Chi Liu, PhD
Professor of Radiology and Biomedical ImagingDownloadHi-Res Photo
Cards
Appointments
Contact Info
About
Titles
Professor of Radiology and Biomedical Imaging
Biography
Chi Liu received his Ph.D. in 2008 from Johns Hopkins University with emphasis on quantitative SPECT/CT imaging. Following his graduate work, he was a postdoctoral fellow at University of Washington, specializing in oncological PET/CT studies with emphasis on compensation algorithms for respiratory motion. In 2010, he joined Yale University as a faculty member. He is board certified in Nuclear Medicine physics and instrumentation by the American Board of Science in Nuclear Medicine. His current research focuses on quantitative cardiac and oncological PET/CT and SPECT/CT imaging, including deep learning algorithms, reconstruction algorithms, data correction, dynamic imaging, and translational imaging. The translational and clinical applications of these projects include early detection of chemotherapy-induced cardiotoxicity, multimodality imaging of heart failure, and eliminating respiratory motion variability for assessing response to therapy. Many of the imaging technologies developed in his lab has been or is being implemented in clinical PET and SPECT scanners. In 2012, he was awarded with the Bruce Hasegawa Young Investigator Medical Imaging Science Award from the IEEE Nuclear Medical and Imaging Sciences Council for “contributions to the imaging physics of SPECT/CT and PET/CT, with emphasis in quantitative imaging and motion correction”. He was the President of Physics, Instrumentation, and Data Sciences Council (PIDSC) of the Society of Nuclear Medicine and Molecular Imaging (SNMMI) between 2022-2023, is currently the Immediate Past President of PIDSC.
Appointments
Radiology & Biomedical Imaging
ProfessorPrimary
Other Departments & Organizations
- Bioimaging Sciences
- Center for Brain & Mind Health
- Chi Liu Lab
- Medical Physics
- Positron Emission Tomography (PET)
- Radiology & Biomedical Imaging
- Yale Translational Research Imaging Center (Y-TRIC)
- Yale Ventures
Education & Training
- Postdoctoral Fellow
- University of Washington (2010)
- PhD
- Johns Hopkins University (2008)
- Board Certification
- Nuclear Medicine Physics and Instrumentation, American Board of Science in Nuclear Medicine
Research
Overview
Machine learning and deep learning in imaging applications
Radiation dose reduction methods in PET/SPECT/CT
Motion correction methods for PET/CT and SPECT/CT
ORCID
0000-0002-7007-1037
Research at a Glance
Yale Co-Authors
Frequent collaborators of Chi Liu's published research.
Publications Timeline
A big-picture view of Chi Liu's research output by year.
Albert Sinusas, MD
Jean-Dominique Gallezot, PhD
Edward J Miller, MD, PhD
Ming-Kai Chen, MD, PhD
Yi-Hwa Liu, PhD
John Onofrey, PhD
125Publications
2,090Citations
Publications
2024
Noise-aware dynamic image denoising and positron range correction for Rubidium-82 cardiac PET imaging via self-supervision
Xie H, Guo L, Velo A, Liu Z, Liu Q, Guo X, Zhou B, Chen X, Tsai Y, Miao T, Xia M, Liu Y, Armstrong I, Wang G, Carson R, Sinusas A, Liu C. Noise-aware dynamic image denoising and positron range correction for Rubidium-82 cardiac PET imaging via self-supervision. Medical Image Analysis 2024, 100: 103391. PMID: 39579623, DOI: 10.1016/j.media.2024.103391.Peer-Reviewed Original ResearchConceptsImage denoisingPositron range correctionDynamic framesSelf-supervised methodsSuperior visual qualityLow signal-to-noise ratioCardiac PET imagingDenoising methodSignal-to-noise ratioSelf-supervisionVisual qualityHigh-energy positronsRange correctionsDenoisingNoise levelImage spatial resolutionImage qualityDefect contrastPET imagingImage quantificationRadioactive isotopesPatient scansQuantitative accuracyImagesFramePatlak-Guided Self-Supervised Learning for Dynamic PET Denoising
Liu Q, Guo X, Tsai Y, Gallezot J, Chen M, Guo L, Xie H, Pucar D, Young C, Panin V, Carson R, Liu C. Patlak-Guided Self-Supervised Learning for Dynamic PET Denoising. 2024, 00: 1-2. DOI: 10.1109/nss/mic/rtsd57108.2024.10655866.Peer-Reviewed Original ResearchConceptsPre-trained modelsSelf-supervised learning methodSuperior noise reductionNoise reductionDynamic framesImage quality improvementUpsampling blockSignal-to-noise ratioWeight initializationWeak supervisionDynamic PET datasetsEnhanced noise reductionUNet modelLearning methodsTraining schemeTemporal dataStatic imagesDenoisingReconstruction methodPET datasetsLesion signal-to-noise ratioSize constraintsLesion SNRImagesReconDose-aware Diffusion Model for 3D Low-count Cardiac SPECT Image Denoising with Projection-domain Consistency
Xie H, Gan W, Chen X, Zhou B, Liu Q, Xia M, Guo X, Liu Y, An H, Kamilov U, Wang G, Sinusas A, Liu C. Dose-aware Diffusion Model for 3D Low-count Cardiac SPECT Image Denoising with Projection-domain Consistency. 2024, 00: 1-1. DOI: 10.1109/nss/mic/rtsd57108.2024.10655170.Peer-Reviewed Original ResearchConceptsImage denoisingImage denoising performanceDeep learning techniquesNoise-levelDenoising performanceDenoising resultsNeural networkLearning techniquesSPECT imagesLow count levelsSPECT scansDenoisingSampling stepIterative reconstructionNoise amplitudeImagesInjected dosePatient studiesDiffusion modelRadiation exposureCardiology studiesSPECTNetworkStochastic natureMLEMDeep Learning-based Dynamic PET Intra-frame Motion Correction and Integration with Inter-frame Motion Estimation
Guo X, Tsai Y, Liu Q, Guo L, Valadez G, Dvornek N, Liu C. Deep Learning-based Dynamic PET Intra-frame Motion Correction and Integration with Inter-frame Motion Estimation. 2024, 00: 1-1. DOI: 10.1109/nss/mic/rtsd57108.2024.10657268.Peer-Reviewed Original ResearchConceptsIntra-frame motionMotion correctionGated imagesLearning-based registration approachesDeep learning-based worksInter-frame motion estimationConventional image registrationLearning-based worksImage registrationMotion estimation processMotion estimation frameworkInter-frame registrationRespiratory gatingImprove image sharpnessInter-frameInference timeMotion estimationReconstructed framesDynamic PET datasetsGeneralization abilityPET imagingConventional registrationDynamic PET imagesImprove image qualityComputational inefficiencyAnatomically and Metabolically Informed Deep Learning Low-Count PET Image Denoising
Xia M, Xie H, Liu Q, Guo L, Ouyang J, Bayerlein R, Spencer B, Badawi R, Li Q, Fakhri G, Liu C. Anatomically and Metabolically Informed Deep Learning Low-Count PET Image Denoising. 2024, 00: 1-2. DOI: 10.1109/nss/mic/rtsd57108.2024.10657099.Peer-Reviewed Original ResearchConceptsDeep learningOver-smoothed imagesDL training processesHigh-count imagesImage denoisingDenoised imageLow-count dataSemantic informationSemantic classesSegmentation guidanceTraining processPET/CT systemHistogram distributionImage qualitySegmentation toolPositron emission tomographyImagesDenoisingDatasetHistogramPriorsRadiation exposure2.5D Multi-view Averaging Diffusion Model for 3D Medical Image Translation: Application to Low-count PET Reconstruction with CT-less AC
Chen T, Hou J, Xie H, Chen X, Zhou Y, Xia M, Duncan J, Liu C, Zhou B. 2.5D Multi-view Averaging Diffusion Model for 3D Medical Image Translation: Application to Low-count PET Reconstruction with CT-less AC. 2024, 00: 1-2. DOI: 10.1109/nss/mic/rtsd57108.2024.10658551.Peer-Reviewed Original ResearchCitationsConceptsLow-dose PETStandard-dose PETImage-to-image translationPositron emission tomographyAttenuation correctionPET reconstructionOverall radiation doseCT acquisitionState-of-the-art deep learning methodsRadiation hazardRadiation doseCNN-based methodsState-of-the-artMedical image translationPatient studiesDiffusion modelDeep learning methodsHigh computation costHuman patient studiesClinical imaging toolImage translationBaseline methodsMulti-viewCNN-basedMultiple viewsPOUR-Net: A Population-Prior-Aided Over-Under-Representation Network for Low-Count PET Attenuation Map Generation
Zhou B, Hou J, Chen T, Zhou Y, Chen X, Xie H, Liu Q, Guo X, Xia M, Tsai Y, Panin V, Toyonaga T, Duncan J, Liu C. POUR-Net: A Population-Prior-Aided Over-Under-Representation Network for Low-Count PET Attenuation Map Generation. 2024, 00: 1-1. DOI: 10.1109/nss/mic/rtsd57108.2024.10658051.Peer-Reviewed Original ResearchConceptsPET attenuation correctionLow-dose PETAttenuation correctionU-mapAttenuation mapElevated radiation doseRadiation doseEfficient feature extractionRadiation exposurePET imagingFinely detailed featuresBaseline methodsMitigate radiation exposureFeature extractionCorrectionMap generationGeneration machinesGeneration of Synthetic brain PET images of synaptic density from MRI and FDG-PET using a Multi-stage U-Net
Zheng X, Worhunsky P, Liu Q, Zhou B, Chen X, Guo X, Xie H, Sun H, Zhang J, Toyonaga T, Mecca A, O’Dell R, van Dyck C, Carson R, Radhakrishnan R, Liu C. Generation of Synthetic brain PET images of synaptic density from MRI and FDG-PET using a Multi-stage U-Net. 2024, 00: 1-2. DOI: 10.1109/nss/mic/rtsd57108.2024.10655600.Peer-Reviewed Original ResearchPDM: A Plug-and-Play Perturbed Multi-path Diffusion Module for Simultaneous Medical Image Segmentation Improvement and Uncertainty Estimation
Zhou B, Chen T, Hou J, Zhou Y, Xie H, Liu C, Duncan J. PDM: A Plug-and-Play Perturbed Multi-path Diffusion Module for Simultaneous Medical Image Segmentation Improvement and Uncertainty Estimation. Lecture Notes In Computer Science 2024, 15241: 259-268. DOI: 10.1007/978-3-031-73284-3_26.Peer-Reviewed Original ResearchConceptsEfficient plug-and-play moduleDenoising diffusion probabilistic modelPlug-and-play moduleDiffusion probabilistic modelState-of-the-artMedical image analysisDeep modelsSegmentation datasetUncertainty estimationSegmentation resultsImproved segmentationSegmentation modelMorphological operationsBinary segmentationSegmentation improvementProbabilistic modelUncertainty mapsDiffusion moduleReverse pathPerturbation segmentsSegmental inputsImage analysisSegmentsInputModulationDuDoCFNet: Dual-Domain Coarse-to-Fine Progressive Network for Simultaneous Denoising, Limited-View Reconstruction, and Attenuation Correction of Cardiac SPECT
Chen X, Zhou B, Guo X, Xie H, Liu Q, Duncan J, Sinusas A, Liu C. DuDoCFNet: Dual-Domain Coarse-to-Fine Progressive Network for Simultaneous Denoising, Limited-View Reconstruction, and Attenuation Correction of Cardiac SPECT. IEEE Transactions On Medical Imaging 2024, 43: 3110-3125. PMID: 38578853, PMCID: PMC11539864, DOI: 10.1109/tmi.2024.3385650.Peer-Reviewed Original ResearchConceptsMulti-task learning methodCross-domainLimited-viewLearning methodsCoarse-to-fine estimationProgressive networkDual domainCross-modal feature fusionDual-domain networkProgressive learning strategyCross-modal informationSimultaneous denoisingFeature fusionSingle-photon emission computed tomographyImage domainCardiac single-photon emission computed tomographyReconstruction accuracyDenoisingHardware expenseFusion mechanismAccelerated scansImage noiseM-mapSuperior accuracyNetwork
Academic Achievements & Community Involvement
honor Bruce Hasegawa Young Investigator Medical Imaging Science Award
International AwardIEEE Nuclear Medical and Imaging Sciences CouncilDetails10/31/2012United Stateshonor Junior Faculty Scholar Award
Yale University AwardYale Center for Clinical InvestigationDetails07/01/2011United Stateshonor Young Investigators Award Honorable Mention
International AwardSociety of Nuclear Medicine, Computer and Instrumentation CouncilDetails06/05/2010United States
News
News
- October 15, 2024
CMITT presentations at upcoming IEEE NSS/MIC/RTSD conference
- April 01, 2024
Yale Faculty Present Groundbreaking Clinical Research at the 2024 American College of Cardiology Scientific Sessions
- February 20, 2023Source: Yale Ventures
Roberts Innovation Fund to Support 10 Bold SEAS Faculty Inventions
- October 02, 2020
Yale PET Center Receives $10.2M BRAIN Initiative Grant to Build New Scanner
Get In Touch
Contacts
Email
Academic Office Number