检索结果-内蒙古大学图书馆

22nd IEEE International Symposium on Biomedical Imaging, ISBI 2025

作者： Ravishankar, Hariharan Paluru, Naveen Sudhakar, Prasad Yalavarthy, Phaneendra K. Indian Institute of Science Department of Computational and Data Sciences Bangalore India Science and Technology Organization Ge HealthCare Bangalore India

ISBN: (纸本)9798331520526

Spectral-Domain Optical Coherence Tomography (SDOCT) is the most common and cost-effective variant of OCT for diagnosing retinal diseases. Deep learning models that are utilized for automated diagnosis of these images often exhibit poor performance - across data-sets with varying acquisition settings, limiting their clinical utility. This work, for the first time proposes 'Inference Time Adaptation (ITA)' with lightweight deep learning models for SDOCT analysis, enhancing model deployability and generalizability. Furthermore, this work extends ITA, introducing 'Inference Time Adaptation with Domain Transforms (ITADT)' to improve the generalizability of deep learning models for retinal disease diagnosis. Domain specific augmentations are applied to test samples to enable regularized model adaptation, enhancing classification performance. The proposed approach is model agnostic and can be applied to any trained deep learning model. The proposed methods achieve significant performance improvement of > 10% on average in accuracy, compared with the un-adapted/original deep learning model on benchmark data-sets for classification of retinal diseases. © 2025 IEEE.

关键词： Deep Learning Generalizability Inference Time Adaptation Optical Coherence Tomography Retinal Diseases

来源：评论

学校读者我要写书评

暂无评论

Display Field-of-View Agnostic Robust Ct Kernel Synthesis Using Model-Based Deep Learning 22

Display Field-of-View Agnostic Robust Ct Kernel Synthesis Us...

引用

22nd IEEE International Symposium on Biomedical Imaging, ISBI 2025

作者： Aggarwal, Hemant Kumar Jerald, Antony Yalavarthy, Phaneendra K. Langoju, Rajesh Das, Bipul Science and Technology Organization Ge HealthCare Bangalore India Indian Institute of Science Department of Computational and Data Sciences Bangalore India

ISBN: (纸本)9798331520526

In X-ray computed tomography (CT) imaging, the choice of reconstruction kernel is crucial as it significantly impacts the quality of clinical images. Different kernels influence spatial resolution, image noise, and contrast in various ways. Clinical applications involving lung imaging often require images reconstructed with both soft and sharp kernels. The reconstruction of images with different kernels require raw sinogram data and storing images for all kernels increases processing time and storage requirements. The Display Field-of-View (DFOV) adds complexity to kernel synthesis, as data acquired at different DFOVs exhibit varying levels of sharpness and details. This work introduces an efficient, DFOV-agnostic solution for image-based kernel synthesis using model-based deep learning. The proposed method explicitly integrates CT kernel and DFOV characteristics into the forward model. Experimental results on clinical data, along with quantitative analysis of the estimated modulation transfer function using wire phantom data, clearly demonstrate the utility of the proposed method in real time. Additionally, a comparative study with a direct learning network, that lacks forward model information, shows that the proposed method is more robust to DFOV variations. © 2025 IEEE.

关键词： Kernel Synthesis Model-based Deep Learning X-ray Computed Tomography

来源：评论

学校读者我要写书评

暂无评论

Efficient Modulo Multiplier

Electronics

引用

Electronics 2023年第1期27卷 18-24页

作者： Ahmed, Rekib Uddin Thabah, Sheba Diamond Haque, Mridul Saha, Prabir Department of Electronics and Communication Engineering National Institute of Technology Meghalaya Shillong793003 India Department of Electrical Engineering Indian Institute of Technology Bombay Mumbai India IBM Systems Bangalore India Department of Computational and Data Science Indian Institute of Science Bangalore560012 India

The paper presents the methodology to compute modulo multiplication with the moduli set 2n, 2n−1, 2n+1. In addition to this, designs of the modulo multipliers, namely 2n, 2n−1, and 2n+1 (with n = 4, 8, and 16), have been proposed which are based on half adders, full adders, 4:3 compressor, 7:3 compressor, and the multi-column compressor namely 5,5:4. The gate level design of 4:3 compressor is carried out by solving the truth table using the K-map reduction. To verify the functionalities we have implemented the proposed modulo multipliers using VHDL coding in Xilinx 14.2 design suite. Simulation using Virtex-6 device has been performed to estimate delay, power consumption, and power-delay product (PDP). Moreover, the modulo multipliers are simulated in Cadence RC compiler using 0.18 μm technology to estimate the area. One of the major contributions to the arts of this work is in the partial product reduction stage which utilizes the multi-column 5,5:4 compressor to reduce power and area. The modulo 2n−1 multiplier of operand size 4-bit shows an improvement of 66.34% in terms of area over the best-reported paper. On the other hand, the modulo 2n+1 multiplier of operand size 4-bit shows an improvement of 58.59% terms of in area and the same of operand size 8-bit shows an improvement of 22.72% over the best-reported paper. The proposed algorithms of moduli multiplication are applicable to Booth multiplication of signed numbers. © 2023, Electronics. All Rights Reserved.

关键词： Adders

来源：评论

学校读者我要写书评

暂无评论

Transfer Learning and Fine-tuned CNN Architecture for Dog Breed Classification 3

Transfer Learning and Fine-tuned CNN Architecture for Dog Br...

引用

3rd International Conference on Artificial Intelligence For Internet of Things, AIIoT 2024

作者： Gunaranjan, K. Vijay, K. Naveen, P. St. Joseph's College of Engineering Department of Artificial Intelligence and Data Science Chennai India SRM institute of science and technology Department of computational intelligence Chennai India

ISBN: (纸本)9798350372120

This study presents a comprehensive approach to dog breed classification, leveraging transfer learning algorithms and Convolutional Neural Networks (CNNs). The dataset utilized for training and evaluation is sourced from the Dog Breed Identification dataset, encompassing 120 distinct breeds. A pre-trained CNN model is employed, and through the application of transfer learning, the proposed models achieve an impressive accuracy rate of 99%. The primary objective is to enable users to submit images of dogs, with the model accurately identifying the specific breed from the diverse dataset. The research underscores the efficacy of transfer learning and pretrained models in achieving exceptional accuracy for the nuanced task of dog breed classification, contributing to the advancements in image recognition within the context of diverse canine breeds. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Emotion Detection Mastery: EEG Signal Analysis with Balanced vs. Unbalanced data

Emotion Detection Mastery: EEG Signal Analysis with Balanced...

引用

2024 International Conference on Brain Computer Interface and Healthcare Technologies, iCon-BCIHT 2024

作者： Menaka, Jadov Sreelekshmi, G.I. Poulose, Alwin School of Data Science Indian Institute of Science Education and Research Kerala Thiruvananthapuram695551 India University of Kerala Department of Computational Biology and Bioinformatics Kerala Thiruvananthapuram695581 India

ISBN: (纸本)9798331540067

Emotion is crucial in daily life, influencing stress levels, decision-making, relationships, and overall mental health. Recognizing emotions accurately is valuable across sectors like healthcare, education, and beyond. This paper presents a comparative study on emotion recognition, applying various deep learning models with the DEAP dataset, which includes EEG recordings from 32 participants who watched 40 short music video clips. We extracted time-domain, frequency-domain, and wavelet features from the preprocessed EEG data to capture nuanced patterns within the signals. The models assessed in this study include convolutional neural networks (CNN), long short-term memory networks (LSTM), gated recurrent units (GRU), and hybrid architectures, each trained on both balanced and unbalanced datasets. To address class imbalance, data was augmented using the Synthetic Minority Oversampling Technique (SMOTE). Results indicate that the Attention GRU model achieved slightly higher performance by utilizing its attention mechanism to capture extended dependencies within the data. Additionally, model accuracy improved by 5-9% when trained on balanced datasets, underscoring the importance of data balancing in enhancing classification outcomes. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Expressivity and Generalization: Fragment-Biases for Molecular GNNs 41

Expressivity and Generalization: Fragment-Biases for Molecul...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Wollschläger, Tom Kemper, Niklas Hetzel, Leon Sommer, Johanna Günnemann, Stephan School of Computation Information & Technology Technical University of Munich Germany Munich Data Science Institute Germany Helmholtz Center for Computational Health Munich Germany

Although recent advances in higher-order Graph Neural Networks (GNNs) improve the theoretical expressiveness and molecular property predictive performance, they often fall short of the empirical performance of models that explicitly use fragment information as inductive bias. However, for these approaches, there exists no theoretic expressivity study. In this work, we propose the Fragment-WL test, an extension to the well-known Weisfeiler & Leman (WL) test, which enables the theoretic analysis of these fragment-biased GNNs. Building on the insights gained from the Fragment-WL test, we develop a new GNN architecture and a fragmentation with infinite vocabulary that significantly boosts expressiveness. We show the effectiveness of our model on synthetic and real-world data where we outperform all GNNs on Peptides and have 12% lower error than all GNNs on ZINC and 34% lower error than other fragment-biased models. Furthermore, we show that our model exhibits superior generalization capabilities compared to the latest transformer-based architectures, positioning it as a robust solution for a range of molecular modeling tasks. Copyright 2024 by the author(s)

关键词： Molecular docking

来源：评论

学校读者我要写书评

暂无评论

Cancellable biometrics based on the index-of-maximum hashing with random sparse binary encoding

引用

Multimedia Tools and Applications 2024年第21期83卷 59915-59942页

作者： Kim, Jihyeon Park, Jaewoo Low, Cheng Yaw Teoh, Andrew Beng Jin School of Electrical and Electronic Engineering Yonsei University Seoul Korea Republic of AiV Co 89 Nambusunhwan-ro 356-gil Seocho-gu Seoul Korea Republic of Center for Mathematical and Computational Sciences Data Science Group Institute for Basic Science Daejeon Korea Republic of

The wide deployment of biometrics has prompted enormous security and privacy concerns regarding biometric template protection. Cancellable biometrics is one of the most representative countermeasures that maps a biometric feature vector into its protected template non-invertible yet replaceable when the previously transformed template is compromised. This paper proposes a two-stage cancellable biometric scheme on top of the Index-of-Maximum (IoM) hashing, dubbed Random Sparse Binary Encoded IoM (RSBE-IoM) hashing. Unlike the vanilla IoM hashing that operates only on the first-ranked feature indices, the RSBE-IoM hashing employs the first and the second-ranked features for additional discriminability gain. We also devise an instance-specific Random Sparse Binary (RSB) encoder to project the IoM-hashed vectors (both first and second-ranked) into a random sparse binary bitstring. We also introduce an instance-specific seeding mechanism to the RSB encoder such that a one-time seed is generated on the fly and not physically cached upon enrolment. We comprehensively evaluate the proposed RSBE-IoM scheme with six fingerprint benchmark datasets, i.e., DB 1 to 3 for each FVC 2002 and FVC 2004. In addition to accuracy performance, our discussions cover other important perspectives, including computational cost, biometric template protection criteria, and privacy and security concerns. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2023.

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

Advances in an r-adaptive high-order method for high-speed flows: multi-domain coupling and uncertainty quantification

Advances in an r-adaptive high-order method for high-speed f...

引用

AIAA science and Technology Forum and Exposition, AIAA SciTech Forum 2025

作者： Van Heyningen, R. Loek Nguyen, Ngoc Cuong Peraire, Jaime Blonigan, Patrick Department of Aeronautics and Astronautics Massachusetts Institute of Technology CambridgeMA02139 United States Computational Data Science Sandia National Laboratories LivermoreCA United States

ISBN: (数字)9781624107238

ISBN: (纸本)9781624107238

This work presents an adaptive high-order method for the simulation of parametrized high-speed flows. To obviate the proliferation of degrees of freedom inherent in the discontinuous Galerkin discretization chosen here, we use a hybridized method that decreases the size of the global system along with r-adaptive meshing. For efficiency and robustness, we use a strategy of artificial viscosity continuation, wherein the amount of added dissipation is ramped down until physicality and smoothness constraints are violated. We demonstrate this procedure on problems with greater physical complexity than considered in prior works by incorporating coupled aerothermal interactions between fluid and solid domains. In addition, we have previously found that this choice of mesh adaptation method aids in the creation of surrogate models for flows with shocks. We evaluate whether these advantages persist for viscous flows with multiphysics couplings, demonstrating the use of this method for the forward propagation of uncertain boundary conditions for high-speed flows. © 2025, American institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved.

关键词： Viscous flow

来源：评论

学校读者我要写书评

暂无评论

Unusual metallic state in superconducting A15-type La4H23

引用

National science Review 2024年第12期11卷 94-103页

作者： Jianning Guo Dmitrii Semenok Grigoriy Shutov Di Zhou Su Chen Yulong Wang Kexin Zhang Xinyue Wu Sven Luther Toni Helm Xiaoli Huang Tian Cui State Key Laboratory of Superhard Materials College of Physics Jilin University Center for High Pressure Science and Technology Advanced Research (HPSTAR) Department of High-Temperature Superconductivity Skolkovo Institute of Science and Technology Skolkovo Innovation CenterComputational and Data Science and Engineering HochfeldMagnetlabor Dresden(HLD-EMFL) and Würzburg-Dresden Cluster of Excellence Helmholtz-Zentrum Dresden-Rossendorf(HZDR) School of Physical Science and Technology Ningbo University

Hydride superconductors continue to fascinate the communities of condensed matter physics and material scientists because they host the promising near room-temperature superconductivity. Current research has concentrated on the new hydride superconductors with the enhancement of the superconducting transition temperature（Tc）. The multiple extreme conditions（high pressure/temperature and magnetic field） will introduce new insights into hydride superconductors. The study of transport properties under very high magnetic fields facilitates the understanding of superconductivity in conventional hydride *** the present work, we report experimental evidence of an unusual metal state in a newly synthesized cubic A15-type La4H23that exhibits superconductivity with a Tc reaching 105 K at 118 GPa. A large negative magnetoresistance is observed in strong pulsed magnetic fields in the non-superconducting state of this compound below 40 K. Moreover, we construct the full magnetic phase diagram of La4H23up to 68 T at high pressure. The present work reveals anomalous electronic structural properties of A15-La4H23under high magnetic fields, and therefore has great importance with regard to advancing the understanding of quantum transport behaviors in hydride superconductors.

关键词： high-pressure superconductivity hydrides negative magnetoresistance

来源：评论

学校读者我要写书评

暂无评论

Learning predictable and robust neural representations by straightening image sequences 38

Learning predictable and robust neural representations by st...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Niu, Xueyan Savin, Cristina Simoncelli, Eero P. Center for Neural Science New York University United States Center for Data Science New York University United States Center for Computational Neuroscience Flatiron Institute United States

Prediction is a fundamental capability of all living organisms, and has been proposed as an objective for learning sensory representations. Recent work demonstrates that in primate visual systems, prediction is facilitated by neural representations that follow straighter temporal trajectories than their initial photoreceptor encoding, which allows for prediction by linear extrapolation. Inspired by these experimental findings, we develop a self-supervised learning (SSL) objective that explicitly quantifies and promotes straightening. We demonstrate the power of this objective in training deep feedforward neural networks on smoothly-rendered synthetic image sequences that mimic commonly-occurring properties of natural videos. The learned model contains neural embeddings that are predictive, but also factorize the geometric, photometric, and semantic attributes of objects. The representations also prove more robust to noise and adversarial attacks compared to previous SSL methods that optimize for invariance to random augmentations. Moreover, these beneficial properties can be transferred to other training procedures by using the straightening objective as a regularizer, suggesting a broader utility of straightening as a principle for robust unsupervised learning. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

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：