In this work, we propose a novel method for the regularization of blind deconvolution algorithms. The proposed method employs example-based machine learning techniques for modeling the space of point spread functions. During an iterative blind deconvolution process, a prior term attracts the point spread function estimates to the learned point spread function space. We demonstrate the usage of this regularizer within a Bayesian blind deconvolution framework and also integrate into the latter a method for noise reduction, thus creating a complete blind deconvolution method. The application of the proposed algorithm is demonstrated on synthetic and real-world three-dimensional images acquired by a wide-field fluorescence microscope, where the need for blind deconvolution algorithms is indispensable, yielding excellent results.
BackgroundThe performance of a prototype novel digital single-photon emission computed tomography (SPECT) camera with multiple pixelated CZT detectors and high sensitivity collimators (Digital SPECT; Valiance X12 prototype, Molecular Dynamics) was evaluated in various clinical settings.Images obtained in the prototype system were compared to images from an analog camera fitted with high-resolution collimators. Clinical feasibility, image quality, and diagnostic performance of the prototype were evaluated in 36 SPECT studies in 35 patients including bone (n = 21), brain (n = 5), lung perfusion (n = 3), and parathyroid (n = 3) and one study each of sentinel node and labeled white blood cells. Images were graded on a scale of 1–4 for sharpness, contrast, overall quality, and diagnostic confidence.ResultsDigital CZT SPECT provided a statistically significant improvement in sharpness and contrast in clinical cases (mean score of 3.79 ± 0.61 vs. 3.26 ± 0.50 and 3.92 ± 0.29 vs. 3.34 ± 0.47 respectively, p < 0.001 for both). Overall image quality was slightly higher for the digital SPECT but not statistically significant (3.74 vs. 3.66).ConclusionCZT SPECT provided significantly improved image sharpness and contrast compared to the analog system in the clinical settings evaluated. Further studies will evaluate the diagnostic performance of the system in large patient cohorts in additional clinical settings.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.