Medical image fusion based on nuclear norm minimization

Abstract

Medical image fusion plays an important role in diagnosis and treatment of diseases such as image-guided radiotherapy and surgery. Although numerous medical image fusion methods have been proposed, most approaches have not touched the low rank nature of matrix formed by medical image, which usually lead to fusion image distortion and image information loss. These methods also often lack universality when dealing with different kinds of medical images. In this article, we propose a novel medical image fusion to overcome aforementioned issues on existing methods with the aid of low rank matrix approximation with nuclear norm minimization (NNM) constraint. The workflow of our method is described as: firstly, nonlocal similar patches across the medical image are searched by block matching for local patch in source images. Second, a fused matrix is stacking by shared nonlocal similarity patches, then the low rank matrix approximation methods under nuclear norm minimization can be used to recover low rank feature of fused matrix. Finally, fused image can be gotten by aggregating all the fused patches. Experimental results show that the proposed method is superior to other methods in both subjectively visual performance and objective criteria.

Multimodal medical image fusion based on discrete Tchebichef moments and pulse coupled neural network

“…In the experiments, the proposed method is compared with following five fusion algorithms, including NSCT‐based multimodal medical image fusion using pulse‐coupled neural network and modified spatial frequency (NSCT‐PCNN‐SF; Das and Kundu, ), medical image fusion based on nuclear norm minimization (NNM; Liu et al, ), image fusion with guided filtering (GFF; Li et al, ), a general framework for image fusion based on multiscale transform and sparse representation (LP‐SR; Liu et al, ), image fusion algorithm based on spatial frequency motivated pulse coupled neural networks in nonsubsampled contourlet transform domain (NSCT‐SF‐PCNN; Qu et al, ). The size of the blocks is determined by experiments.…”

No reference quality evaluation of medical image fusion

“…Eight MIF algorithms are used to produce 160 fused images. MIF algorithms include discrete tchebichef moments and pulse coupled neural network (DTM‐PCNN), nuclear norm minimization (NNM), laplacian pyramid and sparse representation (LP‐SR), cross‐scale coefficient selection(CSCS), guided filtering (GFF), pulse coupled neural network with modified spatial frequency based on NSCT (NSCT‐PCNN‐SF), improved sum modified laplacian (ISML), convolutional sparse representation (CSR), where DTM‐PCNN, NNM, CSCS, NSCT‐PCNN‐SF, ISML and CSR are designed specifically for multimodal medical image, LP‐SR and GFF are general‐purpose image fusion algorithms which designed for multimodal medical image, multiexposure image, and multifocus image. In implementation, the default parameters are used.…”

IGM‐based perceptual multimodal medical image fusion using free energy motivated adaptive PCNN

“…In the experiments, the presented fusion strategy is compared with following five fusion algorithms, including multimodal medical image fusion based on NSCT and PCNN with modified SF (NSCT‐PCNN‐MSF), NNM based on nuclear norm minimization (NNM), GFF based on guided filtering (GFF), image fusion metric with Laplace transform and sparse representation (LP‐SR), CSR based on convolutional sparse representation (CSR). For fair comparison, we use the parameters that were reported by the authors to yield the best fusion results.…”

“…These pair images are divided into five groups. Group 1 (G1): CT and MRI-T1; Group 2 (G2): CT and MRI-T2; Group 3 (G3): MRI-T1 and MRI-T2; Group 4 (G4): MRI-T2 and PET; Group 5 (G5): MRI-T2 and SPECT.In the experiments, the presented fusion strategy is compared with following five fusion algorithms, including multimodal medical image fusion based on NSCT and PCNN with modified SF 13 (NSCT-PCNN-MSF), NNM based on nuclear norm minimization25 (NNM), GFF based on guided filtering 26 (GFF), image fusion metric with Laplace transform and sparse representation 10 (LP-SR), CSR F IGUR E 2 Schematic diagram of our proposed fusion strategy F IGUR E 3 Source multimodal medical images used in our experiments…”