检索结果-内蒙古大学图书馆

Performance of multimodal prediction models for intracerebral hemorrhage outcomes using real-world data

International Journal of Medical Informatics 2025年 202卷 105989-105989页

作者： Matsumoto, Koutarou Suzuki, Masahiro Ishihara, Kazuaki Tokunaga, Koki Matsuda, Katsuhiko Chen, Jenhui Yamashiro, Shigeo Soejima, Hidehisa Nakashima, Naoki Kamouchi, Masahiro Department of Health Care Administration and Management of Medical Sciences Kyushu University Fukuoka Japan Graduate Degree Program of Applied Data Sciences Sophia University Tokyo Japan Department of Computer Science and Information Engineering Chang Gung University Taoyuan Taiwan Department of Pharmacy Saiseikai Kumamoto Hospital Kumamoto Japan Department of Radiology Saiseikai Kumamoto Hospital Kumamoto Japan Division of Neurosurgery Saiseikai Kumamoto Hospital Kumamoto Japan Institute for Medical Information Research and Analysis Saiseikai Kumamoto Hospital Kumamoto Japan Medical Information Center Kyushu University Hospital Fukuoka Japan Department of Medical Informatics of Medical Sciences Kyushu University Fukuoka Japan Center for Cohort Studies of Medical Sciences Kyushu University Fukuoka Japan

Background: We aimed to develop and validate multimodal models integrating computed tomography (CT) images, text and tabular clinical data to predict poor functional outcomes and in-hospital mortality in patients with intracerebral hemorrhage (ICH). These models were designed to assist non-specialists in emergency settings with limited access to stroke specialists. Methods: A retrospective analysis of 527 patients with ICH admitted to a Japanese tertiary hospital between April 2019 and February 2022 was conducted. Deep learning techniques were used to extract features from three-dimensional CT images and unstructured data, which were then combined with tabular data to develop an L1-regularized logistic regression model to predict poor functional outcomes (modified Rankin scale score 3–6) and in-hospital mortality. The model's performance was evaluated by assessing discrimination metrics, calibration plots, and decision curve analysis (DCA) using temporal validation data. Results: The multimodal model utilizing both imaging and text data, such as medical interviews, exhibited the highest performance in predicting poor functional outcomes. In contrast, the model that combined imaging with tabular data, including physiological and laboratory results, demonstrated the best predictive performance for in-hospital mortality. These models exhibited high discriminative performance, with areas under the receiver operating curve (AUROCs) of 0.86 (95% CI: 0.79–0.92) and 0.91 (95% CI: 0.84–0.96) for poor functional outcomes and in-hospital mortality, respectively. Calibration was satisfactory for predicting poor functional outcomes, but requires refinement for mortality prediction. The models performed similar to or better than conventional risk scores, and DCA curves supported their clinical utility. Conclusion: Multimodal prediction models have the potential to aid non-specialists in making informed decisions regarding ICH cases in emergency departments as part of clinical dec

关键词： Logistic regression

来源：评论

学校读者我要写书评

暂无评论

Faster GröBner Bases Via Domain-Specific Ordering in Parameter Identifiability of Ode Models

SSRN

引用

SSRN 2023年

作者： Bessonov, Mariya Ilmer, Ilia Konstantinova, Tatiana Ovchinnikov, Alexey Pogudin, Gleb Soto, Pedro Brookhaven National Laboratory United States Department of Mathematics CUNY NYC College of Technology New YorkNY United States Ph.D. Program in Computer Science CUNY Graduate Center New YorkNY United States Department of Mathematics CUNY Queens College New YorkNY United States LIX CNRS École Polytechnique Institute Polytechnique de Paris Paris France Mathematical Institute University of Oxford Oxford United Kingdom

We consider a specific class of polynomial systems that arise in parameter identifiability problems of models of ordinary differential equations (ODE) and discover a method for speeding up the Gr¨obner basis computation by using specific variable ordering and weights coming from the structure of the ODE model. We provide empirical results that show improvement across different symbolic computing frameworks. © 2023, The Authors. All rights reserved.

关键词： Ordinary differential equations

来源：评论

学校读者我要写书评

暂无评论

Learning Vector Quantized Shape Code for Amodal Blastomere Instance Segmentation

Learning Vector Quantized Shape Code for Amodal Blastomere I...

引用

IEEE International Symposium on Biomedical Imaging

作者： Won-Dong Jang Donglai Wei Xingxuan Zhang Brian Leahy Helen Yang James Tompkin Dalit Ben-Yosef Daniel Needleman Hanspeter Pfister School of Engineering and Applied Sciences Harvard University MA USA Boston College MA USA Department of Molecular and Cellular Biology Harvard University MA USA Harvard Graduate Program in Biophysics Harvard University MA USA Department of Computer Science Brown University RI USA Tel Aviv Sourasky Medical Center Fertility & IVF Institute Israel Dept. of Cell and Developmental Biology Sagol School of Neuroscience Tel Aviv University Israel

Blastomere instance segmentation is important for analyzing embryos’ abnormality. To measure the accurate shapes and sizes of blastomeres, their amodal segmentation is necessary. Amodal instance segmentation aims to recover an object’s complete silhouette even when the object is not fully visible. For each detected object, previous methods directly regress the target mask from input features. However, images of an object under different amounts of occlusion should have the same amodal mask output, making it harder to train the regression model. To alleviate the problem, we propose to classify input features into intermediate shape codes and recover complete object shapes. First, we pre-train the Vector Quantized Variational Autoencoder (VQ-VAE) model to learn these discrete shape codes from ground truth amodal masks. Then, we incorporate the VQ-VAE model into the amodal instance segmentation pipeline with an additional refinement module. We also detect an occlusion map to integrate occlusion information with a backbone feature. As such, our network faithfully detects bounding boxes of amodal objects. On an internal embryo cell image benchmark, the proposed method outperforms previous state-of-the-art methods. To show generalizability, we show segmentation results on the public KINS natural image benchmark. Our method would enable accurate measurement of blastomeres in In Vitro Fertilization (IVF) clinics, potentially increasing the IVF success rate.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Performance comparison of InGaN-based 40–80 μm micro-LEDs fabricated with and without plasma etching

引用

Materials Today Advances 2024年 22卷

作者： Lo, Yu-Yun Chen, Yi-Ho Hsu, Yun-Cheng Lee, Tzu-Yi Hung, Yu-Ying Kao, Yu-Cheng Zan, Hsiao-Wen Wuu, Dong- Sing Kuo, Hao-Chung Samukawa, Seiji Horng, Ray-Hua Institute of Electronics National Yang Ming Chiao Tung University 1001 University Road Hsinchu30010 Taiwan Department of Photonics College of Electrical and Computer Engineering National Yang Ming Chiao Tung University Hsinchu30010 Taiwan Graduate Degree Program of College of Electrical and Computer Engineering National Yang Ming Chiao Tung University Hsinchu30010 Taiwan Department of Materials Science and Engineering National Chung Hsing University Taichung40227 Taiwan Department of Applied Materials and Optoelectronic Engineering National Chi Nan University Nantou54561 Taiwan Semiconductor Research Center Hon Hai Research Institute Taipei11492 Taiwan Institute of Communications Engineering National Yang Ming Chiao Tung University Hsinchu30010 Taiwan

The fabrication of InGaN-based blue 4✕4 array micro-LEDs (μLEDs) with 40 μm ✕40 μm chip size and 2✕2 array μLEDs with 80 μm ✕80 μm chip size etching by the inductive coupled plasma reactive ion etching (ICPRIE) and defect-free neutral beam etching (NBE) processes was studied in this work. In μLEDs of this size, the influence of defects formation in the sidewalls on EQE was evaluated. There was almost no difference in EQE between μLEDs array etched by the NBE process no matter 40 μm ✕40 μm and 80 μm ✕80 μm, but the dependence was observed in the ICPRIE. Even with this size, it was found that the size effect of EQE is smaller than that in case of using ICPRIE for defect-free neutral beam etching. This impact is substantial since μLED predominantly operated at low current density, around 1–5 A/cm2. Consequently, the reduction of defect density, encompassing both internal and sidewall defects, becomes imperative even in 40–80 μm InGaN-based μLEDs. This not only improves the overall efficiency of μLEDs but also fortifies the brightness stability of μLED displays if process etching by NBE. It was also found that the etching shape had an influence on EQE. It could be attributed to fact that the etching profile angle of NBE was more vertical than that of ICPRIE. Because the different angles of the mesa resulted in different light intensity. The μLEDs emitting with a wavelength of 450 nm, the light extraction efficiency and intensity at a mesa angle 58° of NBE etching μLEDs was about 8% lower than those of an angle (38°) of ICPRIE etching μLEDs by simulation. © The Authors

关键词： Light emitting diodes

来源：评论

学校读者我要写书评

暂无评论

Faster Gröbner bases for Lie derivatives of ODE systems via monomial orderings

arXiv

引用

arXiv 2022年

作者： Bessonov, Mariya Ilmer, Ilia Konstantinova, Tatiana Ovchinnikov, Alexey Pogudin, Gleb Soto, Pedro Department of Mathematics CUNY NYC College of Technology New YorkNY United States Ph.D. Program in Computer Science CUNY Graduate Center New YorkNY United States Department of Mathematics CUNY Queens College New YorkNY United States LIX CNRS École Polytechnique Institute Polytechnique de Paris Paris France

Symbolic computation for systems of differential equations is often computationally expensive. Many practical differential models have a form of polynomial or rational ODE system with specified outputs. A basic symbolic approach to analyze these models is to compute and then symbolically process the polynomial system obtained by sufficiently many Lie derivatives of the output functions with respect to the vector field given by the ODE system. In this paper, we present a method for speeding up Gröbner basis computation for such a class of polynomial systems by using specific monomial ordering, including weights for the variables, coming from the structure of the ODE model. We provide empirical results that show improvement across different symbolic computing frameworks and apply the method to speed up structural identifiability analysis of ODE models. © 2022, CC BY-NC-SA.

关键词： Polynomials

来源：评论

学校读者我要写书评

暂无评论

Improved MVDR beamforming using LSTM speech models to clean spatial clustering masks

arXiv

引用

arXiv 2020年

作者： Ni, Zhaoheng Grezes, Felix Trinh, Viet Anh Mandel, Michael I. Computer Science Program Graduate Center City University of New York United States Computer and Information Science Brooklyn College City University of New York United States

Spatial clustering techniques can achieve significant multichannel noise reduction across relatively arbitrary microphone configurations, but have difficulty incorporating a detailed speech/noise model. In contrast, LSTM neural networks have successfully been trained to recognize speech from noise on single-channel inputs, but have difficulty taking full advantage of the information in multi-channel recordings. This paper integrates these two approaches, training LSTM speech models to clean the masks generated by the Model-based EM Source Separation and Localization (MESSL) spatial clustering method. By doing so, it attains both the spatial separation performance and generality of multi-channel spatial clustering and the signal modeling performance of multiple parallel single-channel LSTM speech enhancers. Our experiments show that when our system is applied to the CHiME-3 dataset of noisy tablet recordings, it increases speech quality as measured by the Perceptual Evaluation of Speech Quality (PESQ) algorithm and reduces the word error rate of the baseline CHiME-3 speech recognizer, as compared to the default BeamformIt beamformer. © 2020, CC BY.

关键词： Cluster analysis

来源：评论

学校读者我要写书评

暂无评论

Homogenization and design of acoustic Willis metasurfaces

引用

Physical Review B 2021年第5期103卷 054306-054306页

作者： Hussein Esfahlani Yarden Mazor Andrea Alù Photonics Initiative Advanced Science Research Center City University of New York New York New York 10031 USA Kymeta Corporation Redmond Washington 98052 USA Department of Electrical and Computer Engineering The University of Texas at Austin Austin Texas 78712 USA Physics Program Graduate Center of the City University of New York New York New York 10016 USA

Metasurfaces are ultrathin planar arrays of carefully tailored subwavelength particles that enable agile and flexible manipulation of the impinging waves. Originally introduced in optics, their application to acoustic waves has recently opened exciting opportunities for exotic sound control. In conventional acoustic inclusions, the interactions with the impinging pressure and velocity are decoupled, limiting the functionalities that arrays of them can achieve. While the coupling between these two quantities in symmetry-breaking inclusions, known as Willis coupling, has been discussed for several years, only recently has it been realized that these phenomena can become nonperturbative in suitably tailored resonant scatterers. Here, we explore the opportunities that these Willis meta-atoms open in the context of acoustic metasurfaces, offering additional knobs to manipulate and tailor sound. The general response of Willis metasurface is analytically derived, yielding fundamental bounds and optimal surface responses enabling full control of the impinging acoustic wave front.

关键词： Acoustic metamaterials Acoustic wave phenomena Polarizability

来源：评论

学校读者我要写书评

暂无评论

Turing Tumble Is P(SPACE)-Complete 11th

Turing Tumble Is P(SPACE)-Complete

引用

11th International Conference on Algorithms and Complexity, CIAC 2019

作者： Johnson, Matthew P. Department of Computer Science Lehman College Ph.D. Program in Computer Science The Graduate Center City University of New York New YorkNY United States

ISBN: (纸本)9783030174019

Turing Tumble is a toy gravity-fed mechanical computer (similar to the classic, but including additional types of pieces such as gears), in which marbles roll down a board, along paths determined by the locations of ramps, toggles and gears, which are placed by the "programmer," and by their current states, which are altered by the passing marbles. Aaronson proved that a decision problem (viz., will any marbles reach the sink?) is -Complete, i.e., equivalent to evaluating comparator circuits, and posed the question of what additional functionality would raise the machine’s computational power beyond, speculating that a capability for toggles to affect one another’s states (which gears happen to provide) might suffice. This turns out to be so: we show, though a simple reduction from a variant of the circuit value problem (), that the decision problem is -Complete. The two models also differ in complexity when exponentially (or unboundedly) many marbles are permitted: while remains in, becomes -Complete. © 2019, Springer Nature Switzerland AG.

关键词： Marble

来源：评论

学校读者我要写书评

暂无评论

Monitoring the contribution of development teams in the evolution of technical debt items in software projects 25

Monitoring the contribution of development teams in the evol...

引用

25th Americas Conference on Information Systems, AMCIS 2019

作者： dos Santos, Sandro L.A.C. Rios, Nicolli Mendonça, Manoel Spínola, Rodrigo O. Graduate Program in Systems and Computer Salvador University Salvador Brazil Department of Computer Science Federal University of Bahia Salvador Brazil Department of Computer Science Federal University of Bahia Fraunhofer Project Center Salvador Brazil Graduate Program in Systems and Computer Salvador University State University of Bahia Salvador Brazil

The adequate monitoring of technical debt (TD) is necessary to increase the chances of success of software projects, which also depend on the effective monitoring of the practitioners involved in their development. Several strategies have been proposed to support TD management. However, the relationship between TD and how much developers contribute to its existence has not yet been investigated. This paper presents a strategy to monitor the contribution of development teams to the TD accumulation in software projects. The proposed strategy was evaluated using the Technology Acceptance Model. The results indicated that the strategy is useful for supporting TD monitoring activities, with 97% of participants indicating that they would use it. In addition, about 76% of respondents agreed that the strategy is user-friendly and allows greater productivity and agility in monitoring the relationship between TD and members of the development team. © 2019 Association for Information Systems. All rights reserved.

关键词： Human resource management

来源：评论

学校读者我要写书评

暂无评论

Improving attention mechanism in graph neural networks via cardinality preservation 29

Improving attention mechanism in graph neural networks via c...

引用

29th International Joint Conference on Artificial Intelligence, IJCAI 2020

作者： Zhang, Shuo Xie, Lei Ph.D. Program in Computer Science Graduate Center City University of New York United States Department of Computer Science Hunter College City University of New York United States Helen and Robert Appel Alzheimer's Disease Research Institute Feil Family Brain and Mind Research Institute Weill Cornell Medicine Cornell University United States

ISBN: (纸本)9780999241165

Graph Neural Networks (GNNs) are powerful for the representation learning of graph-structured data. Most of the GNNs use a message-passing scheme, where the embedding of a node is iteratively updated by aggregating the information from its neighbors. To achieve a better expressive capability of node influences, attention mechanism has grown to be popular to assign trainable weights to the nodes in aggregation. Though the attention-based GNNs have achieved remarkable results in various tasks, a clear understanding of their discriminative capacities is missing. In this work, we present a theoretical analysis of the representational properties of the GNN that adopts the attention mechanism as an aggregator. Our analysis determines all cases when those attention-based GNNs can always fail to distinguish certain distinct structures. Those cases appear due to the ignorance of cardinality information in attention-based aggregation. To improve the performance of attention-based GNNs, we propose cardinality preserved attention (CPA) models that can be applied to any kind of attention mechanisms. Our experiments on node and graph classification confirm our theoretical analysis and show the competitive performance of our CPA models. The code is available online: https://***/zetayue/CPA. © 2020 Inst. Sci. inf., Univ. Defence in Belgrade. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：