A novel image fusion algorithm based on 2D scale-mixing complex wavelet transform and Bayesian MAP estimation for multimodal medical images

Abstract

In this paper, we propose a new image fusion algorithm based on two-dimensional Scale-Mixing Complex Wavelet Transform (2D-SMCWT). The fusion of the detail 2D-SMCWT coe±cients is performed via a Bayesian Maximum a Posteriori (MAP) approach by considering a trivariate statistical model for the local neighboring of 2D-SMCWT coe±cients. For the approximation coe±cients, a new fusion rule based on the Principal Component Analysis (PCA) is applied. We conduct several experiments using three di®erent groups of multimodal medical images to evaluate the performance of the proposed method. The obtained results prove the superiority of the proposed method over the state of the art fusion methods in terms of visual quality and several commonly used metrics. Robustness of the proposed method is further tested against di®erent types of noise. The plots of fusion metrics establish the accuracy of the proposed fusion method.

Application of Image Fusion in Diagnosis and Treatment of Liver Cancer

“…DWT has been widely used in CT/MRI and MRI/PET medical image fusion [61][62][63][64][65][66][67][68]. However, DWT is known to be sensitive to the translation/shift of input signals, and therefore, translation among signals may exert a negative impact on effectiveness.…”

A Survey on Multi-Scale Medical images Fusion Techniques: Brain Diseases

“…Radiologists in particular are required to read more and more cases with more and more images per case [28][29][30][31][32][33][34][35]. Shortages in radiologists especially for example specialists in rural and medically underserved areas compound the problem.…”

“…In general ,brain diseases diagnose/treat considered hot topic today due to relevant with human live and body healthy and also it is main sensitive organ as it can control whole body, so the researcher put high effort in the literature review of computer vision and medical image processing like ( analysis, fusion, segmentation, classification, enhancements ).in the fusion process, some of them focus on how to maximize the single image fused 'informative details [33,34],the other start looking for suppress the artifacts like noise [35,36,37] ,blur [38,39,40].The rest emphases on the how to preserved edge ,boundary, smoothly, sharpness [41,42,43,44].from my point ,brain disease diagnosing/treatment is critical work in the computer vision and cannot bear any Proportion of error in diagnosing diseases and that consider as lack and vary of fused medical image requirements in one time like (high visually, free noise, more informative, high contrast, edge preserved, accurate anatomy of tissues) in the previous work of medical image fusion algorithms for the application of brain diseases diagnosing and treatments [40][41][42][43][44] . In summary, the treatment of the diagnostic image data is performed by a physicist, who analyzes and aggregates them according to his knowledge.…”