Performance Evaluation of Advanced Image Fusion Algorithms in Gamma Knife Treatment Planning

Multisensor image fusion is a process of combining information from multiple sensors. A newly noninvasive neurosurgical method that requires image fusion is the Gamma Knife. In this paper we present and evaluate advanced image fusion algorithms for Matlab platform and DICOM images helping efficiently the innovative Gamma Knife treatment planning. This method is first time ever introduced in clinical practice in Hellas mostly for radiological brain neurosurgery. One of the most difficult tasks is to precisely fuse the CT and MR images so that the proper information from both images is included in one image in order to help the specialist produce the treatment planning of the patient. We present several level grayscale image fusion methods: average, pca, select maximum, select minimum, discrete wavelet transform and Laplacian, filter – subtract – decimate, ratio, contrast, gradient, morphological pyramid and a shift invariant discrete wavelet transform method in Matlab platform. We tested these methods qualitatively and quantitatively. The quantitative criteria we used are the Root Mean Square Error (RMSE), the Normalized Least – Square Error (NLSE), the Mutual Information (MI), the Standard Deviation (SD), the Entropy (E), the Difference Entropy (DE), the Cross Entropy (CE) and the Spatial Frequency (SF).