Review
Computer Science, Artificial Intelligence
E. A. Nismi Mol, M. B. Santosh Kumar
Summary: This paper reviews different computational approaches, including rule-based and learning-based methods, and explores various techniques, features, tools, datasets, and evaluation metrics adopted for knowledge extraction from the most relevant literature.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Artificial Intelligence
Yangning Li, Yinghui Li, Xi Chen, Hai-Tao Zheng, Ying Shen
Summary: This article addresses two major problems in Open Relation Extraction (OpenRE): insufficient capacity to discriminate between known and novel relations, and the inability to label human-readable and meaningful types for novel relations. To solve these issues, the Active Relation Discovery (ARD) framework, which includes relational outlier detection and active learning, is proposed. Extensive experiments demonstrate the superior performance of ARD in both conventional and general OpenRE settings.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Information Systems
Xuechao Du, Xiang Pan, Yinzhi Cao, Boyuan He, Gan Fan, Yan Chen, Daigang Xu
Summary: The legitimacy of Android apps accessing private information depends on whether the app provides sufficient semantics to justify the access. Existing analysis methods are limited to coarse-grained app-level analysis and cannot address the correctness of specific app behaviors under different runtime contexts. To address this, we propose FlowCog, an automated system that extracts semantics related to information flows and correlates them with given flows.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Wilson Lau, Kevin Lybarger, Martin L. Gunn, Meliha Yetisgen
Summary: This paper presents a study on semantic annotation of radiology reports using deep learning models. The authors built a corpus of 500 annotated reports and extracted trigger words and argument entities using the advanced deep learning architecture BERT. They predicted the linkages between triggers and argument entities and demonstrated the model's generalizability through testing on an external validation set.
JOURNAL OF DIGITAL IMAGING
(2023)
Article
Computer Science, Artificial Intelligence
Heyan Huang, Ming Lei, Chong Feng
Summary: This paper proposes using corpus subgraphs and sentence subgraphs to obtain linguistic knowledge and classification knowledge, building a relation knowledge graph to extract relations from sentences, and treating multiple relation extraction as a reasoning process for knowledge completion.
Article
Computer Science, Artificial Intelligence
Ming Lei, Heyan Huang, Chong Feng
Summary: The proposed MGSR model successfully bridges the semantic gap between low-level and high-level semantic abstraction by learning multi-granularity semantic representations at word, phrase, and sentence levels. By segmenting sentences into entity chunks and context chunks, and utilizing different self-attention mechanisms for learning semantic representations, the model achieves superior performance in relation extraction tasks.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Computer Science, Interdisciplinary Applications
Giridhar Kaushik Ramachandran, Kevin Lybarger, Yaya Liu, Diwakar Mahajan, Jennifer J. Liang, Ching-Huei Tsou, Meliha Yetisgen, Ozlem Uzuner
Summary: An accurate and detailed record of patient medications, including changes, is crucial for healthcare providers to provide appropriate care. This study focuses on the automatic extraction of medication change information from clinical notes. The proposed systems, based on BERT models, successfully improve the classification performance of medication changes compared to previous work.
JOURNAL OF BIOMEDICAL INFORMATICS
(2023)
Article
Biochemistry & Molecular Biology
Sicheng Zhou, Nan Wang, Liwei Wang, Ju Sun, Anne Blaes, Hongfang Liu, Rui Zhang
Summary: This study evaluated the generalizability of CancerBERT, a Transformer-based clinical NLP model, in a breast cancer phenotype extraction task. The results showed that CancerBERT achieved the best performance among different clinical institutes, and the model developed in one institute and fine-tuned in another also performed well.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2023)
Article
Computer Science, Interdisciplinary Applications
Ying Hu, Yanping Chen, Yongbin Qin, Ruizhang Huang
Summary: Biomedical Relation Extraction (BioRE) is an important task in automatically extracting semantic relations for given entity pairs. Current popular methods often use pretrained language models for feature extraction, but they suffer from overlapping semantics. This study proposes an Entity-oriented Representation (EoR) model that enhances the discriminability between entity pairs and achieves state-of-the-art performance in multiple BioRE tasks.
JOURNAL OF BIOMEDICAL INFORMATICS
(2023)
Article
Computer Science, Interdisciplinary Applications
Nicholas Altieri, Briton Park, Mara Olson, John DeNero, Anobel Y. Odisho, Bin Yu
Summary: The study aims to build an accurate machine learning system for classifying tumor attributes from cancer pathology reports. By using enriched labeling schemes and hierarchical methods, the efficiency of machine learning for pathology report attribute classification can be greatly increased.
JOURNAL OF BIOMEDICAL INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Changsen Yuan, Heyan Huang, Chong Feng, Qianwen Cao
Summary: The PGCN-EA model addresses the issues of redundant syntactic information and wrong dependency parsing results in GCN by utilizing piecewise adjacency matrix based on entity pair and Edge-level Attention. The model achieves the best PR curve on a benchmark dataset compared to seven baseline models.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Kelvin Lai, Jeremy R. Porter, Mike Amodeo, David Miller, Michael Marston, Saman Armal
Summary: This study developed a hybrid Named-Entity Recognition (NER) model using Natural Language Processing (NLP) to extract detailed flooding information and risk reduction projects from newspapers across the United States. The model achieved high accuracy in extracting information such as street closures, project costs, and metrics, expanding upon previous works and covering a wide geographical area.
INFORMATION PROCESSING & MANAGEMENT
(2022)
Article
Medicine, General & Internal
Gyuseon Song, Su Jin Chung, Ji Yeon Seo, Sun Young Yang, Eun Hyo Jin, Goh Eun Chung, Sung Ryul Shim, Soonok Sa, Moongi Simon Hong, Kang Hyun Kim, Eunchan Jang, Chae Won Lee, Jung Ho Bae, Hyun Wook Han
Summary: This study developed a natural language processing (NLP) pipeline that can automatically extract clinical information about gastric diseases from esophagogastroduodenoscopy (EGD) reports. The pipeline's accuracy and reliability have been validated and demonstrated. The study results also revealed the demographic characteristics of patients with gastric diseases and the extent and locations of the diseases.
JOURNAL OF CLINICAL MEDICINE
(2022)
Article
Computer Science, Interdisciplinary Applications
Abbas Akkasi, Mari-Francine Moens
Summary: Identifying causal relationships between events or entities in biomedical texts is crucial for creating scientific knowledge bases and is a fundamental task in NLP. Despite being an open problem in artificial intelligence, there is increasing research attention on this issue, with new techniques like deep neural networks showing promise in addressing it. Enhancements in state-of-the-art systems can be achieved through data augmentation techniques such as random oversampling to address class imbalance.
JOURNAL OF BIOMEDICAL INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Qian Wan, Shangheng Du, Yaqi Liu, Jing Fang, Luona Wei, Sannyuya Liu
Summary: The paper proposes a document-level relation extraction model based on the Hierarchical Dependency Tree and Bridge Path (HDT-BP), which improves the performance of relation extraction by independently extracting fine-grained features for each hierarchy and modeling the bridge path feature.
KNOWLEDGE-BASED SYSTEMS
(2023)
Review
Health Care Sciences & Services
Su Golder, Robin Stevens, Karen O'Connor, Richard James, Graciela Gonzalez-Hernandez
Summary: Social media data is increasingly used in health research. This study aims to identify different methods to extract race or ethnicity from social media and report on the challenges of using these methods. Through a scoping review, the study found that there is currently no standard approach to extract or infer the race or ethnicity of Twitter users, and there are challenges in terms of accuracy and ethical issues.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2022)
Article
Public, Environmental & Occupational Health
Sirish Namilae, Yuxuan Wu, Anuj Mubayi, Ashok Srinivasan, Matthew Scotch
Summary: Conventional models fail to explain superspreading patterns on flights, with passenger movement playing a significant role in infection spread. The use of FFP2/N95 masks is more effective in reducing infection risk, and leaving middle seats vacant is also effective. The results emphasize the importance of implementing stricter guidelines to reduce aviation-related transmission.
TRAVEL MEDICINE AND INFECTIOUS DISEASE
(2022)
Article
Public, Environmental & Occupational Health
Ari Z. Klein, Steven Meanley, Karen O'Connor, Jose A. Bauermeister, Graciela Gonzalez-Hernandez
Summary: By developing an automated natural language processing (NLP) pipeline, MSM at risk of HIV acquisition can be identified on Twitter, laying the groundwork for targeted PrEP-related interventions for this population on a large scale.
JMIR PUBLIC HEALTH AND SURVEILLANCE
(2022)
Article
Public, Environmental & Occupational Health
Su Golder, Davy Weissenbacher, Karen O'Connor, Sean Hennessy, Robert Gross, Graciela Gonzalez Hernandez
Summary: Social media analysis revealed that the main reason for discontinuation of statin therapy was patient experience of adverse events, with musculoskeletal and connective tissue disorders being the most common. 60% of posters identified as female, with the most common age category being 55-64 years. The unique patient perspectives found on social media may provide valuable insights for interventions to reduce medication discontinuation.
Article
Immunology
Jeb Jones, Justin Knox, Steven Meanley, Cui Yang, David W. Lounsbury, Terry T. Huang, Jose Bauermeister, Graciela Gonzalez-Hernandez, Victoria Frye, Christian Grov, Viraj Patel, Stefan D. Baral, Patrick S. Sullivan, Sheree R. Schwartz
Summary: The use of digital technology in HIV-related interventions and implementation strategies is significant and presents both challenges and opportunities. This article explores five case studies that highlight the role of technology in HIV-related implementation research, emphasizing the importance of study design, outcome measurement, and equity.
JAIDS-JOURNAL OF ACQUIRED IMMUNE DEFICIENCY SYNDROMES
(2022)
Article
Psychology, Clinical
Melanie L. Kornides, Sarah Badlis, Katharine J. Head, Mary Putt, Joseph Cappella, Graciela Gonzalez-Hernadez
Summary: Nearly a quarter of #HPV Tweets contain disinformation or misinformation about the HPV vaccine, with adverse health effects, mandatory vaccination, and vaccine inefficacy being the most prevalent categories. These misleading tweets are more likely to be retweeted than supportive tweets.
JOURNAL OF BEHAVIORAL MEDICINE
(2023)
Article
Microbiology
Courtney L. Collins, Simona Kraberger, Rafaela S. Fontenele, Temitope O. C. Faleye, Deborah Adams, Sangeet Adhikari, Helen Sandrolini, Sarah Finnerty, Rolf U. Halden, Matthew Scotch, Arvind Varsani
Summary: In this study, multiple viral infections including anelloviruses, papillomavirus, and influenza viruses were identified from nasopharyngeal swabs.
MICROBIOLOGY RESOURCE ANNOUNCEMENTS
(2022)
Letter
Health Care Sciences & Services
Ari Z. Klein, Shriya Kunatharaju, Karen O'Connor, Graciela Gonzalez-Hernandez
JOURNAL OF MEDICAL INTERNET RESEARCH
(2023)
Letter
Health Care Sciences & Services
Ari Z. Klein, Shriya Kunatharaju, Karen O'Connor, Graciela Gonzalez-Hernandez
JOURNAL OF MEDICAL INTERNET RESEARCH
(2023)
Article
Microbiology
Ainsley R. Chapman, Jillian M. Wright, Nicole A. Kaiser, Peter M. Jones, Erin M. Driver, Rolf U. Halden, Arvind Varsani, Matthew Scotch, Temitope O. C. Faleye
Summary: This paper describes the genome of MAZ-Nov-2020, a microvirus identified from municipal wastewater in Maricopa County, Arizona, USA, in November 2020. The genome is 4,696 nucleotides long, with a GC content of 56% and a coverage of 3,641x. It encodes major capsid protein, endolysin, replication initiator protein, and two hypothetical proteins, one of which is predicted to be a membrane-associated multiheme cytochrome c.
MICROBIOLOGY RESOURCE ANNOUNCEMENTS
(2023)
Review
Health Care Sciences & Services
Su Golder, Karen O'Connor, Yunwen Wang, Graciela Gonzalez Hernandez
Summary: This study aims to evaluate and characterize the use of social media in adverse drug event detection and pharmacovigilance compared to other data sources. By comparing social media data with other sources, the added value of social media in monitoring adverse drug events can be concluded.
JMIR RESEARCH PROTOCOLS
(2023)
Article
Computer Science, Information Systems
Saif Khairat, Sue S. Feldman, Arif Rana, Mohammad Faysel, Saptarshi Purkayastha, Matthew Scotch, Christina Eldredge
Summary: This article presents the foundational domains and corresponding competencies developed by AMIA's Academic Forum Baccalaureate Education Committee (BEC) for undergraduate health informatics education.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2023)
Article
Infectious Diseases
Temitope O. C. Faleye, Erin M. Driver, Devin A. Bowes, Abriana Smith, Nicole A. Kaiser, Jillian M. Wright, Ainsley R. Chapman, Rolf U. Halden, Arvind Varsani, Matthew Scotch
Summary: In this study, CPV genomes were sequenced from dog feces collected in poop bags, and a variant of CPV-2c with amino acid substitutions in NS1 and NS2 was identified in Arizona, USA in June 2022. This genome is the only CPV genome described in the USA from the 2022 season, despite reports of CPV outbreaks and fatalities in dogs. Further studies and experimental research are needed to enhance our understanding of the evolutionary process of CPV.
TRANSBOUNDARY AND EMERGING DISEASES
(2023)
Article
Health Care Sciences & Services
Ari Z. Klein, Karen O'Connor, Lisa D. Levine, Graciela Gonzalez-Hernandez
Summary: This study examined the utility of Twitter data for analyzing the outcomes of pregnancies where beta-blockers were taken. The results suggest that Twitter can be a useful resource for cohort studies on drug safety during pregnancy.
JMIR FORMATIVE RESEARCH
(2022)
Article
Health Care Sciences & Services
Ari Z. Klein, Karen O'Connor, Graciela Gonzalez-Hernandez
Summary: This preliminary study aimed to use Twitter data to identify a cohort for epidemiologic studies of COVID-19 vaccination during pregnancy. By developing regular expressions and utilizing natural language processing tools, the study identified users who reported receiving COVID-19 vaccination during pregnancy and their pregnancy outcomes. Manual verification confirmed a portion of users received vaccination during pregnancy and reported outcomes, suggesting that Twitter can serve as a complementary resource to generate acceptance of COVID-19 vaccination in pregnant populations.
JMIR FORMATIVE RESEARCH
(2022)
