检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Wu, Xinxing Peng, Chong Charnigo, Richard Cheng, Qiang Institute for Biomedical Informatics University of Kentucky LexingtonKY40506 United States Department of Computer Science and Engineering Qingdao University Shandong266071 China Department of Statistics University of Kentucky LexingtonKY40536 United States

Interpreting critical variables involved in complex biological processes related to survival time can help understand prediction from survival models, evaluate treatment efficacy, and develop new therapies for patients. Currently, the predictive results of deep learning (DL)-based models are better than or as good as standard survival methods, they are often disregarded because of their lack of transparency and little interpretability, which is crucial to their adoption in clinical applications. In this paper, we introduce a novel, easily deployable approach, called EXplainable CEnsored Learning (EXCEL), to iteratively exploit critical variables and simultaneously implement (DL) model training based on these variables. First, on a toy dataset, we illustrate the principle of EXCEL;then, we mathematically analyze our proposed method, and we derive and prove tight generalization error bounds;next, on two semi-synthetic datasets, we show that EXCEL has good anti-noise ability and stability;finally, we apply EXCEL to a variety of real-world survival datasets including clinical data and genetic data, demonstrating that EXCEL can effectively identify critical features and achieve performance on par with or better than the original models. It is worth pointing out that EXCEL is flexibly deployed in existing or emerging models for explainable survival data in the presence of right censoring. Copyright © 2022, The Authors. All rights reserved.

关键词： Toys

来源：评论

学校读者我要写书评

暂无评论

Implementing Trust in Non-Small Cell Lung Cancer Diagnosis with a Conformalized Uncertainty-Aware AI Framework in Whole-Slide Images

arXiv

引用

arXiv 2024年

作者： Zhang, Xiaoge Wang, Tao Yan, Chao Najdawi, Fedaa Zhou, Kai Ma, Yuan Cheung, Yiu-Ming Malin, Bradley A. Department of Industrial and Systems Engineering The Hong Kong Polytechnic University Kowloon Hong Kong Department of Biomedical Informatics Vanderbilt University Medical Center NashvilleTN United States Department of Pathology Microbiology and Immunology Vanderbilt University Medical Center NashvilleTN United States Department of Computing The Hong Kong Polytechnic University Kowloon Hong Kong Department of Mechanical Engineering Research Institute for Intelligent Wearable Systems The Hong Kong Polytechnic University Kowloon Hong Kong Department of Computer Science Hong Kong Baptist University Kowloon Tong Kowloon Hong Kong Department of Computer Science Vanderbilt University NashvilleTN United States Department of Biostatistics Vanderbilt University Medical Center NashvilleTN United States

Ensuring trustworthiness is fundamental to the development of artificial intelligence (AI) that is considered societally responsible, particularly in cancer diagnostics, where a misdiagnosis can have dire consequences. Current digital pathology AI models lack systematic solutions to address trustworthiness concerns arising from model limitations and data discrepancies between model deployment and development environments. To address this issue, we developed TRUECAM, a framework designed to ensure both data and model trustworthiness in non-small cell lung cancer subtyping with whole-slide images. TRUECAM integrates 1) a spectral-normalized neural Gaussian process for identifying out-of-scope inputs and 2) an ambiguity-guided elimination of tiles to filter out highly ambiguous regions, addressing data trustworthiness, as well as 3) conformal prediction to ensure controlled error rates. We systematically evaluated the framework across multiple large-scale cancer datasets, leveraging both task-specific and foundation models, illustrate that an AI model wrapped with TRUECAM significantly outperforms models that lack such guidance, in terms of classification accuracy, robustness, interpretability, and data efficiency, while also achieving improvements in fairness. These findings highlight TRUECAM as a versatile wrapper framework for digital pathology AI models with diverse architectural designs, promoting their responsible and effective applications in real-world settings. © 2024, CC BY.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

QoE-Aware Edge-Assisted Machine Learning-Based Fall Detection and Prediction with FBGs

QoE-Aware Edge-Assisted Machine Learning-Based Fall Detectio...

引用

IEEE International Conference on Communications (ICC)

作者： Matilde Rocha Hao Ran Chi Nélia Alberto Paulo André Paulo Antunes Ayman Radwan M. Fátima Domingues Instituto de Telecomunicações and University of Aveiro Aveiro Portugal Department of Electrical and Computer Engineering Instituto de Telecomunicações Instituto Superior Técnico University of Lisbon Lisbon Portugal I3N Department of Physics University of Aveiro Aveiro Portugal Department of Electronics Telecommunications and Informatics University of Aveiro Aveiro Portugal Department of Biomedical Engineering Khalifa University Abu Dhabi United Arab Emirates

Considering that fall accidents are one of the leading causes of non-natural death of elders, it is crucial to design and to implement home’ fall detection systems. Current home monitoring systems are targeting this challenge, pursuing non-invasive, low latency, and simplified fall detection algorithms. Therefore, in this paper, edge-enabled non-wearable and non-invasive fall detection system is proposed. Concretely, outperforming the conventional invasive/privacy-sensitive fall detection technologies, the proposed system comprises four photonic-based accelerometers solely relying on the fiber Bragg grating (FBG) technology, which monitor the vibrations induced by the body impact in the platform by the Bragg wavelength shifts. A newly-developed support vector machine-based multi-class fall detection algorithm is proposed, based on the data collected by the accelerometers. Moreover, feasibility analysis of the proposed fall detection algorithm also reveals the possibility of fall prediction, given the slipping as the pre-falling phenomenon. Experimental results showcase that the proposed fall detection algorithm achieves overall accuracy up to 96.5%, with average processing time achieved as 21.3 ms, indicating the sufficiency to provide high quality of experience (QoE) fall detection services. Besides, fall prediction based on the pre-falling case study of slipping is discussed, revealing that fall can be predicted~197.5 ms beforehand, which is sufficient for further fall prevention (e.g., airbag).

关键词：

来源：评论

学校读者我要写书评

暂无评论

Question answering system for factoid questions about COVID-19 with natural language processing approach

引用

AIP Conference Proceedings 2024年第1期3026卷

作者： Indriati Indriati Randy Cahya Wihandika Putra Pandu Adikara Barlian Henryranu Prasetio Lailil Muflikhah Informatics Engineering Department Faculty of Computer Science Brawijaya University Veteran No.8 Malang-Indonesia 65145

In March 2020, Indonesia faced fast-spreading COVID-19 Disease. In this regard, self-isolation serves as a solution for patients with mild symptoms due to the government’s limited isolation facilities. Self-isolation is different from COVID-19 treatment in the hospital, where patients are fully treated by health professionals. Information clarity is pivotal amid the pandemics due to a vast number of misleading information. Addressing this issue, it is important to keep self-isolation patients receiving adequate and accurate information about COVID-19. Question Answering System (QAS) with Natural Language Processing emerges as one of the solutions to provide information for users. Named Entity Recognition is one of the methods to search for the answer for factoid questions, allowing users to obtain accurate answers. A number of question types were set as keywords with 50% to 72% accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DISENTANGLING THE PROSODY AND SEMANTIC INFORMATION WITH PRE-TRAINED MODEL FOR IN-CONTEXT LEARNING BASED ZERO-SHOT VOICE CONVERSION

arXiv

引用

arXiv 2024年

作者： Chen, Zhengyang Wang, Shuai Zhang, Mingyang Liu, Xuechen Yamagishi, Junichi Qian, Yanmin Auditory Cognition and Computational Acoustics Lab MoE Key Lab of Artificial Intelligence AI Institute Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China Shenzhen Research Institute of Big Data Shenzhen China Chinese University of Hong Kong Shenzhen China National Institute of Informatics Tokyo Japan

Voice conversion (VC) aims to modify the speaker's timbre while retaining speech content. Previous approaches have tokenized the outputs from self-supervised into semantic tokens, facilitating disentanglement of speech content information. Recently, in-context learning (ICL) has emerged in text-to-speech (TTS) systems for effectively modeling specific characteristics such as timbre through context conditioning. This paper proposes an ICL capability enhanced VC system (ICL-VC) employing a mask and reconstruction training strategy based on flow-matching generative models. Augmented with semantic tokens, our experiments on the LibriTTS dataset demonstrate that ICL-VC improves speaker similarity. Additionally, we find that k-means is a versatile tokenization method applicable to various pre-trained models. However, the ICL-VC system faces challenges in preserving the prosody of the source speech. To mitigate this issue, we propose incorporating prosody embeddings extracted from a pre-trained emotion recognition model into our system. Integration of prosody embeddings notably enhances the system's capability to preserve source speech prosody, as validated on the Emotional Speech Database. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Quantum advantage for noisy channel discrimination

引用

Physical Review A 2022年第3期105卷 032401-032401页

作者： Zane M. Rossi Jeffery Yu Isaac L. Chuang Sho Sugiura Department of Physics Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Department of Physics Department of Electrical Engineering and Computer Science and Co-Design Center for Quantum Advantage Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Physics and Informatics Laboratory NTT Research Inc. 940 Stewart Dr. Sunnyvale California 94085 USA

Many quantum mechanical experiments can be viewed as multiround interactive protocols between known quantum circuits and an unknown quantum process. Fully quantum “coherent” access to the unknown process is known to provide an advantage in many discrimination tasks compared to when only incoherent access is permitted, but it is unclear if this advantage persists when the process is noisy. Here, we show that a quantum advantage can be maintained when distinguishing between two noisy single-qubit rotation channels. Numerical and analytical calculations reveal a distinct transition between the performance of fully coherent and fully incoherent protocols as a function of noise strength. Moreover, the size of the region of coherent quantum advantage shrinks inverse polynomially in the number of channel uses, and in an intermediate regime an improved strategy is a hybrid of fully coherent and fully incoherent subroutines. The fully coherent protocol is based on quantum signal processing, suggesting a generalizable algorithmic framework for the study of quantum advantage in the presence of realistic noise.

关键词： Quantum algorithms Quantum channels Quantum computation Quantum information processing Quantum sensing Quantum simulation

来源：评论

学校读者我要写书评

暂无评论

Prediction of Stroke Risk Within 7-Years Follow Up Using Machine Learning Models*

Prediction of Stroke Risk Within 7-Years Follow Up Using Mac...

引用

Data Science and engineering in Healthcare, Medicine and Biology, IEEE EMBS Special Topic Conference on

作者： Konstantina-Helen Tsarapatsani Antonis Sakellarios Vasilis D Tsakanikas Henrik Rudolf Hans J. Trampisch Vasileios C. Pezoulas George K. Matsopoulos Dimitrios I. Fotiadis Biomedical Research Institute Foundation for Research and Technology-Hellas (FORTH) Dept. of Materials Science and Engineering Unit of Medical Technology and Intelligent Information Systems University of Ioannina GR 45110 Ioannina Greece University Medical Center Rostock Institute for Biostatistics and Informatics in Medicine and Ageing Research Department of Medical Informatics Biometry and Epidemiology Ruhr University Bochum Bochum Germany School of Electrical and Computer Engineering National Technical University of Athens Athens Greece

A stroke, also known as brain attack, occurs when blood supply to your brain is interrupted. Primary prevention relies on prompt prediction of a stroke. While currently there are several clinical risk scores, machine learning (ML) models seems to be more suitable tools for accurate prediction of stroke events. Therefore, this work focuses on the prediction of stroke within 7 years follow-up in patients who have not suffered from a stroke or TIA event at baseline. LightGBM (LGBM), Extreme Grading Boosting (XGBoost), Support Vector Machine (SVM) and Decision Tree were employed in the getABI dataset, which includes 5,897 participants. The performance of models was calculated by Accuracy (ACC), Sensitivity (SENS), Specificity (SPE) and area under the receiver operating characteristic curve (AUC) of each model. According to the comparison analysis’s results, LGBM has been shown to be the most trustworthy algorithm, with accuracy 68 %. Moreover, sex, age, status of peripheral artery disease (PAD), history of myocardial infarction, angina pectoris, amputation and diabetes and pulse status of different arteries can be used as a simple and cost-effective way to predict *** Relevance: A fatal medical emergency, stroke may be anticipated using artificial intelligence, and the sooner it is predicted, the more cerebrovascular disease occurrences can be avoided.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Study Experimental of Design Considerations of Constant Input Current Source

引用

Journal of engineering Science and Technology Review 2022年第1期15卷 68-73页

作者： Puspasari, Fitri Sismanto Ashari, Ahmad Department of Physics Universitas Gadjah Mada Sekip Utara BLS 21 Yogyakarta55281 Indonesia Department of Electrical Engineering and Informatics Universitas Gadjah Mada Yogyakarta55281 Indonesia Department of Computer Science and Electronics Universitas Gadjah Mada Sekip Utara BLS 21 Yogyakarta55281 Indonesia

Resistivity geoelectrical method is a method to determine the value of rock resistivity. The measuring tool for resistivity geoelectrical method is a resistivity meter. One of the important features in designing resistivity meter is providing the constant current for various types of resistive loads. The difference on rocks resistance values is caused by the factor of rock types, temperature, and mineral composition. This research is aimed at designing and analyzing a set of constant current source applied at a circuit of current measurement tool based on LM324. To deal with this problem, five variations of current source were built as well as tested. The research shows the constant current was successfully built with five constant current variations with voltage 30V and tested on the loads of 56 Ohm to 1000 KOhm. The test on integrated constant current circuit measurement, and voltage worked properly on resistance (22 – 46000) Ohm. © 2022. School of Science, IHU. All rights reserved.

关键词： Electric current measurement

来源：评论

学校读者我要写书评

暂无评论

Morphosyntactic Probing of Multilingual BERT Models

arXiv

引用

arXiv 2023年

作者： Acs, Judit Hamerlik, Endre Schwartz, Roy Smith, Noah A. Kornai, Andras Budapest Hungary Budapest University of Technology and Economics Budapest Hungary Department of Applied Informatics Comenius University in Bratislava Faculty of Mathematics Physics and Informatics Bratislava Slovakia Hebrew University of Jerusalem Jerusalem Israel Paul G. Allen School of Computer Science and Engineering University of Washington SeattleWA United States Allen Institute for Artificial Intelligence SeattleWA United States

We introduce an extensive dataset for multilingual probing of morphological information in language models (247 tasks across 42 languages from 10 families), each consisting of a sentence with a target word and a morphological tag as the desired label, derived from the Universal Dependencies treebanks. We find that pre-trained Transformer models (mBERT and XLM-RoBERTa) learn features that attain strong performance across these tasks. We then apply two methods to locate, for each probing task, where the disambiguating information resides in the input. The first is a new perturbation method that "masks" various parts of context;the second is the classical method of Shapley values. The most intriguing finding that emerges is a strong tendency for the preceding context to hold more information relevant to the prediction than the following context. © 2023, CC BY.

关键词： Perturbation techniques

来源：评论

学校读者我要写书评

暂无评论

E-Learning: A Successful Learning Environment

E-Learning: A Successful Learning Environment

引用

computer Science and Emerging Technologies (CSET), International Conference on

作者： Fayiz Momani Amer Ahmad Hatamleh Wan Mohd Amir Fazamin Wan Hamzah Julaily Aida Jusoh Muhammad D. Zakaria Sunanda Das E-business & Commerce Department Faculty of Management and Financial Sciences University of Petra (UOP) Jordan Head of Departments Faculty of Business and Financial Science MIS BI and Data Analysis Irbid National University Irbid Jordan Faculty of Informatics and Computing (UniSZA) Malaysia Faculty of Informatics and Computing UniSZA Malaysia Cyber Security School of Computer Science and Engineering Jain University Bangalore India

Education is developing with the economy. This led to the creation of e-Iearning tools for instructors and students. In e-learning, electronic resources are used to teach. E-learning can be taught in or out of school, but computers and the Internet are essential. That is, e-learning involves delivering education to a big group of people at once or over time. So, overall, in this research is about E-learning that make a successful environment. We chose this term since it is widely used in E-learning research to improve student academic achievement. Due to the difficulties of accessing adequate study materials during the Covid-19 epidemic, distance learning has grown more popular among students. This study proposes a unique learning technique that involves choosing and organizing important learning items using a recommender system. We also studied our method's efficacy. Using a recommended system to help online learning activities was proven to be effective. Software development methodology is a collection of principles and processes used to build software. Systematic software development is the goal. The research approach utilized for this topic is Lean Development Methodology. We also examine the advantages and disadvantages of e-learning. This research also introduces various new strategies and methodologies that may be employed in the future to improve system-learning. It also presents system that we choose and why we choose it. So, it aims to eliminate waste and increase production. Following the principles can help developers avoid nonproductive chores while still producing high-quality results. The ultimate objective is to develop an efficient and error-free system. Finally, E-learning is a social revolution. It's part of a bigger rethinking of how we teach future employees and students. The eLearning platform enabled the learning data visualization dashboard beautifully. User-centered design lets the dashboard adapt to its users' demands.

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：