Deep learning(DL)has proven to be important for computed tomography(CT)image denoising.However,such models are usually trained under supervision,requiring paired data that may be difficult to obtain in practice.Diffus...Deep learning(DL)has proven to be important for computed tomography(CT)image denoising.However,such models are usually trained under supervision,requiring paired data that may be difficult to obtain in practice.Diffusion models offer unsupervised means of solving a wide range of inverse problems via posterior sampling.In particular,using the estimated unconditional score function of the prior distribution,obtained via unsupervised learning,one can sample from the desired posterior via hijacking and regularization.However,due to the iterative solvers used,the number of function evaluations(NFE)required may be orders of magnitudes larger than for single-step samplers.In this paper,we present a novel image denoising technique for photon-counting CT by extending the unsupervised approach to inverse problem solving to the case of Poisson flow generative models(PFGM)++.By hijacking and regularizing the sampling process we obtain a single-step sampler,that is NFE=1.Our proposed method incorporates posterior sampling using diffusion models as a special case.We demonstrate that the added robustness afforded by the PFGM++framework yields significant performance gains.Our results indicate competitive performance compared to popular supervised,including state-of-the-art diffusion-style models with NFE=1(consistency models),unsupervised,and non-DL-based image denoising techniques,on clinical low-dose CT data and clinical images from a prototype photon-counting CT system developed by GE HealthCare.展开更多
Microphysiological systems(MPS)are advanced in vitro platforms engineered to replicate in vivo conditions forstudying human biology,disease mechanisms,and drug responses with greater physiological relevance.Fluorescen...Microphysiological systems(MPS)are advanced in vitro platforms engineered to replicate in vivo conditions forstudying human biology,disease mechanisms,and drug responses with greater physiological relevance.Fluorescencesensing is widely used as a functional readout in MPS due to its high sensitivity,selectivity,and stability.However,conventional fluorescence sensing systems often rely on bulky instrumentation with limited integration,whichrestricts continuous in situ monitoring,scalable high-throughput analysis,and spatially resolved investigation in multiorgan-on-a-chip models.To address these limitations,we present a highly miniaturized,fully integrated optical systemwith a 1 mm2 footprint,enabling continuous in situ fluorescence monitoring of three-dimensional microtissues inclose proximity.The system integrates microscale illumination and sensing units for fluorescence excitation andselective detection,an optical element for guided light propagation,and a microcage for mechanical confinement ofmicrotissues.To demonstrate its capabilities,we integrated the miniaturized optical system with an MPS-relevantplatform to monitor fluorescence signals in transgenic mouse pancreatic islets expressing genetically encoded calciumindicators.The integrated platform enables real-time,continuous monitoring of islet responses to potassium chloridestimulation and tracking of calcium oscillations for over two hours,providing valuable information about thefunctional status of the pancreatic islets.Our work enhances the analytical capabilities of MPS through the integrationof miniaturized on-chip quantitative assessment tools,enabling precise,in situ,and continuous monitoring ofbiological activities in close proximity.展开更多
基金supported by MedTechLabs,GE HealthCare,the Swedish Research council,No.2021-05103the Göran Gustafsson foundation,No.2114.
文摘Deep learning(DL)has proven to be important for computed tomography(CT)image denoising.However,such models are usually trained under supervision,requiring paired data that may be difficult to obtain in practice.Diffusion models offer unsupervised means of solving a wide range of inverse problems via posterior sampling.In particular,using the estimated unconditional score function of the prior distribution,obtained via unsupervised learning,one can sample from the desired posterior via hijacking and regularization.However,due to the iterative solvers used,the number of function evaluations(NFE)required may be orders of magnitudes larger than for single-step samplers.In this paper,we present a novel image denoising technique for photon-counting CT by extending the unsupervised approach to inverse problem solving to the case of Poisson flow generative models(PFGM)++.By hijacking and regularizing the sampling process we obtain a single-step sampler,that is NFE=1.Our proposed method incorporates posterior sampling using diffusion models as a special case.We demonstrate that the added robustness afforded by the PFGM++framework yields significant performance gains.Our results indicate competitive performance compared to popular supervised,including state-of-the-art diffusion-style models with NFE=1(consistency models),unsupervised,and non-DL-based image denoising techniques,on clinical low-dose CT data and clinical images from a prototype photon-counting CT system developed by GE HealthCare.
基金supported by the Swedish Foundation for Strategic Research(SSF Grant project no.RMX18-0066)The Family Erling-Persson Foundation,The Jonas&Christina af Jochnick Foundation,ERC-2018-AdG 834860-EYELETSThe Swedish Research Council.H.K.acknowledges funding from the Wenner-Gren foundation(UPD2021-0185)。
文摘Microphysiological systems(MPS)are advanced in vitro platforms engineered to replicate in vivo conditions forstudying human biology,disease mechanisms,and drug responses with greater physiological relevance.Fluorescencesensing is widely used as a functional readout in MPS due to its high sensitivity,selectivity,and stability.However,conventional fluorescence sensing systems often rely on bulky instrumentation with limited integration,whichrestricts continuous in situ monitoring,scalable high-throughput analysis,and spatially resolved investigation in multiorgan-on-a-chip models.To address these limitations,we present a highly miniaturized,fully integrated optical systemwith a 1 mm2 footprint,enabling continuous in situ fluorescence monitoring of three-dimensional microtissues inclose proximity.The system integrates microscale illumination and sensing units for fluorescence excitation andselective detection,an optical element for guided light propagation,and a microcage for mechanical confinement ofmicrotissues.To demonstrate its capabilities,we integrated the miniaturized optical system with an MPS-relevantplatform to monitor fluorescence signals in transgenic mouse pancreatic islets expressing genetically encoded calciumindicators.The integrated platform enables real-time,continuous monitoring of islet responses to potassium chloridestimulation and tracking of calcium oscillations for over two hours,providing valuable information about thefunctional status of the pancreatic islets.Our work enhances the analytical capabilities of MPS through the integrationof miniaturized on-chip quantitative assessment tools,enabling precise,in situ,and continuous monitoring ofbiological activities in close proximity.
