检索结果-内蒙古大学图书馆

WebQuorumChain: A web framework for quorum-based health care model learning

informatics in Medicine Unlocked 2024年 50卷 101590页

作者： Shao, Xiyan Pham, Anh Kuo, Tsung-Ting UCSD Department of Computer Science and Engineering University of California San Diego La Jolla CA United States UCSD Health Department of Biomedical Informatics University of California San Diego La Jolla CA United States Department of Biomedical Informatics and Data Science School of Medicine Yale University New Haven CT United States Department of Surgery School of Medicine Yale University New Haven CT United States

Background: Institutions interested in collaborative machine learning to enhance healthcare may be deterred by privacy concerns. Decentralized federated learning is a privacy-preserving and security-robust tool to promote cross-institutional learning, however, such frameworks require complex setups and advanced technical expertise. Here, we aim to improve their utilization by offering an intuitive, user-friendly, and secure system that integrates both front-end and back-end functionalities. Method: We develop WebQuorumChain, an integrated system built upon the QuorumChain schema. We test the system on a 2-site network using two publicly available health datasets and measure the average vertical and horizontal-ensemble AUCs per dataset across 30 trials, as well as the average execution time of the system. Results: Our system achieved consistently high AUCs for each dataset (0.94–0.96), with reasonable total execution times ranging from 5 to 20 min, inclusive of modeling and all other system overheads. The front-end displays event logs generated from back-end layers in real time, in sync with the progress of the underlying algorithm. Conclusions: We develop a web-based system that supplies users with visual tools to configure the federated learning network, manage training sessions, and inspect the learning process. WebQuorumChain helps schedule and monitor low-level processes without violating the fundamental security promises of cross-institutional decentralized machine learning. The system also maintains predictive accuracy and runtime efficiency in the presence of additional layers. WebQuorumChain will help promote meaningful collaboration among healthcare institutions, who can retain full control of their data privacy while contributing to data-driven discoveries. © 2024 The Authors

关键词： Clinical research information systems Health information infrastructure Machine learning Predictive modeling Privacy and security

来源：评论

学校读者我要写书评

暂无评论

Echo-E3Net: Efficient Endo-Epi Spatio-Temporal Network for Ejection Fraction Estimation

arXiv

引用

arXiv 2025年

作者： Heidari, Moein Bozorgpour, Afshin Zarif-Fakharnia, AmirHossein Merhof, Dorit Hacihaliloglu, Ilker School of Biomedical Engineering University of British Columbia VancouverBC Canada Faculty of Informatics and Data Science University of Regensburg Regensburg Germany School of Electrical Engineering Iran University of Science and Technology Tehran Iran Department of Radiology University of British Columbia VancouverBC Canada Department of Medicine University of British Columbia VancouverBC Canada

Left ventricular ejection fraction (LVEF) is a critical metric for assessing cardiac function, widely used in diagnosing heart failure and guiding clinical decisions. Despite its importance, conventional LVEF estimation remains time-consuming and operator-dependent. Recent deep learning advancements have enhanced automation, yet many existing models are computationally demanding, hindering their feasibility for real-time clinical applications. Additionally, the interplay between spatial and temporal features is crucial for accurate estimation but is often overlooked. In this work, we propose Echo-E3Net, an efficient Endo-Epi spatio-temporal network tailored for LVEF estimation. Our method introduces the Endo-Epi Cardial Border Detector (E2CBD) module, which enhances feature extraction by leveraging spatial and temporal landmark cues. Complementing this, the Endo-Epi Feature Aggregator (E2FA) distills statistical descriptors from backbone feature maps, refining the final EF prediction. These modules, along with a multi-component loss function tailored to align with the clinical definition of EF, collectively enhance spatial-temporal representation learning, ensuring robust and efficient EF estimation. We evaluate Echo-E3Net on the EchoNet-Dynamic dataset, achieving a RMSE of 5.15 and an R2 score of 0.82, setting a new benchmark in efficiency with 6.8 million parameters and only 8.49G Flops. Our model operates without pre-training, data augmentation, or ensemble methods, making it well-suited for real-time point-of-care ultrasound (PoCUS) applications. Our Code is publicly available on GitHub. © 2025, CC BY-NC-SA.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Efficient and Effective Adaptation of Large Language Models for Sequential Recommendation

arXiv

引用

arXiv 2023年

作者： Peng, Bo Burns, Ben Chen, Ziqi Parthasarathy, Srinivasan Ning, Xia Department of Computer Science and Engineering The Ohio State University Columbus United States Department of Biomedical Informatics The Ohio State University Columbus United States Translational Data Analytics Institute The Ohio State University Columbus United States

In recent years, with large language models (LLMs) achieving state-of-the-art performance in context understanding, increasing efforts have been dedicated to developing LLM-enhanced sequential recommendation (SR) methods. Considering that most existing LLMs are not specifically optimized for recommendation tasks, adapting them for SR becomes a critical step in LLM-enhanced SR methods. Though numerous adaptation methods have been developed, it still remains a significant challenge to adapt LLMs for SR both efficiently and effectively. To address this challenge, in this paper, we introduce a novel side sequential network adaptation method, denoted as SSNA, for LLM-enhanced SR. SSNA features three key designs to allow both efficient and effective LLM adaptation. First, SSNA learns adapters separate from LLMs, while fixing all the pre-trained parameters within LLMs to allow efficient adaptation. In addition, SSNA adapts the top-a layers of LLMs jointly, and integrates adapters sequentially for enhanced effectiveness (i.e., recommendation performance). We compare SSNA against five state-of-the-art baseline methods on five benchmark datasets using three LLMs. The experimental results demonstrate that SSNA significantly outperforms all the baseline methods in terms of recommendation performance, and achieves substantial improvement over the best-performing baseline methods at both run-time and memory efficiency during training. Our analysis shows the effectiveness of integrating adapters in a sequential manner. Our parameter study demonstrates the effectiveness of jointly adapting the top-a layers of LLMs. © 2023, CC BY.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Multi-modality Meets Re-learning: Mitigating Negative Transfer in Sequential Recommendation

arXiv

引用

arXiv 2023年

作者： Peng, Bo Parthasarathy, Srinivasan Ning, Xia The Department of Computer Science and Engineering The Ohio State University ColumbusOH43210 United States The Department of Computer Science and Engineering The Translational Data Analytics Institute The Ohio State University ColumbusOH43210 United States The Department of Biomedical Informatics The Department of Computer Science and Engineering The Translational Data Analytics Institute The Ohio State University ColumbusOH43210 United States

Learning effective recommendation models from sparse user interactions represents a fundamental challenge in developing sequential recommendation methods. Recently, pre-training-based methods have been developed to tackle this challenge. Though promising, in this paper, we show that existing methods suffer from the notorious negative transfer issue, where the model adapted from the pre-trained model results in worse performance compared to the model learned from scratch in the task of interest (i.e., target task). To address this issue, we develop a method, denoted as ANT, for transferable sequential recommendation. ANT mitigates negative transfer by 1) incorporating multi-modality item information, including item texts, images and prices, to effectively learn more transferable knowledge from related tasks (i.e., auxiliary tasks);and 2) better capturing task-specific knowledge in the target task using a re-learning-based adaptation strategy. We evaluate ANT against eight state-of-the-art baseline methods on five target tasks. Our experimental results demonstrate that ANT does not suffer from the negative transfer issue on any of the target tasks. The results also demonstrate that ANT substantially outperforms baseline methods in the target tasks with an improvement of as much as 15.2%. Our analysis highlights the superior effectiveness of our re-learning-based strategy compared to fine-tuning on the target tasks. © 2023, CC BY.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Session Introduction: TOWARDS ETHICAL biomedical informatics: LEARNING FROM OLELO NOEAU, HAWAIIAN PROVERBS 28

Session Introduction: TOWARDS ETHICAL BIOMEDICAL INFORMATICS...

引用

28th Pacific Symposium on Biocomputing, PSB 2023

作者： Washington, Peter Y. Puniwai, Noelani Kamaka, Martina Gürsoy, Gamze Tatonetti, Nicholas Brenner, Steven E. Wall, Dennis P. Department Of Information Computer Sciences University Of Hawaii At Manoa HonoluluHI96822 United States Kamakakuokalani University Of Hawaii At Manoa HonoluluHI96822 United States Department Of Native Hawaiian Health University Of Hawaii At Manoa HonoluluHI96822 United States Department Of Biomedical Informatics Columbia University New YorkNY10032 United States Department Of Plant Microbial Biology University Of California Berkeley BerkeleyCA94720 United States Biomedical Data Science And Psychiatry Behavioral Sciences Stanford University StanfordCA94305 United States

ISBN: (纸本)9789811270604

Innovations in human-centered biomedical informatics are often developed with the eventual goal of real-world translation. While biomedical research questions are usually answered in terms of how a method performs in a particular context, we argue that it is equally important to consider and formally evaluate the ethical implications of informatics solutions. Several new research paradigms have arisen as a result of the consideration of ethical issues, including but not limited for privacy-preserving computation and fair machine learning. In the spirit of the Pacific Symposium on Biocomputing, we discuss broad and fundamental principles of ethical biomedical informatics in terms of Olelo Noeau, or Hawaiian proverbs and poetical sayings that capture Hawaiian values. While we emphasize issues related to privacy and fairness in particular, there are a multitude of facets to ethical biomedical informatics that can benefit from a critical analysis grounded in ethics. © 2022 The Authors.

关键词： Ethical technology

来源：评论

学校读者我要写书评

暂无评论

CAUSAL RULE ENSEMBLE METHOD FOR ESTIMATING HETEROGENEOUS TREATMENT EFFECT WITH CONSIDERATION OF MAIN EFFECTS

arXiv

引用

arXiv 2023年

作者： Hiraishi, Mayu Wan, Ke Tanioka, Kensuke Yadohisa, Hiroshi Shimokawa, Toshio Clinical Study Support Center Wakayama Medical University Hospital Graduate School of Culture and Information Science Doshisha University Japan Department of Medical Data Science Wakayama Medical University Japan Department of Biomedical Sciences and Informatics Doshisha University Japan Department of Culture and Information Science Doshisha University Japan

This study proposes a novel framework based on the RuleFit method to estimate Heterogeneous Treatment Effect (HTE) in a randomized clinical trial. To achieve this, we adopted S-learner of the metaalgorithm for our proposed framework. The proposed method incorporates a rule term for the main effect and treatment effect, which allows HTE to be interpretable form of rule. By including a main effect term in the proposed model, the selected rule is represented as an HTE that excludes other effects. We confirmed a performance equivalent to that of another ensemble learning methods through numerical simulation and demonstrated the interpretation of the proposed method from a real data application. Copyright © 2023, The Authors. All rights reserved.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

OPTIC: Optimizing Patient-Provider Triaging & Improving Communications in Clinical Operations using GPT-4 data Labeling and Model Distillation.

arXiv

引用

arXiv 2025年

作者： Santamaria-Pang, Alberto Tuan, Frank Campbell, Ross Zhang, Cindy Jindal, Ankush Surapur, Roopa Holloman, Brad Hanisch, Deanna Buckley, Rae Cooney, Carisa Tarapov, Ivan Peairs, Kimberly S. Hasselfeld, Brian Greene, Peter Microsoft Health AI RedmondWA United States Adjunct Faculty Department of Biomedical Informatics & Data Science Johns Hopkins University BaltimoreMD United States Johns Hopkins Medicine BaltimoreMD United States University of Washington SeattleWA United States Department of Biomedical Informatics & Data Science Johns Hopkins University BaltimoreMD United States Microsoft BaltimoreMD United States Department of Plastic and Reconstructive Surgery Johns Hopkins Medicine BaltimoreMD United States Department of Plastic and Reconstructive Surgery Johns Hopkins University BaltimoreMD United States Department of Medicine Johns Hopkins University BaltimoreMD United States

Background: The COVID-19 pandemic has led to significant increases in the use of telemedicine as well as patient messaging via electronic medical portals (patient medical advice requests, or PMARs). While these systems offer benefits such as increased access to healthcare for patients, they have also created a substantial burden for healthcare providers due to surges in PMARs. This study addresses the need for an efficient tool for message triaging to reduce physician burden and improve patient-provider communication. Methods: We developed OPTIC (Optimizing Patient-Provider Triaging & Improving Communications in Clinical Operations), a robust tool for message triaging using GPT-4 for data labeling and BERT for model distillation. The study utilized a dataset of 405,487 patient messaging encounters at Johns Hopkins Medicine between January to June 2020. We employed prompt engineering strategies with GPT-4 to create a high-quality labelled dataset, which was then used to train a BERT model for classifying messages as "Admin" or "Clinical". Results: The BERT model achieved an accuracy of 88.85% on the test set derived from validated GPT-4 labeling, with a 88.29% sensitivity, 89.38% specificity, and an F1 score of 0.8842. BERTopic analysis identified 81 distinct topics within the test data, with the model achieving over 80% accuracy in classification for 58 topics. The system was successfully deployed through Epic's Nebula Cloud Platform, demonstrating its applicability in healthcare settings. Conclusion: OPTIC offers a scalable and efficient tool for triaging PMARs in health systems. Leveraging advanced natural language processing techniques can reduce the administrative burden on healthcare providers and streamline patient care coordination. © 2025, CC BY.

关键词： Labeled data

来源：评论

学校读者我要写书评

暂无评论

BiomedRAG: A Retrieval augmented Large Language Model for Biomedicine

arXiv

引用

arXiv 2024年

作者： Li, Mingchen Kilicoglu, Halil Xu, Hua Zhang, Rui Division of Computational Health Sciences Department of Surgery University of Minnesota MinneapolisMN United States School of Information Sciences University of Illinois at Urbana-Champaign Champaign United States Section of Biomedical Informatics and Data Science School of Medicine Yale University New HavenCT United States

Large Language Models (LLMs) have swiftly emerged as vital resources for different applications in the biomedical and healthcare domains;however, these models encounter issues such as generating inaccurate information or hallucinations. Retrieval-augmented generation provided a solution for these models to update knowledge and enhance their performance. In contrast to previous retrieval-augmented LMs, which utilize specialized cross-attention mechanisms to help LLM encode retrieved text, BiomedRAG adopts a simpler approach by directly inputting the retrieved chunk-based documents into the LLM. This straightforward design is easily applicable to existing retrieval and language models, effectively bypassing noise information in retrieved documents, particularly in noise-intensive tasks. Moreover, we demonstrate the potential for utilizing the LLM to supervise the retrieval model in the biomedical domain, enabling it to retrieve the document that assists the LM in improving its predictions. Specifically, BIOMEDRAG retrieves relevant documents from a specifically curated, diverse chunk database through a unique, purpose-built chunk scoring mechanism by a tunable scorer. This retrieval process then integrates the selected information directly into the LLM's input, facilitating the generation of expected output, such as structured knowledge. Our experiments reveal that with the tuned scorer, BIOMEDRAG attains superior performance across 4 biomedical NLP tasks, encompassing information extraction (triple extraction, relation extraction), text classification, and link prediction leveraging over 8 datasets. For instance, in the triple extraction task, BIOMEDRAG outperforms other triple extraction systems with micro-F1 scores of 81.42 and 88.83 on GIT and ChemProt corpora, respectively. The outstanding performance underscores the potential of BIOMEDRAG in constructing effective biomedical intervention systems for various tasks. © 2024, CC BY.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Grounding Realizable Entities

arXiv

引用

arXiv 2024年

作者： Rabenberg, Michael Benson, Carter Donato, Federico He, Yongqun Huffman, Anthony Babcock, Shane Beverley, John National Center for Ontological Research United States University at Buffalo Department of Biomedical informatics United States University at Buffalo Department of Philosophy United States University of Michigan Medical School United States KadSci LLC United States Institute for Artificial Intelligence and Data Science United States

Ontological representations of qualities, dispositions, and roles have been refined over the past decade, clarifying subtle distinctions in life science research. After articulating a widely-used characterization of these entities within the context of Basic Formal Ontology (BFO), we identify gaps in this treatment and motivate the need for supplementing the BFO characterization. By way of supplement, we propose definitions for grounding relations holding between qualities and dispositions, and dispositions and roles, illustrating our proposal by representing subtle aspects of host-pathogen interactions. © 2024, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Unified Representation of Genomic and biomedical Concepts through Multi-Task, Multi-Source Contrastive Learning

arXiv

引用

arXiv 2024年

作者： Yuan, Hongyi Liu, Suqi Cho, Kelly Liao, Katherine Pereira, Alexandre Cai, Tianxi Department of Biomedical Informatics Harvard Medical School United States Department of Statistics and Data Science Tsinghua University China VA Boston Healthcare System United States Department of Medicine Brigham and Women’s Hospital United States Department of Medicine Harvard Medical School United States

biomedical concepts, along with genomic features such as single-nucleotide polymorphisms (SNPs) and gene expression, are crucial for understanding the genetic and biomedical relationships in modern medicine. Genome-wide association studies (GWAS), expression quantitative trait loci (eQTL), and phenome-wide association studies (PheWAS) derived from biobank data provide valuable insights into the interplay between genomic and clinical features. However, harmonizing existing biological and clinical knowledge from different sources to enable a better understanding of diseases and treatments remains challenging. Variations in biomedical concept coding systems and differences in SNP selections across data sources create barriers to effective integration. To address this challenge, we introduce GENomic Encoding REpresentation with Language Model (GENEREL), a framework designed to bridge genetic and biomedical knowledge bases. What sets GENEREL apart is its ability to fine-tune language models to infuse biological knowledge behind clinical concepts such as diseases and medications. This fine-tuning enables the model to capture complex biomedical relationships more effectively, enriching the understanding of how genomic data connects to clinical outcomes. By constructing a unified embedding space for biomedical concepts and a wide range of common SNPs from sources such as patient-level data, biomedical knowledge graphs, and GWAS summaries, GENEREL aligns the embeddings of SNPs and clinical concepts through multi-task contrastive learning. This allows the model to adapt to diverse natural language representations of biomedical concepts while bypassing the limitations of traditional code mapping systems across different data sources. Our experiments demonstrate GENEREL’s ability to effectively capture the nuanced relationships between SNPs and clinical concepts. GENEREL also emerges to discern the degree of relatedness, potentially allowing for a more refined identification of

关键词： Gene expression

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：