is suggested in [13], the main advantage of the parallel multichannel reconstruction is the gain in processing time.…”

Section: Introductionmentioning

confidence: 99%