检索结果-内蒙古大学图书馆

ieee Southwest symposium on Image analysis and Interpretation

作者： Harshit Parmar Eric Walden Texas Tech Neuroimaging Institute Texas Tech University Lubbock USA

T-distributed stochastic neighbor embedding (tSNE) based postprocessing techniques have been used for visualization of temporal information from functional Magnetic Resonance imaging (fMRI) data. With advancements in multiband pulse sequence, finer temporal sampling can be achieved during fMRI data acquisition. The synergy between fast sampled fMRI data and t-SNE can be used to detect intra-trial brain state changes which might correspond to mind wandering or task switching. In this article we propose an intra-trial brains state detection algorithm suitable for multiband fMRI data which is less susceptible to noise and random fluctuations. The proposed approach was tested on a gambling decision-making task to identify the instance of brain state change during the task. Functional connectivity analysis revealed that the switching from decision making phase to reward anticipation phase caused a brain state change which was reliably and repeatedly detected by the proposed algorithm. The proposed technique can be used to validate task switching or quantify the degree of involvement during task trials.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ActiveAnno3D - An Active Learning Framework for Multi-Modal 3D Object Detection

ActiveAnno3D - An Active Learning Framework for Multi-Modal ...

引用

ieee symposium on Intelligent Vehicle

作者： Ahmed Ghita Bjørk Antoniussen Walter Zimmer Ross Greer Christian Creß Andreas Møgelmose Mohan M. Trivedi Alois C. Knoll Chair of Robotics and Artificial Intelligence the Technical University of Munich Laboratory for Intelligent and Safe Automobiles the University of California San Diego Visual Analysis and Perception Lab Aalborg Universitet

ISBN: (数字)9798350348811

ISBN: (纸本)9798350348828

The curation of large-scale datasets is still costly and requires much time and resources. data is often manually labeled, and the challenge of creating high-quality datasets remains. In this work, we fill the research gap using active learning for multi-modal 3D object detection. We propose ActiveAnno3D, an active learning framework to select data samples for labeling that are of maximum informativeness for training. We explore various continuous training methods and integrate the most efficient method regarding computational demand and detection performance. Furthermore, we perform extensive experiments and ablation studies with BEVFusion and PV-RCNN on the nuScenes and TUM Traffic Intersection (TUMTraf-I) dataset. We show that we can achieve almost the same performance with PV-RCNN and the entropy-based query strategy when using only half of the training data (77.25 mAP compared to 83.50 mAP) of the TUMTraf-I dataset. BEVFusion achieved an mAP of 64.31 when using half of the training data and 52.88 mAP when using the complete nuScenes dataset. We integrate our active learning framework into the proAnno labeling tool to enable AI-assisted data selection and labeling and minimize the labeling costs. Finally, we provide code, weights, and visualization results on our website.

关键词： Training Three-dimensional displays Costs Codes Intelligent vehicles Training data data visualization

来源：评论

学校读者我要写书评

暂无评论

Visual linkage and interactive features of Evidente for an enhanced analysis of SNP-based phylogenies

Visual linkage and interactive features of Evidente for an e...

引用

25th Eurographics Conference on visualization, EuroVis 2023

作者： Paz, M. Witte Harbig, T. Varga, D. Kränzle, E. Nieselt, K. University of Tübingen Institute for Bioinformatics and Medical Informatics Tübingen Germany

ISBN: (纸本)9783038682202

Phylogenetic trees of a set of bacterial strains are often used to analyze their evolutionary relationships and they are commonly based on genomic features, such as single nucleotide polymorphisms (SNPs). Evidente - a recently published tool - provides visual and analytical linkage across a phylogenetic tree, SNP data and metadata, and integrates them into one interactive visual analytics platform. In contrast to other approaches, Evidente shows how SNPs agree with the tree structure. Evidente is part of the TueVis server (https://***/). Here, we give an overview of the tasks supported by Evidente. The version of Evidente described in the publication can seamlessly visualize up to 150 strains. Thus, we introduce further enhancements for larger trees, such as data-driven aggregation and semantic zooming. © 2023 The Authors.

关键词： Genome

来源：评论

学校读者我要写书评

暂无评论

Visual analysis of Spatio-Temporal Event Predictions: Investigating the Spread Dynamics of Invasive Species

引用

ieee TRANSACTIONS ON BIG data 2021年第3期7卷 497-509页

作者： Seebacher, Daniel Hauessler, Johannes Hundt, Michael Stein, Manuel Mueller, Hannes Engelke, Ulrich Keim, Daniel A. Univ Konstanz Data Anal & Visualizat Grp D-78464 Constance Germany Fed State Baden Wurttemberg LUBW State Off Environm Measurements & Nat Conservat D-76185 Karlsruhe Germany CSIRO Data61 Canberra ACT 2601 Australia

Invasive species are a major cause of ecological damage and commercial losses. A current problem spreading in North America and Europe is the vinegar fly Drosophila suzukii. Unlike other Drosophila, it infests non-rotting and healthy fruits and is therefore of concern to fruit growers, such as vintners. Consequently, large amounts of data about the occurrence of D. suzukii have been collected in recent years. However, there is a lack of interactive methods to investigate this data. We employ ensemble-based classification to predict areas susceptible to the occurrence of D. suzukii and bring them into a spatio-temporal context using maps and glyph-based visualizations. Following the information-seeking mantra, we provide a visual analysis system Drosophigator for spatio-temporal event predictions, enabling the investigation of the spread dynamics of invasive species. We demonstrate the usefulness of our approach in three use cases and an evaluation with more than 30 domain experts.

关键词： data visualization Big data Temperature measurement visualization Europe Insects Industries Geographic information systems data visualization supervised learning

来源：评论

学校读者我要写书评

暂无评论

Conditional Visuo-Textual Prompt Learning for Medical Image analysis

Conditional Visuo-Textual Prompt Learning for Medical Image ...

引用

ieee International symposium on Biomedical Imaging

作者： Raffaele Mineo Giovanni Bellitto Federica Proietto Salanitri Rutger Hendrix Ulas Bagci Daniela Giordano Concetto Spampinato Simone Palazzo Department of Electrical Electronic and Computer Engineering University of Catania Italy Department of Radiology and BME Northwestern University Chicago IL USA

ISBN: (数字)9798331520526

ISBN: (纸本)9798331520533

The paper presents CTVTL-CLIP, a conditional visuo-textual prompt tuning method designed to adapt large mul-timodal models for highly specialized applications, where the required semantic representations are not initially encoded in the models' latent space. CTVTL-CLIP jointly optimizes CLIP's text and visual encoders using sets of soft prompts that condition the representations on domain-specific knowledge and features, with the support of a lightweight pretrained network to condition textual learnable tokens. This approach enhances the model's ability to accurately understand medical images, aligning textual descriptions with visual representations even in cases with limited annotated data. CTVTL-CLIP was tested on an endoscopy image dataset containing multiple gastric lesions, significantly outperforming traditional classifiers such as CNNs and vision transformers, while using fewer learnable parameters. It was also evaluated on two additional medical image analysis tasks, including skin lesion classification and stenosis classification in angiographies, showing improved performance compared to state-of-the-art methods that require training more parameters. The combination of CTVTL-CLIP's efficiency and superior performance under-scores its practical potential for real-world medical applications, where data scarcity and model efficiency are critical challenges.

关键词： Training visualization Adaptation models Image analysis Biological system modeling Transformers data models Skin Lesions Tuning

来源：评论

学校读者我要写书评

暂无评论

Abstract and Concrete Materials: What to use for visualization Onboarding for a Treemap visualization? 22

Abstract and Concrete Materials: What to use for Visualizati...

引用

15th International symposium on Visual Information Communication and Interaction, VINCI2022

作者： Stoiber, Christina Grassinger, Florian Aigner, Wolfgang St.Pölten University of Applied Sciences Austria

ISBN: (纸本)9781450398077

Visual exploration of large and complex data is becoming increasingly important in different domains. However, non-experts in the field of visual data analysis often have problems with correctly reading and interpreting information from visualization idioms that are new to them. To make new forms of visualizations understandable and interpretable for a broad range of audiences, visualization onboarding methods can support users. However, it is unclear whether concrete or abstract materials yield better results to foster learning. In order to answer this question, we conducted a within-subject study with 40 students to compare abstract and concrete onboarding messages for a treemap visualization. The results show that (1) concrete onboarding messages are more helpful than abstract, whereas the length of the abstract messages is ranked higher;(2) abstract onboarding messages lead to more valuable descriptions;and (3) either concrete or abstract onboarding messages can lead to high valuable insights. © 2022 Owner/Author.

关键词： visualization

来源：评论

学校读者我要写书评

暂无评论

Breast lesion detection and visualization utilizing artificial intelligence and the H-scan

Breast lesion detection and visualization utilizing artifici...

引用

ieee International Ultrasonics symposium (IUS)

作者： Baek, Jihye O'Connell, Avice M. Parker, Kevin J. Univ Rochester Dept Elect & Comp Engn Rochester NY USA Univ Rochester Dept Imaging Sci Med Ctr Rochester NY USA

ISBN: (数字)9781665466578

ISBN: (纸本)9781665466578

W e incorporated raw ultrasound parameters into artificial intelligence-based breast cancer diagnosis to achieve improved accuracy compared to radiologists and deep learning (DL). 76 patients with suspicious breast lesions were ultrasound-imaged using a Samsung RS85 system equipped with a 9.4 MHz center frequency transducer. The patients underwent biopsy, and the biopsy results were used as a reference gold standard: n=53 for benign and n=23 for malignant. Ten radiologists reviewed the ultrasound images and provided BI-RADS (Breast Imaging Reporting and data System) scores. A previously trained DL product with a modified fully convolutional network and GoogLeNet contoured the breast lesion boundaries. Within the contoured lesions, ultrasound parameters were extracted from the radiofrequency, envelope, and log-compressed data: (1) H-scan color level, (2) lesion boundary shape using convex hull, (3) B-mode boundary standard deviation (STD), (4) B-mode STD, and (5) Burr distribution b. To quantify breast condition, multiparametric analysis combining the 5 features was performed using principal component analysis, resulting in the first principal component (PC1). The PC1 outputs within a lesion were overlaid on B-mode images. We calculated the area under the curve (AUC) to evaluate performance. We compared AUC results from radiologists, the DL product, and our PC1 quantification. The PC1 showed the highest AUC. Further, utilizing this PC1 quantification, we visualized the localized probability of malignancy, illustrating BI-RADS score differences using a color display. Overall, we demonstrated the potential of utilizing raw ultrasound parameters to improve DL performance and to achieve higher diagnostic accuracy than radiologists for breast cancer.

关键词： Biophysical ultrasound feature Breast cancer diagnosis Machine learning Multiparametric analysis Multiparametric imaging Tissue characterization

来源：评论

学校读者我要写书评

暂无评论

Mitigating CH4 Emission From Hydroelectric Reservoirs Using GEE And G-res: Case Study From A Tropical Reservoir 3

Mitigating CH4 Emission From Hydroelectric Reservoirs Using ...

引用

3rd ieee India Geoscience and Remote Sensing symposium, InGARSS 2023

作者： Sarkar, Kounik De Ghosh, Surajit Bhattacharyya, Soumya Chowdhury, Anuva National Institute of Technology Durgapur Department of Earth and Environmental Studies Durgapur India International Water Management Institute Water Data Science Research Group Colombo Sri Lanka National Institute of Technology Durgapur Department of Civil Engineering Durgapur India Birla Institute of Technology Mesra Department of Remote Sensing Ranchi India

ISBN: (纸本)9798350325591

In spite of being known as a cleaner source of energy, large hydropower reservoirs are found that they emit a considerable amount of GHGs. Quantifying and mitigating the impact of GHGs are essential as they contribute to global warming. Cloud-based tools like GEE and G-res, Earth Observation data, and in-situ reservoir water level have been used to quantify methane (CH4) emissions from Maithon reservoir, Jharkhand, India. Mitigation techniques and developing an alternative energy source from the emission were also discussed. Initial analysis showed that a 30 cm increment in operating level could reduce CH4 degassing emissions by 5.87±1.43%. The present workflow can be used for preparing emission baselines from large reservoirs to strengthen hydropower policy to accomplish the SDGs. © 2023 ieee.

关键词： Global warming

来源：评论

学校读者我要写书评

暂无评论

Research hotspots and trends of Rural revitalization assisted by Farmers' cooperatives -CiteSpace analysis based on 2097 articles 21

Research hotspots and trends of Rural revitalization assiste...

引用

21st International symposium on Distributed Computing and Applications for Business Engineering and Science, DCABES 2022

作者： Wang, Chenyu Lu, Qian School of Business Guilin University of Electronic Technology Guilin China Business School Guilin University of Electronic Technology Guilin China

ISBN: (纸本)9781665454629

In order to clarify the current situation of rural revitalization assisted by farmers' cooperatives, dig out the research hotspot of rural revitalization assisted by farmers' cooperatives, and explore the trend of rural revitalization assisted by farmers' cooperatives, the paper carries out visual analysis of the literature in this field, in order to provide reference for further research. CiteSpacevisualization software was used to analyze the literatures collected by CNKI from 2000 to 2022. The results showed that the authors showed the characteristics of 'large dispersion and smallconcentration'. The research strength of 'cooperatives and rural revitalization' is not evenly distributed, and the overall trend is biased towards the central and southern regions of China. The backbone of research is mainly research institutes and universities, and the degree of cooperation between institutions needs to be improved. Through the keywords co-occurrence network visualization map, four research hotspots are obtained, namely, rural revitalization, farmer cooperatives, rural finance, agricultural and rural modernization. Industrial integration and industrial poverty alleviation have become new emerging directions. © 2022 ieee.

关键词： Agriculture

来源：评论

学校读者我要写书评

暂无评论

Saliency Prediction in the Deep Learning Era: Successes and Limitations

引用

ieee TRANSACTIONS ON PATTERN analysis AND MACHINE INTELLIGENCE 2021年第2期43卷 679-700页

作者： Borji, Ali MarkableAI Inc Brooklyn NY 11201 USA

Visual saliency models have enjoyed a big leap in performance in recent years, thanks to advances in deep learning and large scale annotated data. Despite enormous effort and huge breakthroughs, however, models still fall short in reaching human-level accuracy. In this work, I explore the landscape of the field emphasizing on new deep saliency models, benchmarks, and datasets. A large number of image and video saliency models are reviewed and compared over two image benchmarks and two large scale video datasets. Further, I identify factors that contribute to the gap between models and humans and discuss the remaining issues that need to be addressed to build the next generation of more powerful saliency models. Some specific questions that are addressed include: in what ways current models fail, how to remedy them, what can be learned from cognitive studies of attention, how explicit saliency judgments relate to fixations, how to conduct fair model comparison, and what are the emerging applications of saliency models.

关键词： Predictive models Computational modeling Benchmark testing data models visualization Deep learning Task analysis Visual saliency eye movement prediction attention video saliency benchmark deep learning

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：