Gong K.Cheng-Liao J.Wang G.Chen K.T.Catana C.Qi J.TZE-HSIANG CHEN2022-05-242022-05-242018https://www.scopus.com/inward/record.uri?eid=2-s2.0-85035797973&doi=10.1109%2fTMI.2017.2776324&partnerID=40&md5=ee0acc0782a7e1dc1f980acace839b30https://scholars.lib.ntu.edu.tw/handle/123456789/611669Positron emission tomography (PET) is a functional imaging modality widely used in oncology, cardiology, and neuroscience. It is highly sensitive, but suffers from relatively poor spatial resolution, as compared with anatomical imaging modalities, such as magnetic resonance imaging (MRI). With the recent development of combined PET/MR systems, we can improve the PET image quality by incorporating MR information into image reconstruction. Previously, kernel learning has been successfully embedded into static and dynamic PET image reconstruction using either PET temporal or MRI information. Here, we combine both PET temporal and MRI information adaptively to improve the quality of direct Patlak reconstruction. We examined different approaches to combine the PET and MRI information in kernel learning to address the issue of potential mismatches between MRI and PET signals. Computer simulations and hybrid real-patient data acquired on a simultaneous PET/MR scanner were used to evaluate the proposed methods. Results show that the method that combines PET temporal information and MRI spatial information adaptively based on the structure similarity index has the best performance in terms of noise reduction and resolution improvement. ? 1982-2012 IEEE.ElectronsFeature extractionHospital data processingImage enhancementImage qualityImage reconstructionMagnetic resonance imagingMedical computingPositronsKernelKernel methodsMagnetic Resonance Imaging (MRI)Positron emission tomography (PET)Resolution improvementSpatial informationsStructural similarityStructure similarityPositron emission tomographyfluorodeoxyglucoseArticlecomputer simulationevaluation studygray matterhumanimage analysisimage qualityimage reconstructionintermethod comparisonkernel methodneoplasmneuroimagingnoise reductionnuclear magnetic resonance imagingpositron emission tomographyquantitative analysissignal processingspatiotemporal analysistotal quality managementwhite matteralgorithmbraindiagnostic imagingimage processingimaging phantomproceduresstatistical modelAlgorithmsBrainHumansImage Processing, Computer-AssistedMagnetic Resonance ImagingModels, StatisticalPhantoms, ImagingPositron-Emission Tomographyjournal article10.1109/TMI.2017.27763242-s2.0-85035797973