We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Usage of biorthogonal wavelet filtering algorithm in data processing of biomedical images.
- Authors
Chang, Xiaoyi; Li, Yuebin; Bai, Ting; Qu, Tianrong; Gao, Jungang; Zhao, Chao
- Abstract
Though bi-orthogonal filtering (BWF) algorithm has been applied in image processing, the image after processing still has the phenomenon of edge distortion. To improve the image processing effect of BWF algorithm, genetic algorithm was introduced to optimize BWF algorithm, and optimized biorthogonal filtering algorithm (OBWF) was established. Computed tomography (CT), magnetic resonance imaging (MRI), and digital subtraction angiography (DSA) were employed as biomedical imaging data processing objects. The results of OBWF algorithm and other algorithms were analyzed under the same conditions regarding the changes of the peak signal to noise ratio, structural similarity index (SSIM), normalized mean absolute deviation (NMAD), background contrast (C), and mean offset CII index. The results showed that the PSNR and SSIM of the biomedical image data processed by the OBWF algorithm were higher and the NMAD value was lower than those of other algorithms. The C of the biomedical image after processing by the OBWF algorithm was 0.2393 higher than that of the BWF, and the CII was 0.9862 higher than the discrete sequence wavelet transform (DSWT) algorithm. The processing times of OBWF algorithm for CT, MRI, and DSA biomedical image data were 3.029 s, 2.239 s, and 68.745 s, respectively, which were lower than the times of other algorithms (P < 0.001). In short, OBWF could improve the quality of biomedical image data and reduce the running time, which provided a reference for data processing of biomedical images.
- Subjects
IMAGE processing; MAGNETIC resonance imaging; DISCRETE wavelet transforms; SIGNAL-to-noise ratio; COMPUTED tomography; ELECTRONIC data processing; DIGITAL subtraction angiography; IMAGE compression
- Publication
Journal of Supercomputing, 2022, Vol 78, Issue 16, p17920
- ISSN
0920-8542
- Publication type
Article
- DOI
10.1007/s11227-022-04535-y