Hybrid Surface- and Voxel-Based Registration for MR-PET Brain Fusion

Abstract

In this paper, we propose a novel technique of registration using hybrid approach for MR-PET brain image fusion. Hybrid approach uses merits of surface-and voxel-based registration. Thus, our method measures similarities using voxel intensities in MR images corresponding to the feature points of the brain in PET images. Proposed method selects the brain threshold using histogram accumulation ratio in PET images. And then, we automatically segment the brain using the inverse region growing with pre-calculated threshold and extract the feature points of the brain using sharpening filter in PET images. In order to find the optimal location for registration, we evaluate the Hybrid-based Cross-Correlation using the voxel intensities in MR images corresponding to the feature points in PET images. In our experiments, we evaluate our method using software phantom and clinical datasets in the aspect of visual inspection, accuracy, robustness, and computation time. Experimental results show that our method is dramatically faster than the voxel-based registration and more accurate than the surface-based registration. In particular, our method can robustly align two datasets with large geometrical displacement and noise at optimal location.

Medical image fusion: A survey of the state of the art

“…Here, we discuss some application areas where PET is a prime modality considered in the data fusion. Similar to CT and MRI, a major application of PET is in radiology studies for brain diagnosis and treatment [47,135,48,208,140,143]. There are a wide range of the application of image fusion using PET, some of which are for cancer treatments [51,223,220,162,222,209,225,229,131], image segmentation and integration [51,259], 3D tumor simulation [191], gynecological cancer diagnosis [190], inertial electrostatic confinement fusion [270], gross tumor volume detection [200,271], diagnosis of local recurrence of rectal cancer [214], tumor detection and treatment [246,220], pediatric solid extracranial tumors [251], telemedicine [201], breast cancer detection [184,252,253], oral cancer treatment [254], lung cancer diagnosis [256,272], cervical cancer treatment [202], esophageal cancer diagnosis [260], and pancreatic tumors characterization [207,261].…”

“…There is increased interest in using fusion techniques to improve the imaging quality. The use of PET data in combination with some of the existing modalities using the image fusion techniques include MRI-CT-PET-SPECT-DSA-MEG [47,135], MRI-CT-PET [51,191,200,201,202], MRI-SPECT-PET [48], MRI/CT-PET-SPECT [190], MRI-PET [207,208,140,184,209,143,130,131], FDG-PET-CT [229,251,253], PET-CT [273,248,214,246,220,221,239,29,223,68,254,256,252,222,76,274,261], FDG-PET [272], and PET-CT-ultrasound [260].…”

“…The imaging studies reveal several important pieces of information about the brain which are otherwise not visible to human sensory mechanisms. The most commonly used image modalities to study the brain include CT [47,15,16,17,18,135,186,293,294,142,259], MRI [47,15,16,17,85,181,105,135,48,52,186,124,20,136,137,126,49,208,295,54,139,55,210,93,294,141,142,144,296], DSA [47], PET [47,135,48,208,259], SPECT [47,135,…”

“…Medical image fusion in brain studies has been employed for segmentation of brain tissues [18,105,52,126,295,139,55,50,210,144,259,296], pseudo coloring for MRI based brain segmentation [85], visualizing cortical potential fields [181], stereotactic brachytherapy of brain tumors [135], brain tissue map and volume identification [105], image guided neuro-surgery [186,297], development of stereoscopic panoramas of brain images [186], 2D-3D registration of brain images [186,136,208], volumetric fusion with brain images [124], classification of abnormal brain tissues [20], semiautomatic 3D fusion with brain images [136], image fusion with multimodal brain images [298,136,293,208,195,210,141], verification of implanted catheters [293], surface projection maximum mutual information fusion of brain images [137], parametric classification of differential brain activity [299], locating anatomical targets with MRI brain images [295], microelectrode recording and test stimulation [195], multi-classifier fusion based brain image segmentation [139], classification of differential brain activity [300], sensor fusion for surgical navigatio...…”

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

“…The imaging studies reveal several important pieces of information about the brain which are otherwise not visible to human sensory mechanisms. The most commonly used image modalities to study the brain include CT [45], MRI [46,47], PET [48,49], SPECT [50,51].For example, Alzheimer's disease was founded 100 years ago. But during the past 30 years, only researches have been developing in its risk factors, symptoms, causes, and treatments.…”