检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Lu, Ming Y. Chen, Bowen Williamson, Drew F.K. Chen, Richard J. Liang, Ivy Ding, Tong Jaume, Guillaume Odintsov, Igor Zhang, Andrew Le, Long Phi Gerber, Georg Parwani, Anil V. Mahmood, Faisal Department of Pathology Brigham and Women’s Hospital Harvard Medical School BostonMA United States Department of Pathology Massachusetts General Hospital Harvard Medical School BostonMA United States Cancer Program Broad Institute of Harvard and MIT CambridgeMA United States Cancer Data Science Program Dana-Farber Cancer Institute BostonMA United States Department of Biomedical Informatics Harvard Medical School BostonMA United States CambridgeMA United States Health Sciences and Technology Harvard-MIT CambridgeMA United States Harvard John A. Paulson School of Engineering And Applied Sciences Harvard University CambridgeMA United States Department of Pathology Wexner Medical Center Ohio State University ColumbusOH United States Harvard Data Science Initiative Harvard University CambridgeMA United States

The accelerated adoption of digital pathology and advances in deep learning have enabled the development of powerful models for various pathology tasks across a diverse array of diseases and patient cohorts1–13. However, model training is often difficult due to label scarcity in the medical domain and the model’s usage is limited by the specific task and disease for which it is trained1–3, 14, 15. Additionally, most models in histopathology leverage only image data, a stark contrast to how humans teach each other and reason about histopathologic entities. We introduce CONtrastive learning from Captions for Histopathology (CONCH), a visual-language foundation model developed using diverse sources of histopathology images, biomedical text, and notably over 1.17 million image-caption pairs via task-agnostic pretraining. Evaluated on a suite of 13 diverse benchmarks, CONCH can be transferred to a wide range of downstream tasks involving either or both histopathology images and text, achieving state-of-the-art performance on histology image classification, segmentation, captioning, text-to-image and image-to-text retrieval. CONCH represents a substantial leap over concurrent visual-language pretrained systems for histopathology, with the potential to directly facilitate a wide array of machine learning-based workflows requiring minimal or no further supervised fine-tuning. © 2023, CC BY-NC-ND.

关键词： Pathology

来源：评论

学校读者我要写书评

暂无评论

An Autonomous Mobile Robot Path Planner using Spline Curves and Differential Evolution

An Autonomous Mobile Robot Path Planner using Spline Curves ...

引用

Latin America Congress on Computational Intelligence (LA-CCI)

作者： Elias José de Rezende Freitas Miri Weiss Cohen Frederico Gadelha Guimarães Graduate Program in Electrical Engineering Universidade Federal de Minas Gerais Av. Antônio Carlos 6627 Belo Horizonte MG Brazil Federal Institute of Minas Gerais IFMG Department of Software Engineering Braude College of Engineering Karmiel Israel Department of Electrical Engineering Machine Intelligence and Data Science Laboratory (MINDS) Universidade Federal de Minas Gerais UFMG Belo Horizonte MG Brazil

Robots that operate autonomously require a path planner that ensures the robot reaches the desired position in a safe manner. The robot will be able to achieve greater speeds along the path if the path is smooth, minimizing the effort of the controller. This work presents two new path planners, called DE-Spline and DE-NURBS. These planners provide a smooth path directly by the cost function optimization using the algorithm L-SHADE (Success-History based Adaptive of Differential Evolution using linear population size reduction algorithm) and two path’s parametrization: Ferguson cubic spline and NURBS. Additionally, it ensures that the robot reaches the target in the desired orientation. The planners are compared with a Vector Field-based planner, widely used in robot soccer. Based on the results, it has been found that both planners provide a smoother path along the shortest path, by accounting for the robot’s starting angle and its goal angle. Furthermore, DE-NURBS provides a greater degree of planning freedom than DESpline.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Building Ramps and Pathways to data science in K-12, Minority-Serving Institutions, and Two-Year Colleges

引用

Scatterplot 2024年第1期1卷

作者： Rachel Saidi Helen Burn Ben Baumer Jason Black Debzani Deb Zarek Drozda a Math Statistics and Data Science Department Montgomery College – Rockville Campus Rockville MD USA b Mathematics Department Highline College Des Moines Washington Des Moines WA USA c Statistical and Data Sciences Department Smith College Northampton MA USA d School of Business and Interdisciplinary Center for Creativity and Innovation (ICCI) Florida A&M University Tallahassee FL USA e Computer Science Department Center for Applied Data Science (CADS) Winston-Salem State University Winston-Salem NC USA f Data Science 4 Everyone (DS4E) Program University of Chicago Chicago IL USA

Global demand for data scientists and analysts is on a trajectory of sustained growth, underscoring the need to expand enrollment of students in data science and data analytics courses and programs. This article explores diverse pathways to data science for various groups, including high school students, those in two-year colleges, and students in small colleges and minority-serving institutions. This article identifies opportunities and challenges associated with pursuing data science for students in these sectors, including factors such as building partnerships and awareness of the field and difficulties in establishing articulation agreements between institutions. The authors, representing various organizations participating in a panel discussion at the 2023 data science Leadership Summit of the Academic data science Alliance (ADSA), share insights into efforts to develop nontraditional pathways, serving as a resource for schools, colleges, and organizations contemplating creation of data science programs. The panelists’ combined experience spans work with a broad range of organizations including ADSA, American Mathematics Association of Two-Year Colleges, American Statistical Association, data science for Everyone, the Department of Education, HBCU data science Consortium, Mathematical Association of America, National Academies of science, Engineering, and Medicine, HBCU data science Consortium, National science Foundation, and StatPREP.

关键词： data science pathways access K-12 minority-serving institution two-year college

来源：评论

学校读者我要写书评

暂无评论

EMBEDDING SPACE AUGMENTATION FOR WEAKLY SUPERVISED LEARNING IN WHOLE-SLIDE IMAGES

arXiv

引用

arXiv 2022年

作者： Zaffar, Imaad Jaume, Guillaume Rajpoot, Nasir Mahmood, Faisal Department of Computer Science University College London United Kingdom Department of Pathology Brigham and Women's Hospital Harvard Medical School BostonMA United States Department of Pathology Massachusetts General Hospital Harvard Medical School BostonMA United States Cancer Program Broad Institute of Harvard and MIT CambridgeMA United States Data Science Program Dana-Farber Cancer Institute BostonMA United States Tissue Image Analytics Centre Department of Computer Science University of Warwick Coventry United Kingdom Department of Pathology University Hospitals Coventry and Warwickshire NHS Trust Coventry United Kingdom The Alan Turing Institute London United Kingdom

Multiple Instance Learning (MIL) is a widely employed framework for learning on gigapixel whole-slide images (WSIs) from WSI-level annotations. In most MIL based analytical pipelines for WSI-level analysis, the WSIs are often divided into patches and deep features for patches (i.e., patch embeddings) are extracted prior to training to reduce the overall computational cost and cope with the GPUs' limited RAM. To overcome this limitation, we present EMBAUGMENTER, a data augmentation generative adversarial network (DA-GAN) that can synthesize data augmentations in the embedding space rather than in the pixel space, thereby significantly reducing the computational requirements. Experiments on the SICAPv2 dataset show that our approach outperforms MIL without augmentation and is on par with traditional patch-level augmentation for MIL training while being substantially faster. © 2022, CC BY.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

A call to action to address escalating global threats to academic research

引用

The Innovation 2025年第4期6卷 17-19页

作者： Gaelle Piret Fun Man Fung Josie Fullerton Giuseppe Fico Dmitriy Ponkratov Wenxin Chen Daniela Latorre Kirsty Y.Wan Nima Aghaeepour Jules Welgryn Adeel Razi Patricia Silveyra Ahmet Altun Renata Z.Jurkowska Alice C.Hughes Joy Wolfram INSERM U1205 UGA38400 Saint Martin d’HèresFrance School of Chemistry University College DublinD04 C1P1 BelfieldIreland UCD Geary Institute for Public Policy University College DublinD04 N9Y1 DublinIreland School of Cardiovascular&Metabolic Health University of GlasgowG128TA GlasgowUK European Alliance for Medical and Biological Engineering and Science 3001 LeuvenBelgium Life Supporting Technologies Universidad Politécnica de Madrid28040 MadridSpain Siemens Digital Industries Software B908BG SolihullUK Department of Philosophy of Science and Technology University of Science and Technology of ChinaHefei 230026China Institute of Microbiology ETH Zurich8093 ZurichSwitzerland Living Systems Institute&Department of Mathematics and Statistics University of ExeterEX44QD ExeterUK Department of Anesthesiology Painand Perioperative MedicineDepartment of PediatricsDepartment of Biomedical Data ScienceStanford University School of MedicineStanfordCA 94305USA UniversitéParis Nanterre 92000 NanterreFrance Turner Institute for Brain and Mental Health School of Psychological Sciencesand Monash Biomedical ImagingMonash UniversityClaytonVIC 3800Australia Wellcome Centre for Human Neuroimaging University College LondonWC1N 3AR LondonUK CIFAR Azrieli Global Scholars Program TorontoON M5G 1M1Canada Department of Environmental and Occupational Health Indiana University School of Public Health BloomingtonBloomingtonIN 47408USA Department of Pharmacology Sivas Cumhuriyet UniversitySivas 58140Turkey Division of Biomedicine School of BiosciencesCardiff UniversityCF103AX CardiffUK School of Biological Sciences University of Hong KongHong Kong SAR 999077China Australian Institute for Bioengineering and Nanotechnology The University of QueenslandBrisbaneQLD 4072Australia School of Chemical Engineering The University of QueenslandBrisbaneQLD 4072Australia

This article is a call to action to address escalating threats to scientific progress that affect academic researchers across the *** threats include public mistrust of science,challenges in translating academic research to end-user applications,a disconnect between academics and policymakers,emerging barriers to international collaboration,and a reliance on conventional metrics to evaluate academic *** article presents various calls to action informed by exemplary approaches across the globe that serve as frameworks to drive beneficial transformation for researchers,academic institutions,and society.

关键词： academic research calls action global threats public mistrust sciencechallenges translating academic research disconnect academics policymakers call action translation research

来源：评论

学校读者我要写书评

暂无评论

Weakly Supervised Joint Whole-Slide Segmentation and Classification in Prostate Cancer

arXiv

引用

arXiv 2023年

作者： Pati, Pushpak Jaume, Guillaume Ayadi, Zeineb Thandiackal, Kevin Bozorgtabar, Behzad Gabrani, Maria Goksel, Orcun IBM Research Europe Switzerland Department of Pathology Brigham and Women’s Hospital Harvard Medical School United States Department of Pathology Massachusetts General Hospital Harvard Medical School United States Cancer Program Broad Institute of Harvard MIT United States Data Science Program Dana-Farber Harvard Cancer Center United States Computer-Assisted Applications in Medicine ETH Zurich Switzerland Signal Processing Laboratory 5 EPFL Switzerland EPFL Switzerland Department of Information Technology Uppsala University Sweden

The segmentation and automatic identification of histological regions of diagnostic interest offer a valuable aid to pathologists. However, segmentation methods are hampered by the difficulty of obtaining pixel-level annotations, which are tedious and expensive to obtain for Whole-Slide images (WSI). To remedy this, weakly supervised methods have been developed to exploit the annotations directly available at the image level. However, to our knowledge, none of these techniques is adapted to deal with WSIs. In this paper, we propose WHOLESIGHT, a weakly-supervised method, to simultaneously segment and classify WSIs of arbitrary shapes and sizes. Formally, WHOLESIGHT first constructs a tissue-graph representation of the WSI, where the nodes and edges depict tissue regions and their interactions, respectively. During training, a graph classification head classifies the WSI and produces node-level pseudo labels via post-hoc feature attribution. These pseudo labels are then used to train a node classification head for WSI segmentation. During testing, both heads simultaneously render class prediction and segmentation for an input WSI. We evaluated WHOLESIGHT on three public prostate cancer WSI datasets. Our method achieved state-of-the-art weakly-supervised segmentation performance on all datasets while resulting in better or comparable classification with respect to state-of-the-art weakly-supervised WSI classification methods. Additionally, we quantify the generalization capability of our method in terms of segmentation and classification performance, uncertainty estimation, and model calibration. © 2023, CC BY-NC-ND.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

INTOSIS: Interactive Observation of Smartphone Inferred Symptoms for In-The-Wild data

INTOSIS: Interactive Observation of Smartphone Inferred Symp...

引用

IEEE International Conference on Big data

作者： Hamid Mansoor Walter Gerych Luke Buquicchio Abdulaziz Alajaji Kavin Chandrasekaran Emmanuel Agu Elke Rundensteiner Worcester Polytechnic Institute Worcester MA Data Science Program Worcester Polytechnic Institute Worcester MA

ISBN: (数字)9781728162515

ISBN: (纸本)9781728162522

Current research in passive health monitoring utilizes machine learning methods to infer users' symptoms and health status from smartphone-sensed data, which can be gathered on a large scale. However, reasoning about smartphone-sensed health behaviors should engage health experts who may not want to solely rely on computational approaches as they provide limited insights. We designed and proposed a visualization framework for the INTeractive Observation of Smartphone-Inferred Symptoms (INTOSIS), that supports contextualization of symptomatic days by presenting a holistic picture of complex smartphone data for analysts to find c oncerning b ehavior patterns. For instance, while sedentary behavior caused by the flu is concerning, sedentary behaviors on holidays are non-concerning. INTOSIS visualizes multiple smartphone sensor data channels such as geo-location, app usage and screen usage. It uses visual metaphors to effectively represent the data to help analysts derive important human-understandable spatio-temporal contexts and assign health consequences with corresponding semantic labels. INTOSIS provides timeline visualizations of contextual clues such as screen and app usage at night that support an analyst in being able to reason about and then extract plausible explanations for the occurrence of certain symptoms, such as, sleep problems. We validate INTOSIS with intuitive use cases, using a real-world smartphone-sensed dataset, along with expert evaluation.

关键词： Visual analytics Semantics data visualization Psychology Big data Tools Task analysis

来源：评论

学校读者我要写书评

暂无评论

引用

IEEE International Conference on Big data

作者： Jidapa Thadajarassiri Cansu Sen Thomas Hartvigsen Xiangnan Kong Elke Rundensteiner Data Science Program Worcester Polytechnic Institute Worcester USA Worcester Polytechnic Institute Worcester USA

ISBN: (数字)9781728162515

ISBN: (纸本)9781728162522

Publicly available pre-trained word embeddings are rich sources for turning critical high-dimensional representations of huge text data repositories into meaningful compact vectors essential for text mining applications. With many of such pre-trained embedding sources available, each faces limitations in the appropriateness of their language use for the downstream text-mining tasks. Meta-embeddings aim to tackle this ambiguity challenge by fusing multiple embedding sources into one feature space. However, current meta-embedding methods assume vocabularies across sources are similar or even identical; which unfortunately stands in sharp contrast to the fact that many sources barely overlap. Further, these methods encode a meta-embedding for each word by reconstructing its actual embedding values (word-encoder), while valuable information of relationships (distances) among words within each source are not directly considered. In this work, we instead propose a novel relation-encoder learning approach called Similarity-Preserving Meta-Embedding (SimME) that directly integrates word-pair relationships from partially overlapping embedding sources. SimME embeds words such that their similarities are learned from those observed in multiple pre-trained sources. To handle relations between words that are not present in all sources, we introduce maskout, a new loss term, that steers the learning selectively to the sources containing said relations. SimME consistently outperforms state-of-the-art methods by 10% on average and with up to 20% across several core metrics in 4 popular mining tasks on 23 datasets.

关键词： Text mining Training Vocabulary Transforms Big data Turning Task analysis

来源：评论

学校读者我要写书评

暂无评论

On the computational Bayesian survival spatial Dengue hemorrhagic fever (DHF) modelling with Fernandez–steel skew normal conditional autoregressive (FSSN CAR) frailty

引用

AIP Conference Proceedings 2023年第1期2536卷

作者： Dwi Rantini Nur Iriawan Irhamah Musofa Rusli Data Science Technology Study Program Faculty of Advanced Technology and Multidiscipline Universitas Airlangga Surabaya 60115 Indonesia Department of Statistics Faculty of Science and Data Analytics Institut Teknologi Sepuluh Nopember Surabaya 60111 Indonesia Department of Internal Medicine Faculty of Medicine Universitas Airlangga Surabaya 60115 Indonesia

Generally, infectious disease data has a spatial effect, meaning that areas that are close together affect each other. One of the infectious diseases is dengue hemorrhagic fever (DHF). The recovery time of DHF patients is interesting to study. In this research, DHF data in eastern Surabaya was used. The recovery time of this DHF patient can be modeled using one of the statistical methods, namely survival analysis. This recovery time data follows the Weibull distribution. On the data indicated spatial effect, the conditional autoregressive (CAR) model can be used to express dependencies between adjacent areas. Spatial random effects in the survival model were modeled with Normal CAR, Double-Exponential (DE) CAR, and Fernandez-Steel skew Normal (FSSN) CAR. In this research, Cox regression was used and parameter estimation was performed using the Bayesian analysis with Hamiltonian Monte Carlo (HMC) algorithm using the Stan programming language. Based on the comparison of the Watanabe-Akaike information criterion (WAIC), the spatial random effects on the Weibull Cox regression model are best modeled with the FSSN CAR. This is because the FSSN CAR is able to capture error patterns both symmetrical and asymmetrical, not with Normal CAR and DE CAR which can only capture symmetrical error patterns. In this research, several variables that allegedly affect the recovery rate of DHF patients are given. Then, based on the best model, variables that significantly affect the patient's recovery rate are age, the high schools in last education, housewife in the type of occupation, stadium-II in severity level, fever days before entering the hospital, pulse, temperature, and leukocytes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Corrosion Image data Set for Automating Scientific Assessment of Materials 32

Corrosion Image Data Set for Automating Scientific Assessmen...

引用

32nd British Machine Vision Conference, BMVC 2021

作者： Yin, Biao Josselyn, Nicholas Considine, Thomas Kelley, John Rinderspacher, Berend Jensen, Robert Synder, James Zhang, Ziming Rundensteiner, Elke Data Science Program Worcester Polytechnic Institute WorcesterMA United States Weapons and Materials Research Directorate US Army Research Laboratory Aberdeen Proving GroundMD United States Weapons and Materials Research Directorate ARL Northeast Regional Extended Site BurlingtonMA United States Weapons and Materials Research Directorate DEVCOM Army Research Laboratory AdelphiMD United States

The study of material corrosion is an important research area, with corrosion degradation of metallic structures causing expenses up to 4% of the global domestic product annually along with major safety risks worldwide. Unfortunately, large-scale and timely scientific discovery of materials has been hindered by the lack of standardized corrosion experimental data in the public domain for developing machine learning models. Obtaining such data is challenging due to the expert knowledge and time required to conduct these scientific experiments and assess corrosion levels. We curate a novel data set consisting of 600 images annotated with expert corrosion ratings obtained over 10 years of laboratory corrosion testing by material scientists. Based on this data set, we find that non-experts even when rigorously trained with domain guidelines to rate corrosion fail to match expert ratings. Challenges include limited data, image artifacts, and millimeter-precision corrosion. This motivates us to explore the viability of deep learning approaches to tackle this benchmark classification task. We study (i) convolutional neural networks powered with rich domain-specific image augmentation techniques tuned to our data, and (ii) a recent self-supervised representation learning approach either pretrained on ImageNet or trained on our data. We demonstrate that pretrained ResNet-18 and HRNet models with tuned augmentations can reach up to 0.83 accuracy. With this corrosion data set, we open the door for the design of more advanced deep learning models to support this real-world task, while driving innovative new research to bridge computer vision and material innovation. Our data and code are available at: https://***. © 2021. The copyright of this document resides with its authors.

关键词： Corrosion rate

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：