IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics最新文献

{"title":"Identifying Symptoms of Delirium from Clinical Narratives Using Natural Language Processing.","authors":"Aokun Chen, Daniel Paredes, Zehao Yu, Xiwei Lou, Roberta Brunson, Jamie N Thomas, Kimberly A Martinez, Robert J Lucero, Tanja Magoc, Laurence M Solberg, Urszula A Snigurska, Sarah E Ser, Mattia Prosperi, Jiang Bian, Ragnhildur I Bjarnadottir, Yonghui Wu","doi":"10.1109/ichi61247.2024.00046","DOIUrl":"10.1109/ichi61247.2024.00046","url":null,"abstract":"<p><p>Delirium is an acute decline or fluctuation in attention, awareness, or other cognitive function that can lead to serious adverse outcomes. Despite the severe outcomes, delirium is frequently unrecognized and uncoded in patients' electronic health records (EHRs) due to its transient and diverse nature. Natural language processing (NLP), a key technology that extracts medical concepts from clinical narratives, has shown great potential in studies of delirium outcomes and symptoms. To assist in the diagnosis and phenotyping of delirium, we formed an expert panel to categorize diverse delirium symptoms, composed annotation guidelines, created a delirium corpus with diverse delirium symptoms, and developed NLP methods to extract delirium symptoms from clinical notes. We compared 5 state-of-the-art transformer models including 2 models (BERT and RoBERTa) from the general domain and 3 models (BERT_MIMIC, RoBERTa_MIMIC, and GatorTron) from the clinical domain. GatorTron achieved the best strict and lenient F1 scores of 0.8055 and 0.8759, respectively. We conducted an error analysis to identify challenges in annotating delirium symptoms and developing NLP systems. To the best of our knowledge, this is the first large language model-based delirium symptom extraction system. Our study lays the foundation for the future development of computable phenotypes and diagnosis methods for delirium.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"305-311"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11670120/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142900616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learning to Rank Complex Biomedical Hypotheses for Accelerating Scientific Discovery.","authors":"Juncheng Ding, Shailesh Dahal, Bijaya Adhikari, Kishlay Jha","doi":"10.1109/ichi61247.2024.00044","DOIUrl":"10.1109/ichi61247.2024.00044","url":null,"abstract":"<p><p>Hypothesis generation (HG) is a fundamental problem in biomedical text mining that uncovers plausible implicit links ( <math><mi>B</mi></math> terms) between two disjoint concepts of interest ( <math><mi>A</mi></math> and <math><mi>C</mi></math> terms). Over the past decade, many HG approaches based on distributional statistics, graph-theoretic measures, and supervised machine learning methods have been proposed. Despite significant advances made, the existing approaches have two major limitations. First, they mainly focus on enumerating hypotheses and often neglect to rank them in a semantically meaningful way. This leads to wasted time and resources as researchers may focus on hypotheses that are ultimately not supported by experimental evidence. Second, the existing approaches are designed to rank hypotheses with only one intermediate or evidence term (referred as simple hypotheses), and thus are unable to handle hypotheses with multiple intermediate terms (referred as complex hypotheses). This is limiting because recent research has shown that the complex hypotheses could be of greater practical value than simple ones, especially in the early stages of scientific discovery. To address these issues, we propose a new HG ranking approach that leverages upon the expressive power of Graph Neural Networks (GNN) coupled with a domain-knowledge guided Noise-Contrastive Estimation (NCE) strategy to effectively rank both simple and complex biomedical hypotheses. Specifically, the message passing capabilities of GNN allows our approach to capture the rich interactions between biomedical entities and succinctly handle the complex hypotheses with variable intermediate terms. Moreover, the proposed domain knowledge-guided NCE strategy enables the ranking of complex hypotheses based on their coherence with the established biomedical knowledge. Extensive experiment results on five recognized biomedical datasets show that the proposed approach consistently outperforms the existing baselines and prioritizes hypotheses worthy of potential clinical trials.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"285-293"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11920884/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143665495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"<ArticleTitle xmlns:ns0=\"http://www.w3.org/1998/Math/MathML\">An average-case efficient two-stage algorithm for enumerating all longest common substrings of minimum length <ns0:math><ns0:mi>k</ns0:mi></ns0:math> between genome pairs.","authors":"Mattia Prosperi, Simone Marini, Christina Boucher","doi":"10.1109/ichi61247.2024.00020","DOIUrl":"10.1109/ichi61247.2024.00020","url":null,"abstract":"<p><p>A problem extension of the longest common substring (LCS) between two texts is the enumeration of all LCSs given a minimum length <math><mi>k</mi></math> (ALCS- <math><mi>k</mi></math> ), along with their positions in each text. In bioinformatics, an efficient solution to the ALCS- <math><mi>k</mi></math> for very long texts -genomes or metagenomes- can provide useful insights to discover genetic signatures responsible for biological mechanisms. The ALCS- <math><mi>k</mi></math> problem has two additional requirements compared to the LCS problem: one is the minimum length <math><mi>k</mi></math> , and the other is that all common strings longer than <math><mi>k</mi></math> must be reported. We present an efficient, two-stage ALCS- <math><mi>k</mi></math> algorithm exploiting the spectrum of text substrings of length <math><mi>k</mi></math> ( <math><mi>k</mi></math> -mers). Our approach yields a worst-case time complexity loglinear in the number of <math><mi>k</mi></math> -mers for the first stage, and an average-case loglinear in the number of common <math><mi>k</mi></math> -mers for the second stage (several orders of magnitudes smaller than the total <math><mi>k</mi></math> -mer spectrum). The space complexity is linear in the first phase (disk-based), and on average linear in the second phase (disk- and memory-based). Tests performed on genomes for different organisms (including viruses, bacteria and animal chromosomes) show that run times are consistent with our theoretical estimates; further, comparisons with MUMmer4 show an asymptotic advantage with divergent genomes.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"93-102"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11412151/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142302596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assertion Detection in Clinical Natural Language Processing using Large Language Models.","authors":"Yuelyu Ji, Zeshui Yu, Yanshan Wang","doi":"10.1109/ichi61247.2024.00039","DOIUrl":"10.1109/ichi61247.2024.00039","url":null,"abstract":"<p><p>In this study, we aim to address the task of assertion detection when extracting medical concepts from clinical notes, a key process in clinical natural language processing (NLP). Assertion detection in clinical NLP usually involves identifying assertion types for medical concepts in the clinical text, namely certainty (whether the medical concept is positive, negated, possible, or hypothetical), temporality (whether the medical concept is for present or the past history), and experiencer (whether the medical concept is described for the patient or a family member). These assertion types are essential for healthcare professionals to quickly and clearly understand the context of medical conditions from unstructured clinical texts, directly influencing the quality and outcomes of patient care. Although widely used, traditional methods, particularly rule-based NLP systems and machine learning or deep learning models, demand intensive manual efforts to create patterns and tend to overlook less common assertion types, leading to an incomplete understanding of the context. To address this challenge, our research introduces a novel methodology that utilizes Large Language Models (LLMs) pre-trained on a vast array of medical data for assertion detection. We enhanced the current method with advanced reasoning techniques, including Tree of Thought (ToT), Chain of Thought (CoT), and Self-Consistency (SC), and refine it further with Low-Rank Adaptation (LoRA) fine-tuning. We first evaluated the model on the i2b2 2010 assertion dataset. Our method achieved a micro-averaged F-1 of 0.89, with 0.11 improvements over the previous works. To further assess the generalizability of our approach, we extended our evaluation to a local dataset that focused on sleep concept extraction. Our approach achieved an F-1 of 0.74, which is 0.31 higher than the previous method. The results show that using LLMs is a viable option for assertion detection in clinical NLP and can potentially integrate with other LLM-based concept extraction models for clinical NLP tasks.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"242-247"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11908446/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143652454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Developing a computational representation of human physical activity and exercise using open ontology-based approach: a Tai Chi use case.","authors":"Eloisa Nguyen, Rebecca Z Lin, Yang Gong, Cui Tao, Muhammad Tuan Amith","doi":"10.1109/ichi61247.2024.00012","DOIUrl":"10.1109/ichi61247.2024.00012","url":null,"abstract":"<p><p>Many studies have examined the impact of exercise and other physical activities in influencing the health outcomes of individuals. These physical activities entail an intricate sequence and series of physical anatomy, physiological movement, movement of the anatomy, etc. To better understand how these components interact with one another and their downstream impact on health outcomes, there needs to be an information model that conceptualizes all entities involved. In this study, we introduced our early development of an ontology model to computationally describe human physical activities and the various entities that compose each activity. We developed an open-sourced biomedical ontology called the Kinetic Human Movement Ontology that reused OBO Foundry terminologies and encoded in OWL2. We applied this ontology in modeling and linking a specific Tai Chi movement. The contribution of this work could enable modeling of information relating to human physical activity, like exercise, and lead towards information standardization of human movement for analysis. Future work will include expanding our ontology to include more expressive information and completely modeling entire sets of movement from human physical activity.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"31-39"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11503552/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142514161","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analyzing Social Factors to Enhance Suicide Prevention Across Population Groups.","authors":"Richard Li Xu, Song Wang, Zewei Wang, Yuhan Zhang, Yunyu Xiao, Jyotishman Pathak, David Hodge, Yan Leng, S Craig Watkins, Ying Ding, Yifan Peng","doi":"10.1109/ichi61247.2024.00032","DOIUrl":"10.1109/ichi61247.2024.00032","url":null,"abstract":"<p><p>Social factors like family background, education level, financial status, and stress can impact public health outcomes, such as suicidal ideation. However, the analysis of social factors for suicide prevention has been limited by the lack of up-to-date suicide reporting data, variations in reporting practices, and small sample sizes. In this study, we analyzed 172,629 suicide incidents from 2014 to 2020 utilizing the National Violent Death Reporting System Restricted Access Database (NVDRS-RAD). Logistic regression models were developed to examine the relationships between demographics and suicide-related circumstances. Trends over time were assessed, and Latent Dirichlet Allocation (LDA) was used to identify common suicide-related social factors. Mental health, interpersonal relationships, mental health treatment and disclosure, and school/work-related stressors were identified as the main themes of suicide-related social factors. This study also identified systemic disparities across various population groups, particularly concerning Black individuals, young people aged under 24, healthcare practitioners, and those with limited education backgrounds, which shed light on potential directions for demographic-specific suicidal interventions.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"189-199"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11450796/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142382637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing Large Language Models for Clinical Decision Support by Incorporating Clinical Practice Guidelines.","authors":"David Oniani, Xizhi Wu, Shyam Visweswaran, Sumit Kapoor, Shravan Kooragayalu, Katelyn Polanska, Yanshan Wang","doi":"10.1109/ichi61247.2024.00111","DOIUrl":"https://doi.org/10.1109/ichi61247.2024.00111","url":null,"abstract":"<p><p>Large Language Models (LLMs), enhanced with Clinical Practice Guidelines (CPGs), can significantly improve Clinical Decision Support (CDS). However, approaches for incorporating CPGs into LLMs are not well studied. In this study, we develop three distinct methods for incorporating CPGs into LLMs: Binary Decision Tree (BDT), Program-Aided Graph Construction (PAGC), and Chain-of-Thought-Few-Shot Prompting (CoT-FSP), and focus on CDS for COVID-19 outpatient treatment as the case study. Zero-Shot Prompting (ZSP) is our baseline method. To evaluate the effectiveness of the proposed methods, we create a set of synthetic patient descriptions and conduct both automatic and human evaluation of the responses generated by four LLMs: GPT-4, GPT-3.5 Turbo, LLaMA, and PaLM 2. All four LLMs exhibit improved performance when enhanced with CPGs compared to the baseline ZSP. BDT outperformed both CoT-FSP and PAGC in automatic evaluation. All of the proposed methods demonstrate high performance in human evaluation. LLMs enhanced with CPGs outperform plain LLMs with ZSP in providing accurate recommendations for COVID-19 outpatient treatment, highlighting the potential for broader applications beyond the case study.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"694-702"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11909794/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143652456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-Task Deep Neural Networks for Irregularly Sampled Multivariate Clinical Time Series.","authors":"Yuxi Liu, Zhenhao Zhang, Shaowen Qin, Jiang Bian","doi":"10.1109/ichi61247.2024.00025","DOIUrl":"10.1109/ichi61247.2024.00025","url":null,"abstract":"<p><p>Multivariate clinical time series data, such as those contained in Electronic Health Records (EHR), often exhibit high levels of irregularity, notably, many missing values and varying time intervals. Existing methods usually construct deep neural network architectures that combine recurrent neural networks and time decay mechanisms to model variable correlations, impute missing values, and capture the impact of varying time intervals. The complete data matrices thus obtained from the imputation task are used for downstream risk prediction tasks. This study aims to achieve more desirable imputation and prediction accuracy by performing both tasks simultaneously. We present a new multi-task deep neural network that incorporates the imputation task as an auxiliary task while performing risk prediction tasks. We validate the method on clinical time series imputation and in-hospital mortality prediction tasks using two publicly available EHR databases. The experimental results show that our method outperforms state-of-the-art imputation-prediction methods by significant margins. The results also empirically demonstrate that the incorporation of time decay mechanisms is a critical factor for superior imputation and prediction performance. The novel deep imputation-prediction network proposed in this study provides more accurate imputation and prediction results with EHR data. Future work should focus on developing more effective time decay mechanisms for simultaneously enhancing the imputation and prediction performance of multi-task learning models.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"135-140"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11670123/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142900697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fine-grained Patient Similarity Measuring using Contrastive Graph Similarity Networks.","authors":"Yuxi Liu, Zhenhao Zhang, Shaowen Qin, Flora D Salim, Jiang Bian, Antonio Jimeno Yepes","doi":"10.1109/ichi61247.2024.00009","DOIUrl":"10.1109/ichi61247.2024.00009","url":null,"abstract":"<p><p>Predictive analytics using Electronic Health Records (EHRs) have become an active research area in recent years, especially with the development of deep learning techniques. A popular EHR data analysis paradigm in deep learning is patient representation learning, which aims to learn a condensed mathematical representation of individual patients. However, EHR data are often inherently irregular, i.e., data entries were captured at different times as well as with different contents due to the individualized needs of each patient. Most of the work focused on the provision of deep neural networks with attention mechanisms that generate complete patient representations that can be readily used for downstream prediction tasks. However, such approaches fail to take patient similarity into account, which is generally used in clinical reasoning scenarios. This study presents a new Contrastive Graph Similarity Network for similarity calculation among patients in large EHR datasets. Particularly, we apply graph-based similarity analysis that explicitly extracts the clinical characteristics of each patient and aggregates the information of similar patients to generate rich patient representations. Experimental results on real-world EHR databases demonstrate the effectiveness and superiority of our method for the task of vital signs imputation and ICU patient deterioration prediction.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"1-10"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11654828/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142857143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Ethical Approach to Genomic Privacy Preserving Technology Development.","authors":"Lynette Hammond Gerido, Erman Ayday","doi":"10.1109/ichi61247.2024.00102","DOIUrl":"10.1109/ichi61247.2024.00102","url":null,"abstract":"<p><p>Demand for genomic research data and genetic testing results from cancer patients has grown exponentially. When a patient is diagnosed with a hereditary cancer syndrome, standard practice is for providers to encourage patients to discuss their results with their relatives and encourage those relatives to have clinical genetic testing and possibly participate in genetic research. Genomic research data and genetic testing results are being shared and connected in ways never imagined. Genomic data sharing is critical for advancing precision health and increasing diversity in global genome databases. However, these advancements often come with undesirable consequences, which call for additional privacy safeguards and research practices to protect hereditary cancer patients and their families because relatives who may have genomic information in common with the patient causing privacy risks to ripple throughout a kinship network. We propose to address this gap using an interdisciplinary approach integrating bioethical principles (autonomy, non-maleficence, beneficence, respect for persons, and equity) with data science techniques to mitigate privacy risk challenges.</p>","PeriodicalId":73284,"journal":{"name":"IEEE International Conference on Healthcare Informatics. IEEE International Conference on Healthcare Informatics","volume":"2024 ","pages":"638-641"},"PeriodicalIF":0.0,"publicationDate":"2024-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11976530/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143812955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}