检索结果-内蒙古大学图书馆

DBMamba-UNet: dual-branch mamba in vision for 3D hippocampus image segmentation

JOURNAL OF ELECTRONIC imaging 2025年第2期34卷

作者： Yu, Lei Wang, Zunkai Chongqing Normal Univ Sch Comp Sci & Informat Technol Chongqing Peoples R China

Hippocampal segmentation is a crucial task in medical image processing, playing a significant role in the diagnosis and research of neurodegenerative diseases such as Alzheimer's disease. The hippocampus, located in the medial temporal lobe of the human brain, is a bilayer gray matter structure that protrudes into the temporal horn of the lateral ventricle and occupies its medial base. Due to its small size, irregular shape, and indistinct boundaries, accurate segmentation of the hippocampus poses considerable challenges. Currently, most hippocampal segmentation methods rely on magnetic resonance imaging (MRI). However, using structural MRI alone often fails to achieve precise segmentation. To enhance segmentation accuracy, researchers have adopted various approaches. Among these, deep learning-based methods have gained significant attention for their powerful feature extraction and classification capabilities. Nevertheless, these methods tend to confuse edge regions during segmentation, reducing overall accuracy. Transformer-based models, known for capturing long-range dependencies effectively, have also been explored, but their extensive parameter requirements often limit practical applications in segmentation tasks. To address these challenges, we propose a dual-branch Mamba-inspired U-Net (DBMamba-UNet). The dual-branch structure preserves rich detail information, particularly edge features, at high resolutions. Features of different scales from the main and auxiliary branches are fused using the FMamba module before being passed to the upsampling layer. The Mamba-inspired visual architecture, utilized as both encoder and decoder, combines the advantages of transformers and Mamba structures. This design reduces the computational resources required for segmentation tasks while enabling parallel processing. Compared with state-of-the-art algorithms, DBMamba-UNet achieves superior segmentation performance with significantly lower computational costs, delivering bet

关键词： multifeature fusion mamba asymmetric U-Net linear attention semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Radar-based Tomography with Filtered Backprojection Using Attenuation and Time Shift Profiles of a Reference Reflection

Radar-based Tomography with Filtered Backprojection Using At...

引用

International Conference on Electromagnetics in Advanced applications (ICEAA)

作者： Irwin Barengolts Ali Al-Tayar Manuel Funk Kristof Dausien Dennis Pohle Christian Schulz Ilona Rolfes Jan Barowski Institute of Microwave Systems Ruhr University Bochum Germany

ISBN: (数字)9798350360974

ISBN: (纸本)9798350360981

In this paper we present monostatic tomographic imaging using a frequency modulated continuous wave (FMCW) radar system operating from 126 GHz to 182 GHz. Instead of an active receiver, we use a metallic corner reflector that is positioned opposite the radar and behind the test object. Both radar and corner reflector are included in a parallel-beam based measurement setup, consisting of electrical translation stages and a turntable to enable measurements at multiple angular positions. The information for reconstruction was obtained by investigating the reflection of the corner reflector in terms of spatially dependent attenuation and time shift caused by a test object. Here, a 3D-printed cuboid made of polylactic acid (PLA) and a polyvinyl chloride (PVC) pipe were placed in the middle of the measurement setup. The image processing is based on the Fourier slice theorem using the filtered backprojection algorithm, whereby the obtained data from the reference reflection is assumed as a projection separately in case of attenuation or time shift.

关键词： Electric variables measurement Radar x-rays Radar imaging Tomography Position measurement Filtering algorithms Attenuation Reflection image reconstruction

来源：评论

学校读者我要写书评

暂无评论

19th IFIP WG 12.5 International Conference on Artificial Intelligence applications and Innovations, AIAI 2023

19th IFIP WG 12.5 International Conference on Artificial Int...

引用

19th IFIP WG 12.5 International Conference on Artificial Intelligence applications and Innovations, AIAI 2023

ISBN: (纸本)9783031341700

The proceedings contain 129 papers. The special focus in this conference is on Artificial Intelligence applications and Innovations. The topics include: Generating Synthetic Vehicle Speed Records Using LSTM;Measuring the State-Observation-Gap in POMDPs: An Exploration of Observation Confidence and Weighting algorithms;predicting Colour Reflectance with Gradient Boosting and Deep Learning;the Importance of the Current Input in Sequence Modeling;towards Historical Map Analysis Using Deep Learning Techniques;analysis of the Lingering Effects of Covid-19 on Distance Education;exploring the Power of Failed Experiences in Case-Based Reasoning for Improved Decision Making;improving Customer Experience in Call Centers with Intelligent Customer-Agent Pairing;a Sharpe Ratio Based Reward Scheme in Deep Reinforcement Learning for Financial Trading;mind the Gap: Addressing Incompleteness Challenge in Case-Based Reasoning applications;Sentiment Analysis of Tweets on Online Education during COVID-19;image Classification Using Class-Agnostic Object Detection;Ince-PD Model for Parkinson’s Disease Prediction Using MDS-UPDRS I & ii and PDQ-8 Score;medical Knowledge Extraction from Graph-Based Modeling of Electronic Health Records;Predicting ALzheimer's Disease with AI and Brain imaging Data;Pre-trained Model Robustness Against GAN-Based Poisoning Attack in Medical imaging Analysis;semi-supervised Brain Tumor Segmentation Using Diffusion Models;A Methodology for Emergency Calls Severity Prediction: From Pre-processing to BERT-Based Classifiers;a Temporal Metric-Based Efficient Approach to Predict Citation Counts of Scientists;algorithmic Forex Trading Using Q-learning;Comparing Vectorization Techniques, Supervised and Unsupervised Classification Methods for Scientific Publication Categorization in the UNESCO Taxonomy;decoding Customer Behaviour: Relevance of Web and Purchasing Behaviour in Predictive Response Modeling;exploring Pairwise Spatial Relationships for Actions Recognition and

关键词：

来源：评论

学校读者我要写书评

暂无评论

19th IFIP WG 12.5 International Conference on Artificial Intelligence applications and Innovations, AIAI 2023

19th IFIP WG 12.5 International Conference on Artificial Int...

引用

19th IFIP WG 12.5 International Conference on Artificial Intelligence applications and Innovations, AIAI 2023

ISBN: (纸本)9783031341106

关键词：

来源：评论

学校读者我要写书评

暂无评论

19th IFIP WG 12.5 International Conference on Artificial Intelligence applications and Innovations, AIAI 2023

19th IFIP WG 12.5 International Conference on Artificial Int...

引用

19th IFIP WG 12.5 International Conference on Artificial Intelligence applications and Innovations, AIAI 2023

ISBN: (纸本)9783031341069

关键词：

来源：评论

学校读者我要写书评

暂无评论

Accelerated Edge Detection Algorithm for High-Speed applications 9

Accelerated Edge Detection Algorithm for High-Speed Applicat...

引用

9th International Japan-Africa Conference on Electronics, Communications, and Computations, JAC-ECC 2021

作者： Saad, Aya Rafat, Khloud Soltan, Ahmed Darweesh, M. Saeed Nile University School of Engineering and Applied Sciences Giza12588 Egypt Nile University Giza12588 Egypt Nile University Giza12588 Egypt

ISBN: (纸本)9781665482929

Digital image processing (DIP) is a growing field for various applications, such as autonomous vehicles and video surveillance. To improve the performance of DIP systems, image processing algorithms are implemented in hardware rather than software. The idea here is primarily to get a faster system than software imaging or other alternative hardware. Field-programmable gate arrays (FPGAs) have the advantages of parallel processing, low cost, and low power consumption. These semiconductor devices contain many logic blocks that can be programmed to perform everything from basic digital gate-level technology to complex image processing algorithms. This paper provides an enhancement pipeline system architecture using AxI interface to implement image processing algorithms such as Sobel edge detection and mean filters on the Zybo z7 Zynq 7010 board using Verilog HDL language. The system is implemented in a 512×512 image that takes 0.009ms in the processing system. © 2021 IEEE.

关键词： Field programmable gate arrays (FPGA)

来源：评论

学校读者我要写书评

暂无评论

Algorithm of stitching and fusion image in PCB analysis 15

Algorithm of stitching and fusion image in PCB analysis

引用

Conference on Optical Engineering + applications and Optics and Photonics for Information processing xV

作者： Semenishchev, E. Voronin, V. Alepko, A. Urunov, S. Agaian, S. Zelensky, A. Moscow State Technol Univ STANKIN 1a Vadkovsky Moscow 127055 Russia CUNY Dept Comp Sci New York NY 10314 USA

ISBN: (数字)9781510644779

ISBN: (纸本)9781510644779;9781510644762

The article proposes an approach that allows merging a series of images obtained using an electron microscope and fusion data with a camera that records data in the far-infrared spectrum (thermal images). The proposed approach is implemented based on algorithms for stitching images obtained in the visible spectrum. Based on the data on the calibration parameters of the pair cameras, the images obtained by the thermal imaging camera are stitched. The frame overlap is between 30% and 50%. The data received by the thermal imaging camera is noisy and requires primary data processing. To filter thermal imaging images, a multi-criteria method is used in work. The parameters of the method are analyzed in parallel on both pairs of the image. They are used for the subsequent identification of the search boundaries of objects and interframe communication points. For data obtained in the visible spectrum, a simplification algorithm is applied to increase data analysis speed. As test data, we used a combination of images obtained by an electron microscope (a maximum approximation is 300x, the color depth is 8bit, the resolution is 1024 Chi 768) and a thermal imaging camera (image resolution is 320 Chi 240 pixels).

关键词： local features thermal images IR denoising PCB analysis stitching image fusion image

来源：评论

学校读者我要写书评

暂无评论

Human-Centric Multimodal Machine Learning: Recent Advances and Testbed on AI-Based Recruitment

引用

SN Computer Science 2023年第5期4卷 434页

作者： Peña, Alejandro Serna, Ignacio Morales, Aythami Fierrez, Julian Ortega, Alfonso Herrarte, Ainhoa Alcantara, Manuel Ortega-Garcia, Javier Universidad Autónoma de Madrid Madrid 28049 Spain

The presence of decision-making algorithms in society is rapidly increasing nowadays, while concerns about their transparency and the possibility of these algorithms becoming new sources of discrimination are arising. There is a certain consensus about the need to develop AI applications with a Human-Centric approach. Human-Centric Machine Learning needs to be developed based on four main requirements: (i) utility and social good;(ii) privacy and data ownership;(iii) transparency and accountability;and (iv) fairness in AI-driven decision-making processes. All these four Human-Centric requirements are closely related to each other. With the aim of studying how current multimodal algorithms based on heterogeneous sources of information are affected by sensitive elements and inner biases in the data, we propose a fictitious case study focused on automated recruitment: FairCVtest. We train automatic recruitment algorithms using a set of multimodal synthetic profiles including image, text, and structured data, which are consciously scored with gender and racial biases. FairCVtest shows the capacity of the Artificial Intelligence (AI) behind automatic recruitment tools built this way (a common practice in many other application scenarios beyond recruitment) to extract sensitive information from unstructured data and exploit it in combination to data biases in undesirable (unfair) ways. We present an overview of recent works developing techniques capable of removing sensitive information and biases from the decision-making process of deep learning architectures, as well as commonly used databases for fairness research in AI. We demonstrate how learning approaches developed to guarantee privacy in latent spaces can lead to unbiased and fair automatic decision-making process. Our methodology and results show how to generate fairer AI-based tools in general, and in particular fairer automated recruitment systems. © 2023, The Author(s).

关键词： Automated recruitment Bias Biometrics Computer vision Deep learning FairCV Fairness Multimodal Natural language processing

来源：评论

学校读者我要写书评

暂无评论

Jigsaw: A Slice-and-Dice Approach to Non-uniform FFT Acceleration for MRI image Reconstruction 35

Jigsaw: A Slice-and-Dice Approach to Non-uniform FFT Acceler...

引用

35th IEEE International parallel and Distributed processing Symposium (IPDPS)

作者： West, Brendan L. Fessler, Jeffrey A. Wenisch, Thomas F. Univ Michigan Ann Arbor MI 48109 USA

ISBN: (纸本)9781665440660

The Fast Fourier Transform (FFT) is a fundamental algorithm in signal processing;significant efforts have been made to improve its performance using software optimizations and specialized hardware accelerators. Computational imaging modalities, such as MRI, often rely on the Non-uniform Fast Fourier Transform (NuFFT), a variant ()I' the EFT for processing data acquired from non-uniform sampling patterns. The most time-consuming step of the NuFFT algorithm is "gridding," wherein non-uniform samples are interpolated to allow a uniform FFT to be computed over the data. Each non-uniform sample affects a window of non-contiguous memory locations, resulting in poor cache and memory bandwidth utilization. As a result, gridding can account for more than 99.6% of the NuFFT computation time, while the FFT requires less than 0.4%. We present Slice-and-Dice, a novel approach to the NuFFT's gridding step that eliminates the presorting operations required by prior methods and maps more efficiently to hardware. Our GPI! implementation achieves gridding speedups of over 250x and 16x vs prior ***-art CPU and GPU implementations, respectively. We achieve further speedup and energy efficiency gains by implementing Slice-and-Dice in hardware with JIGSAW, a streaming hardware accelerator for non-uniform data gridding. JIGSAW uses stall-free fixed-point pipelines to process M non-uniform samples in approximately M cycles, irrespective of sampling pattern-yielding speedups of over 1500x the CPU baseline and 36x the state-of-the-art GPU implementation, consuming similar to 200 mW power and similar to 12 mm(2) area in 16 nm technology. Slice-and-Dice GPU and JIGSAW ASIC implementations achieve unprecedented end-to-end NuFFT speedups of 8x and 36x compared to the state-of-the-art GPU implementation, respectively.

关键词： Runtime Power demand Fast Fourier transforms Magnetic resonance imaging Software algorithms Graphics processing units Signal processing algorithms

来源：评论

学校读者我要写书评

暂无评论

Radiomics software for breast imaging optimization and simulation studies

引用

PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS 2021年 89卷 114-128页

作者： Marinov, Stoyko Buliev, Ivan Cockmartin, Lesley Bosmans, Hilde Bliznakov, Zhivko Mettivier, Giovanni Russo, Paolo Bliznakova, Kristina Univ Hosp Leuven Dept Radiol Herestr 49 B-3000 Leuven Belgium Mikrosistemi Ltd Varna Bulgaria Katholieke Univ Leuven Dept Imaging & Pathol Med Phys & Qual Assessment Herestr 49 B-3000 Leuven Belgium Med Univ Varna Dept Med Equipment Elect & Informat Technol Healt Varna Bulgaria Univ Napoli Federico II Dipartimento Fis Ettore Pancini Naples Italy Ist Nazl Fis Nucl Sez Napoli Naples Italy

Background and Objective: The development, control and optimisation of new x-ray breast imaging modalities could benefit from a quantitative assessment of the resulting image textures. The aim of this work was to develop a software tool for routine radiomics applications in breast imaging, which will also be available upon request. Methods: The tool (developed in MATLAB) allows image reading, selection of Regions of Interest (ROI), analysis and comparison. Requirements towards the tool also included convenient handling of common medical and simulated images, building and providing a library of commonly applied algorithms and a friendly graphical user interface. Initial set of features and analyses have been selected after a literature search. Being open, the tool can be extended, if necessary. Results: The tool allows semi-automatic extracting of ROIs, calculating and processing a total of 23 different metrics or features in 2D images and/or in 3D image volumes. Computations of the features were verified against computations with other software packages performed with test images. Two case studies illustrate the applicability of the tool - (i) features on a series of 2D 'left' and 'right' CC mammograms acquired on a Siemens Inspiration system were computed and compared, and (ii) evaluation of the suitability of newly proposed and developed breast phantoms for x-ray-based imaging based on reference values from clinical mammography images. Obtained results could steer the further development of the physical breast phantoms. Conclusions: A new image analysis toolbox was realized and can now be used in a multitude of radiomics applications, on both clinical and test images.

关键词： Evaluation of x-ray images Breast images Statistical features Fractal dimension Power law analysis Matrix analyses

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：