检索结果-内蒙古大学图书馆

International Conference on Cybernetics and Intelligent System (ICORIS)

作者： Mahmud Isnan Gregorius Natanael Elwirehardja Kuncahyo Setyo Nugroho Imelda Liana Ritonga Roby Pahala Januario Gultom Elida Afni Lubis Bens Pardamean Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta Indonesia Undergraduate of Nursing Program Universitas Imelda Medan Indonesia Medan Indonesia Undergraduate Pharmacy Program Universitas Imelda Medan Indonesia Medan Indonesia Clinical Pathology Unit Rumah Sakit Imelda Pekerja Indonesia Medan Indonesia Computer Science Department BINUS Graduate Program - Master of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798331539603

ISBN: (纸本)9798331539610

The need for an early screening and Computer-Aided Diagnosis (CAD) system based on Artificial Intelligence (AI) for the field of radiology is essential to realize considering the large impact of lung diseases globally. However, developing AI requires a lot of data, which is a challenge because data in the health sector tends to be limited in quantity and has disparities between diagnostic categories (data imbalance). To overcome this issue, DINO ViT models was trained with the publicly available COVID-19 Radiography database in our previous study. The model was subsequently implemented into a web application, with this study focusing on the development of a prototype for the application. The study found this application can be useful for medical personnel and doctors in carrying out early screening and CAD. Meanwhile, it also highlighted the general unfamiliarity with AI applications among medical staff, emphasizing the need for increased education and training on AI's role in healthcare.

关键词： Training Solid modeling Design automation Pulmonary diseases Prototypes Radiology Predictive models Computer aided diagnosis Artificial intelligence Medical diagnostic imaging

来源：评论

学校读者我要写书评

暂无评论

Biased hypothesis formation from projection pursuit

arXiv

引用

arXiv 2022年

作者： Patterson, John Avery, Chris Grear, Tyler Jacobs, Donald J. Department of Bioinformatics and Genomics University of North Carolina Charlotte United States Department of Physics and Optical Science University of North Carolina Charlotte United States Affiliate Faculty of the UNC Charlotte School of Data Science United States

The effect of bias on hypothesis formation is characterized for an automated data-driven projection pursuit neural network to extract and select features for binary classification of data streams. This intelligent exploratory process partitions a complete vector state space into disjoint subspaces to create working hypotheses quantified by similarities and differences observed between two groups of labeled data streams. data streams are typically time sequenced, and may exhibit complex spatio-temporal patterns. For example, given atomic trajectories from molecular dynamics simulation, the machine's task is to quantify dynamical mechanisms that promote function by comparing protein mutants, some known to function while others are nonfunctional. Utilizing synthetic two-dimensional molecules that mimic the dynamics of functional and nonfunctional proteins, biases are identified and controlled in both the machine learning model and selected training data under different contexts. The refinement of a working hypothesis converges to a statistically robust multivariate perception of the data based on a context-dependent perspective. Including diverse perspectives during data exploration enhances interpretability of the multivariate characterization of similarities and differences. © 2022, CC BY-SA.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Fairness-Aware Interpretable Modeling (FAIM) for Trustworthy Machine Learning in Healthcare

arXiv

引用

arXiv 2024年

作者： Liu, Mingxuan Ning, Yilin Ke, Yuhe Shang, Yuqing Chakraborty, Bibhas Ong, Marcus Eng Hock Vaughan, Roger Liu, Nan Centre for Quantitative Medicine Duke-NUS Medical School Singapore Singapore Department of Anaesthesiology and Perioperative Medicine Singapore General Hospital Singapore Programme in Health Services and Systems Research Duke-NUS Medical School Singapore Singapore Department of Statistics and Data Science National University of Singapore Singapore Department of Biostatistics and Bioinformatics Duke University DurhamNC United States Department of Emergency Medicine Singapore General Hospital Singapore Singapore Institute of Data Science National University of Singapore Singapore Singapore

Background The escalating integration of machine learning in high-stakes fields such as healthcare raises substantial concerns about model fairness and interpretability. Conventional bias-mitigation techniques often sacrifice model performance (e.g., accuracy), limiting their impact on aiding decision-making. This study introduces an interpretable bias mitigation framework, Fairness-Aware Interpretable Modeling (FAIM), aiming to balance model performance and fairness. Methods We propose the FAIM framework, which prioritizes fairness in predictive models that are nearly optimal in performance. Initiating with a performance-optimized model yet without explicit fairness considerations (i.e., "fairness-unaware"), we construct a set of nearly-optimal models (i.e., Integral Rashomon Set) to select a fairer alternative (i.e., "fairness-aware"). To identify this fairness-aware model, we introduce a fairness ranking index (FRI) to comprehensively rank model fairness, which encompasses multiple fairness metrics including equalized odds, equal opportunity, and equality of balanced error rate. This index is complemented by an interactive graphical interface that facilitates the incorporation of clinical expertise to guide model selection with contextualized fairness. Moreover, we integrate SHapley Additive exPlanations (SHAP) within FAIM to compare variable contributions between the fairness-unaware and fairness-aware models, thereby uncovering model modifications made to address fairness. We demonstrated the value of FAIM by predicting hospital admission using two real-world databases, MIMIC-IV-ED and SGH-ED, to test its performance in bias mitigation, particularly for sex and race. We gauged model performance via AUC, sensitivity and specificity, and evaluated model fairness via the aforementioned fairness metrics. Results For both datasets, the fairness-aware models yielded by FAIM automatically excluded race and reduced the contribution of sex and other bias-related variabl

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Fine-scale genetic structure of Tujia and central Han Chinese revealing massive genetic admixture under language borrowing

引用

植物分类学报 2021年第1期59卷 1-20页

作者： Guang-Lin He Ying-Xiang Li Meng-Ge Wang Xing Zou Hui-Yuan Yeh Xiao-Min Yang Zheng Wang Ren-Kuan Tang Su-Min Zhu Jian-Xin Guo Ting Luo Jing Zhao Jin Sun Zi-Yang Xia Hao-Liang Fan Rong Hu Lan-Hai Wei Gang Chen Yi-Ping Hou Chuan-Chao Wang Department of Anthropology and Ethnology Institute of AnthropologyNational Institute for Data Science in Health and Medicineand School of Life SciencesXiamen UniversityXiamen 361005FujianChina Institute of Forensic Medicine West China School of Basic Science and Forensic Medicine Sichuan UniversityChengdu 610041China Department of Anthropology and Ethnology Institute of AnthropologyNational Institute for Data Science in Health and Medicineand School of Life SciencesXiamen UniversityXiamen 361005FujianChina Institute of Forensic Medicine West China School of Basic Science and Forensic Medicine Sichuan UniversityChengdu 610041China Institute of Forensic Medicine West China School of Basic Science and Forensic Medicine Sichuan UniversityChengdu 610041China School of Humanities Nanyang Technological UniversityNanyang 639798Singapore Department of Anthropology and Ethnology Institute of AnthropologyNational Institute for Data Science in Health and Medicineand School of Life SciencesXiamen UniversityXiamen 361005FujianChina Institute of Forensic Medicine West China School of Basic Science and Forensic Medicine Sichuan UniversityChengdu 610041China Department of Forensic Medicine College of Basic Medicine Chongqing Medical UniversityChongqing 400000China Department of Forensic Medicine College of Basic Medicine Chongqing Medical UniversityChongqing 400000China Department of Anthropology and Ethnology Institute of AnthropologyNational Institute for Data Science in Health and Medicineand School of Life SciencesXiamen UniversityXiamen 361005FujianChina Department of Anthropology and Ethnology Institute of AnthropologyNational Institute for Data Science in Health and Medicineand School of Life SciencesXiamen UniversityXiamen 361005FujianChina Department of Anthropology and Ethnology Institute of AnthropologyNational Institute for Data Science in Health and Medicineand School of Life SciencesXiamen UniversityXiamen 361005FujianChina Department of Anthropology and Ethnology In

Archaeological,genetic,and linguistic evidence has supported the idea that northern China is the original center of modern Sino-Tibetan-speaking ***,the demographic history of subsequent southward migration and genetic admixture of Han Chinese with surrounding indigenous populations remain uncharacterized,and the language shifts and assimilations accompanied by movement of people,or just an adaptation of cultural ideas among populations in central China is still unclear,especially for Tibeto-Burman-speaking Tujia and central Han Chinese *** resolve this,we genotyped over 60K genome-wide markers in 505 unrelated individuals from 63 indigenous *** results showed both studied Han and Tujia were at the intermediate position in the modern East Asian North-South genetic cline and there was a correlation between the genetic composition and the *** observed the strong genetic assimilation between Tujia people and central Han Chinese,which suggested massive population movements and genetic admixture under language *** and central Han Chinese could be modeled as a two-way admixture deriving primary ancestry from a northern ancestral population closely related to the ancient DevilsCave and present-day Tibetans and a southern ancestral population closely related to the present-day Tai-Kadai and Austronesian-speaking *** ancestral northern population we suspect to be related to the Neolithic millet farming groups in the Yellow River Basin or central *** showed that the newly genotyped populations in Hubei Province had a higher proportion of DevilsCave or modern Tungusic/Mongolic-related northern ancestries,while the Hunan populations harbored a higher proportion of Austronesian/Tai-Kadai-related southern ancestries.

关键词： demographic history gene flow genetic admixture language borrowing Tibeto-Burman Tujia

来源：评论

学校读者我要写书评

暂无评论

Unsupervised and lightly supervised learning in particle physics

arXiv

引用

arXiv 2024年

作者： Bardhan, Jai Mandal, Tanumoy Mitra, Subhadip Neeraj, Cyrin Patra, Monalisa Center for Computational Natural Sciences and Bioinformatics International Institute of Information Technology Hyderabad500 032 India Indian Institute of Science Education and Research Thiruvananthapuram Kerala Vithura695 551 India Center for Quantum Science and Technology International Institute of Information Technology Hyderabad500 032 India iHub-Data International Institute of Information Technology Hyderabad500 032 India

We review the main applications of machine learning models that are not fully supervised in particle physics, i.e., clustering, anomaly detection, detector simulation, and unfolding. Unsupervised methods are ideal for anomaly detection tasks-machine learning models can be trained on background data to identify deviations if we model the background data precisely. The learning can also be partially unsupervised when we can provide some information about the anomalies at the data level. Generative models are useful in speeding up detector simulations-they can mimic the computationally intensive task without large resources. They can also efficiently map detector-level data to parton-level data (i.e., data unfolding). In this review, we focus on interesting ideas and connections and briefly overview the underlying techniques wherever necessary. Copyright © 2024, The Authors. All rights reserved.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

quantms: A cloud-based pipeline for proteomics reanalysis enables the quantification of 17521 proteins in 9,502 human samples

Research Square

引用

Research Square 2023年

作者： Dai, Chengxin Pfeuffer, Julianus Wang, Hong Sachsenberg, Timo Demichev, Vadim Kohlbacher, Oliver Perez-Riverol, Yasset Chongqing Key Laboratory of Big Data for Bio Intelligence Chongqing University of Posts and Telecommunications Chongqing400065 China Algorithmic Bioinformatics Freie Universität Berlin Berlin Germany Department of Computer Science Applied Bioinformatics University of Tübingen Tübingen72076 Germany Charité - Universitätsmedizin Berlin Berlin10117 Germany European Molecular Biology Laboratory European Bioinformatics Institute Wellcome Genome Campus Hinxton CambridgeCB10 1SD United Kingdom

Public proteomics data is rapidly increasing, creating a computational challenge for large-scale reanalysis. Here, we introduce quantms, an open-source cloud-based pipeline for massively parallel proteomics data analysis. We used quantms to reanalyze 56 of the largest datasets, comprising 26801 instrument files from 9502 human samples, to quantify 17521 based on 1.02 million unique peptides. Based on standard file formats improves the reproducibility and deposition of the data to ProteomeXchange. © 2023, CC BY.

关键词： Mass spectrometry

来源：评论

学校读者我要写书评

暂无评论

Identification and evaluation of multi-antigenic epitopes of immunodominant protein from the selected Crimean–Congo hemorrhagic fever virus genome towards the development of diagnostic and vaccine candidates by reverse vaccinology approach

引用

Journal of Proteins and Proteomics 2024年第4期15卷 625-634页

作者： Jeyachandran, Akshay Muthuvel, Reshma Jagannathan, Selvaraj Pillai, Sarika Baburajan Rajendran, Vijayakumar Pachamuthu, Rahul Gandhi Balakrishnan, Ajithkumar Vedagiri, Hemamalini Channappa, Shivanandappa Kukkaler Premkumar, Ananda Arona Sakthivel, Sivakumar Mandal, Abhishek Department of Microbiology Pasteur Institute of India Coonoor The Nilgiris India Centre for Advanced Studies in Crystallography and Biophysics University of Madras Chennai India Computational and Data Sciences Indian Institute of Science Bangalore India The Centre for Nanoscience and Technology Pondicherry University Puducherry India Department of Bioinformatics Bharathiar University Coimbatore India Pasteur Institute of India Coonoor India Department of Biotechnology Pondicherry University Puducherry India

Crimean–Congo hemorrhagic fever (CCHF) is a severe zoonotic disease caused by the CCHF virus (CCHFV) belonging to the family of Bunyaviridae. It spreads through tick bites or direct contact with infected animals, affecting both humans and animals. Despite its high mortality rate, there is currently no vaccine available for the treatment of CCHF. To address this gap, we focused on the antigenic glycoprotein precursor (GPC) of CCHFV. This present study aims to identify potential T-cell epitopes that could serve as vaccine candidates. Using bioinformatics tools, epitopes were predicted for both major histocompatibility complex (MHC) classes relevant to the Indian population. These predicted epitopes were then evaluated for antigenicity, immunogenicity, toxicity, and conservancy. Ten epitopes with the highest binding scores were selected and modeled. These epitopes were subsequently docked with their respective HLA allele proteins. Notably, two epitopes—KLPPEIITL (MHC-1) and CKVIFYLLIIFGTLG (MHC-2)—demonstrated significant binding affinity with HLA alleles HLA-A*24:02 and HLA-DRB1*15:01, respectively. Furthermore, molecular dynamics simulations revealed that both epitopes exhibited higher stability when interacting with their respective HLA allele proteins. These findings suggest that these epitopes are promising candidates for developing an epitope-based vaccine against Crimean–Congo hemorrhagic fever (CCHF).

关键词：

来源：评论

学校读者我要写书评

暂无评论

ζ-QVAE: A Quantum Variational Autoencoder utilizing Regularized Mixed-state Latent Representations

arXiv

引用

arXiv 2024年

作者： Wang, Gaoyuan Warrell, Jonathan Emani, Prashant S. Gerstein, Mark Program in Computational Biology and Bioinformatics Yale University New HavenCT06520 United States Department of Molecular Biophysics and Biochemistry Yale University New HavenCT06520 United States Department of Computer Science Yale University New HavenCT06520 United States Department of Statistics & Data Science Yale University New HavenCT06520 United States Department of Biomedical Informatics & Data Science Yale University New HavenCT06520 United States NEC Laboratories America PrincetonNJ08540 United States

A major challenge in near-term quantum computing is its application to large real-world datasets due to scarce quantum hardware resources. One approach to enabling tractable quantum models for such datasets involves compressing the original data to manageable dimensions while still representing essential information for downstream analysis. In classical machine learning, variational autoencoders (VAEs) facilitate efficient data compression, representation learning for subsequent tasks, and novel data generation. However, no model has been proposed that exactly captures all of these features for direct application to quantum data on quantum computers. Some existing quantum models for data compression lack regularization of latent representations, thus preventing direct use for generation and control of generalization. Others are hybrid models with only some internal quantum components, impeding direct training on quantum data. To bridge this gap, we present a fully quantum framework, ζ-QVAE, which encompasses all the capabilities of classical VAEs and can be directly applied for both classical and quantum data compression. Our model utilizes regularized mixed states to attain optimal latent representations. It accommodates various divergences for reconstruction and regularization. Furthermore, by accommodating mixed states at every stage, it can utilize the full-data density matrix and allow for a"global"training objective. Doing so, in turn, makes efficient optimization possible and has potential implications for private and federated learning. In addition to exploring the theoretical properties of ζ-QVAE, we demonstrate its performance on representative genomics and synthetic data. Our results consistently indicate that ζ-QVAE exhibits similar or better performance compared to matched classical models. © 2024, CC BY-NC-ND.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

A methodology for hierarchical classification of semantic answer types of questions

A methodology for hierarchical classification of semantic an...

引用

2020 SeMantic AnsweR Type Prediction Task, SMART 2020 at 2020 Semantic Web Challenge, ISWC 2020

作者： Ammar, Ammar Mehryar, Shervin Celebi, Remzi Department of Bioinformatics BiGCaT NUTRIM Maastricht University Netherlands University of Toronto Canada Institute of Data Science Maastricht University Maastricht Netherlands

Question answering systems have recently been integrated with many smart devices and search engines. Answer type prediction plays an important role in question answering systems as it can help filter irrelevant results and improve overall search and retrieval performance. Here, we present our approach for answer type prediction using the datasets provided for the International Semantic Web Conference (ISWC 2020) SMART Task Challenge. Predicting granular answer types for a question from a big knowledge graph is a greater challenge due to the large number of possible classes. Thus, we propose a 3-step approach to tackle the challenge task. We start with building a classifier that predicts the category of the types and build another classifier just for resource types. The latter model will predict the most general (frequent) type for each question, ignoring type hierarchy. We use a multi-class text classification algorithm built-in fastai library for these two models. The models’ accuracies are 0.95 and 0.73 for category and generic type classification respectively in the validation set (20% randomly chosen samples) of the DBPedia dataset. Next, we train a third classifier to find more specific types (sub-classes) for each question based on the previous general predicted types. We achieve 0.62 and 0.61 using NDCG@5 and NDCG@10 metrics respectively for the test set. Copyright © 2020 for this paper by its authors.

关键词： Search engines

来源：评论

学校读者我要写书评

暂无评论

Nonconvex Factorization and Manifold Formulations are Almost Equivalent in Low-rank Matrix Optimization

arXiv

引用

arXiv 2021年

作者： Luo, Yuetian Li, Xudong Zhang, Anru R. Data Science Institute University of Chicago United States School of Data Science Fudan University China Departments of Biostatistics & Bioinformatics Computer Science Electrical & Computer Engineering Mathematics and Statistical Science Duke University United States

In this paper, we consider the geometric landscape connection of the widely studied manifold and factorization formulations in low-rank positive semidefinite (PSD) and general matrix optimization. We establish a sandwich relation on the spectrum of Riemannian and Euclidean Hessians at first-order stationary points (FOSPs). As a result of that, we obtain an equivalence on the set of FOSPs, second-order stationary points (SOSPs), and strict saddles between the manifold and the factorization formulations. In addition, we show the sandwich relation can be used to transfer more quantitative geometric properties from one formulation to another. Similarities and differences in the landscape connection under the PSD case and the general case are discussed. To the best of our knowledge, this is the first geometric landscape connection between the manifold and the factorization formulations for handling rank constraints, and it provides a geometric explanation for the similar empirical performance of factorization and manifold approaches in low-rank matrix optimization observed in the literature. In the general low-rank matrix optimization, the landscape connection of two factorization formulations (unregularized and regularized ones) is also provided. By applying these geometric landscape connections, in particular, the sandwich relation, we are able to solve unanswered questions in literature and establish stronger results in the applications on geometric analysis of phase retrieval, well-conditioned low-rank matrix optimization, and the role of regularization in factorization arising from machine learning and signal processing. © 2021, CC BY-SA.

关键词： Matrix factorization

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：