检索结果-内蒙古大学图书馆

A Feasibility Study on Deep Learning Denoising for Standard-Dose Cardiac-Gated SPECT Images

学校读者我要写书评

暂无评论

A Feasibility Study on Deep Learning Denoising for Standard-...

IEEE Symposium on Nuclear Science (NSS/MIC)

作者： X. Zhang Y. Yang J. G. Brankov P.H. Pretorius M. A. King Department of Electrical and Computer Engineering Illinois Institute of Technology Chicago Chicago Illinois United States of America Department of Radiology University of Massachusetts Medical School Worcester Massachusetts United States of America

Cardiac-gated studies in myocardial perfusion imaging with SPECT are known to suffer from increased noise owing to lowered data counts within individual gate frames. In this work, we investigate the feasibility of applying a deep learning (DL) denoising network for noise suppression in cardiac-gated studies. With the lack of ground truth images in clinical cardiac-gated studies for DL network training, we employ a denoising network that were pre-trained instead on ungated imaging data in reduced-dose studies based on their equivalence in data counts. In the experiments we demonstrated this approach on a set of 50 clinical gated acquisitions. The quantitative results demonstrate that the proposed approach can effectively improve the uniformity of the left ventricular (LV) wall and suppress the frame-to-frame variations caused by imaging noise in gated images. Moreover, the DL processed images also yield volumetric measurements of the LV consistent with reconstruction with conventional Gaussian filtering

关键词：

A pre-training oracle for predicting distances in social networks

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Mahindre, Gunjan Paffenroth, Randy Jayasumana, Anura Karkare, Rasika Electrical and Computer Engineering Colorado State University Fort Collins United States Mathematical Sciences Worcester Polytechnic Institute Worcester United States Data Science Worcester Polytechnic Institute Worcester United States

In this paper, we propose a novel method to make distance predictions in real-world social networks. As predicting missing distances is a difficult problem, we take a two-stage approach. Structural parameters for families of synthetic networks are first estimated from a small set of measurements of a real-world network and these synthetic networks are then used to pre-train the predictive neural networks. Since our model first searches for the most suitable synthetic graph parameters which can be used as an "oracle" to create arbitrarily large training data sets, we call our approach "Oracle Search Pre-training" (OSP). For example, many real-world networks exhibit a Power law structure in their node degree distribution, so a Power law model can provide a foundation for the desired oracle to generate synthetic pre-training networks, if the appropriate Power law graph parameters can be estimated. Accordingly, we conduct experiments on real-world Facebook, Email, and Train Bombing networks and show that OSP outperforms models without pre-training, models pre-trained with inaccurate parameters, and other distance prediction schemes such as Low-rank Matrix Completion. In particular, we achieve a prediction error of less than one hop with only 1% of sampled distances from the social network. OSP can be easily extended to other domains such as random networks by choosing an appropriate model to generate synthetic training data, and therefore promises to impact many different network learning problems. Unpublished w Not for © 2021, CC BY.

关键词： Forecasting

A Pre-training Oracle for Predicting Distances in Social Networks

学校读者我要写书评

暂无评论

A Pre-training Oracle for Predicting Distances in Social Net...

IEEE International Conference on Big Data

作者： Gunjan Mahindre Rasika Karkare Randy Paffenroth Anura Jayasumana Electrical and Computer Engineering Colorado State University Fort Collins CO USA Data Science Worcester Polytechnic Institute Worcester MA USA Mathematical Sciences Computer Science & Data Science Worcester Polytechnic Institute Worcester MA USA

ISBN: (纸本)9781665445993

In this paper, we propose a novel method to make distance predictions in real-world social networks. As predicting missing distances is a difficult problem, we take a two-stage approach. Structural parameters for families of synthetic networks are first estimated from a small set of measurements of a real-world network and these synthetic networks are then used to pre-train the predictive neural networks. Since our model first searches for the most suitable synthetic graph parameters which can be used as an “oracle” to create arbitrarily large training data sets, we call our approach “Oracle Search Pre-training” (OSP). Such an approach enables us to evaluate the robustness of the autoencoder to artificial training data while making predictions on real-world social *** example, many real-world networks exhibit a Power law structure in their node degree distribution, so a Power law model can provide a foundation for the desired oracle to generate synthetic pre-training networks, if the appropriate Power law graph parameters can be estimated. Accordingly, we conduct experiments on real-world Facebook, Email, and Train Bombing networks and show that OSP outperforms models without pre-training, models pre-trained with inaccurate parameters, and other distance prediction schemes such as Low-rank Matrix Completion. In particular, we achieve a prediction error of less than one hop with only 1% of sampled distances from the social network. OSP can be easily extended to other domains such as random networks by choosing an appropriate model to generate synthetic training data, and therefore promises to impact many different network learning problems.

关键词： Analytical models Adaptation models Social networking (online) Neural networks Training data Predictive models Big Data

Tuning spatial entanglement in interacting two-electron quantum dots

学校读者我要写书评

暂无评论

Physical Review B 2020年第4期101卷 045306-045306页

作者： Dung N. Pham Sathwik Bharadwaj L. R. Ram-Mohan Department of Physics Worcester Polytechnic Institute Worcester Massachusetts 01609 USA Center for Computational NanoScience Worcester Polytechnic Institute Worcester Massachusetts 01609 USA Department of Electrical and Computer Engineering Worcester Polytechnic Institute Worcester Massachusetts 01609 USA Department of Mechanical Engineering Worcester Polytechnic Institute Worcester Massachusetts 01609 USA

Confined geometries such as semiconductor quantum dots are promising candidates for fabricating quantum computing devices. When several quantum dots are in proximity, spatial correlation between electrons in the system becomes significant. In this paper, we employ a variational formulation for calculating accurate two-electron wavefunctions in configuration space, irrespective of potential geometry. To evaluate the Coulomb integrals with high accuracy, a numerical integration method using multiple Gauss quadratures is used. With this approach, we investigate the confinement of two electrons in double quantum dots, and evaluate the spatial entanglement. We investigate the dependence of spatial entanglement on various geometrical parameters. We derive the two-particle wavefunctions in the asymptotic limit of the separation distance between quantum dots and obtain universal saturation values for the spatial entanglement. Resonances in the entanglement values due to avoided level crossings of states are observed. We also demonstrate the formation of electron clusters and show that the entanglement value is a good indicator for the formation of such clusters. Further, we show that a precise tuning of the entanglement values is feasible with applied external electric fields.

关键词： Entanglement entropy Quantum dots Quantum wells Semiconductors Variational wave functional methods

Reinforcement learning beyond expectation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Ramasubramanian, Bhaskar Niu, Luyao Clark, Andrew Poovendran, Radha Network Security Lab Department of Electrical and Computer Engineering University of Washington SeattleWA98195 United States Department of Electrical and Computer Engineering Worcester Polytechnic Institute WorcesterMA01609 United States

The inputs and preferences of human users are important considerations in situations where these users interact with autonomous cyber or cyber-physical systems. In these scenarios, one is often interested in aligning behaviors of the system with the preferences of one or more human users. Cumulative prospect theory (CPT) is a paradigm that has been empirically shown to model a tendency of humans to view gains and losses differently. In this paper, we consider a setting where an autonomous agent has to learn behaviors in an unknown environment. In traditional reinforcement learning, these behaviors are learned through repeated interactions with the environment by optimizing an expected utility. In order to endow the agent with the ability to closely mimic the behavior of human users, we optimize a CPT-based cost. We introduce the notion of the CPT-value of an action taken in a state, and establish the convergence of an iterative dynamic programming-based approach to estimate this quantity. We develop two algorithms to enable agents to learn policies to optimize the CPT-value, and evaluate these algorithms in environments where a target state has to be reached while avoiding obstacles. We demonstrate that behaviors of the agent learned using these algorithms are better aligned with that of a human user who might be placed in the same environment, and is significantly improved over a baseline that optimizes an expected utility. © 2021, CC BY.

关键词： Autonomous agents

Improved Performance of a Multipinhole SPECT for DAT Imaging by Increasing Number of Pinholes at the Expense of Increased Multiplexing

学校读者我要写书评

暂无评论

Improved Performance of a Multipinhole SPECT for DAT Imaging...

作者： Konik, Arda Zeraatkar, Navid Kalluri, Kesava S. Auer, Benjamin Fromme, Timothy J. He, Yulun May, Micaehla Furenlid, Lars R. Kuo, Phillip H. King, Michael A. Department of Imaging Dana Farber Cancer Institute BostonMA02215 United States Department of Radiology University of Massachusetts Medical School WorcesterMA01605 United States Department of Robotics Engineering Worcester Polytechnic Institute WorcesterMA01609 United States Department of Computer Science MD Anderson Cancer Center HoustonTX United States Department of Radiology University of Arizona TucsonAZ85724 United States

Single photon emission tomography imaging of dopamine transporters (DATs) in the brain is a widely utilized study to improve the diagnosis of Parkinsonian syndromes, where conventional (parallel-hole and fan-beam) collimators on dual-head scanners are commonly employed. We have designed a multipinhole (MPH) collimator to improve the performance of DAT imaging. The MPH collimator focuses on the striatum and hence offers a better tradeoff for sensitivity and spatial resolution than the conventional collimators within this clinically most relevant region for DAT imaging. Our original MPH design consisted of 9 pinholes with a background-to-striatal (Bkg/Str) projection multiplexing of 1% only. In this simulation study, we investigated whether further improvements in the performance of MPH imaging could be obtained by increasing the number of pinholes, hence by enhancing the sensitivity and sampling, despite the ambiguity in reconstructing images due to increased multiplexing. We performed analytic simulations of the MPH configurations with 9, 13, and 16 pinholes (aperture diameters: 4-6 mm) using a digital phantom modeling DAT imaging. Our quantitative analyses indicated that using 13 (Bkg/Str: 12%) and 16 (Bkg/Str: 22%) pinholes provided better performance than the original 9-pinhole configuration for the acquisition with two or four angular views, but a similar performance with 8 and 16 views. © 2017 IEEE.

关键词： Photons

Evaluating cardiac function in ex vivo heart perfusion using lumped parameter models

学校读者我要写书评

暂无评论

computers in Biology and Medicine 2025年 189卷 109922-109922页

作者： Roig, Raimon Casamitjana Li, Selena S. Asheghan, Mostafa Olverson, George Vincent, Doug Bolger-Chen, Maya Ajenu, Emmanuella Higuita, Manuela Lopera Tessier, Shannon N. Osho, Asishana D'Alessandro, David A. Rabi, S. Alireza Nezami, Farhad R. Division of Cardiac Surgery Department of Surgery Brigham and Women's Hospital Harvard Medical School BostonMA United States Division of Cardiac Surgery Department of Surgery Massachusetts General Hospital Harvard Medical School BostonMA United States Electrical Engineering Department Worcester Polytechnic Institute WorcesterMA United States Ventriflo Inc. PelhamNH United States Center for Engineering in Medicine and Surgery Massachusetts General Hospital Harvard Medical School and Shriners Children's Hospital BostonMA United States

Background: Heart transplant outcomes and survival depend on the ability to implant well-functioning organs, but there remain no reliable, objective measures of cardiac function prior to implantation. The lack of standardized protocols and advanced technologies results in inconsistencies and subjective assessments, increasing the risk for postoperative graft dysfunction, the leading cause of short-term morbidity and mortality after transplant. Ex-vivo heart perfusion (EVHP) provides a platform to evaluate donor hearts prior to implantation, using machine perfusion to reanimate the heart to a beating, physiologic state. The FDA-approved Organ Care System (OCS) is widely utilized for the evaluation and ex vivo preservation of hearts, particularly from donors after circulatory death (DCD). However, it does not permit a physiological assessment of heart function because, while the heart continues to beat, its chambers remain devoid of perfusate and thus are unable to perform any functional work. Method: In this study, we developed and validated a lumped parameter mathematical model to assess donor hearts during ex-vivo perfusion, using a customized, in-house EVHP setup that allows left ventricular loading. Results: We demonstrate the ability of our mathematical model to accurately predict hemodynamic parameters, enabling performance analysis of hearts during EVHP. Our model generates pressure-volume loops, allowing for the computation of ejection fraction, and was verified with experimental measurements taken via echocardiography. Conclusion: This promising tool demonstrates the unique opportunity to utilize mathematical modeling in the assessment of donor hearts, streamlining their performance evaluation. Ultimately, a more accurate assessment of donor hearts on EVHP may improve our utilization of available donor hearts, addressing the donor organ shortage that continues to limit transplant capabilities. © 2025

关键词： Heart

Training deep neural networks via branch-and-bound

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Wu, Yuanwei Zhang, Ziming Wang, Guanghui Department of Electrical Engineering & Computer Science University of Kansas LawrenceKS United States Department of Electrical & Computer Engineering Worcester Polytechnic Institute WorcesterMA United States Department of Computer Science Ryerson University TorontoON Canada

In this paper, we propose BPGrad, a novel approximate algorithm for deep nueral network training, based on adaptive estimates of feasible region via branch-and-bound. The method is based on the assumption of Lipschitz continuity in objective function, and as a result, it can adaptively determine the step size for the current gradient given the history of previous updates. We prove that, by repeating such a branch-and-pruning procedure, it can achieve the optimal solution within finite iterations. A computationally efficient solver based on BPGrad has been proposed to train the deep neural networks. Empirical results demonstrate that BPGrad solver works well in practice and compares favorably to other stochastic optimization methods in the tasks of object recognition, detection, and segmentation. The code is available at https://***/RyanCV/BPGrad. © 2021, CC BY-NC-SA.

关键词： Deep neural networks

B-COLORINGS OF PLANAR AND OUTERPLANAR GRAPHS

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Martin, Ryan R. Ruszinkó, Miklós Sárközy, Gábor N. Department of Mathematics Iowa State University AmesIA United States HUN-REN Alfréd Rényi Institute of Mathematics Budapest Hungary Department of Sciences and Engineering Sorbonne University Abu Dhabi Abu Dhabi United Arab Emirates Faculty of Information Technology and Bionics Pázmány Péter Catholic University Hungary Computer Science Department Worcester Polytechnic Institute WorcesterMA United States

A coloring of the edges of a graph G in which every K1,2 is totally multicolored is known as a proper coloring and a coloring of the edges of G in which every K1,2 and every K2,2 is totally multicolored is called a B-coloring. In this paper, we establish that a planar graph with maximum degree ∆ can be B-colored with max{2∆, 32} colors. This is best-possible for large ∆ because K2,∆ requires 2∆ colors. In addition, there is an example with ∆ = 4 that requires 12 colors. We also establish that an outerplanar graph with maximum degree ∆ can be B-colored with max{∆, 6} colors. This is almost best-possible because ∆ colors are necessary and there is an example with ∆ = 4 that requires 5 *** Codes 05C15, 05C10, 05C35 © 2024, CC BY.

关键词：