17,294 Hits in 4.4 sec

Concept selection for phenotypes and diseases using learn to rank

Nigel Collier, Anika Oellrich, Tudor Groza
2015 Journal of Biomedical Semantics  
We evaluate four pipelines as stand-alone systems and then attempt to optimise semantic-type based performance using several learn-to-rank (LTR) approaches -three pairwise and one listwise.  ...  Phenotypes form the basis for determining the existence of a disease against the given evidence.  ...  Acknowledgements We gratefully acknowledge the kind permission of the ShARE/CLEF eHealth evaluation organisers for facilitating access to the ShARE/CLEF eHealth corpus used in our evaluation.  ... 
doi:10.1186/s13326-015-0019-z pmid:26034558 pmcid:PMC4450611 fatcat:ywypdbtuerdjphay3nd56yntvu

Diagnosis support systems for rare diseases: a scoping review

Carole Faviez, Xiaoyi Chen, Nicolas Garcelon, Antoine Neuraz, Bertrand Knebelmann, Rémi Salomon, Stanislas Lyonnet, Sophie Saunier, Anita Burgun
2020 Orphanet Journal of Rare Diseases  
Diagnosis targets varied from 1 rare disease to all rare diseases. Material used for diagnosis support consisted mostly of phenotype concepts, images or fluids.  ...  Fourteen studies provided online tools, most of which aimed to support the diagnosis of all rare diseases by considering queries based on phenotype concepts.  ...  Acknowledgments The authors acknowledge the ANR for the funding and would like to thank the C'IL-LICO members for their advice for the analysis and interpretation of the results.  ... 
doi:10.1186/s13023-020-01374-z pmid:32299466 pmcid:PMC7164220 fatcat:irqxt4ps45frlp7s6lnzr2g46y

Automated disease cohort selection using word embeddings from Electronic Health Records

Benjamin S Glicksberg, Riccardo Miotto, Kipp W Johnson, Khader Shameer, Li Li, Rong Chen, Joel T Dudley
2018 Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing  
Methodologies that automatically learn features from EHRs have been used for cohort selection as well.  ...  Using medical concepts as a query, we then rank patients by their proximity in the embedding space and automatically extract putative disease cohorts via a distance threshold.  ...  Acknowledgments We would like to thank the Mount Sinai Data Warehouse for facilitating data accessibility and the Mount Sinai Scientific Computing team for infrastructural support.  ... 
pmid:29218877 pmcid:PMC5788312 fatcat:fpydj3oyrjeh7ea3c7czhgjyj4

Automated disease cohort selection using word embeddings from Electronic Health Records

Benjamin S. Glicksberg, Riccardo Miotto, Kipp W. Johnson, Khader Shameer, Li Li, Rong Chen, Joel T. Dudley
2017 Biocomputing 2018  
Methodologies that automatically learn features from EHRs have been used for cohort selection as well.  ...  Using medical concepts as a query, we then rank patients by their proximity in the embedding space and automatically extract putative disease cohorts via a distance threshold.  ...  Acknowledgments We would like to thank the Mount Sinai Data Warehouse for facilitating data accessibility and the Mount Sinai Scientific Computing team for infrastructural support.  ... 
doi:10.1142/9789813235533_0014 fatcat:3y2pdtmhvjbcfe7lc7xe4j6avy

Phe2vec: Automated Disease Phenotyping based on Unsupervised Embeddings from Electronic Health Records [article]

Jessica K De Freitas, Kipp W Johnson, Eddye Golden, Girish N Nadkarni, Joel T Dudley, Erwin P Bottinger, Benjamin S Glicksberg, Riccardo Miotto
2020 medRxiv   pre-print
Objective: We introduce Phe2vec, an automated framework for disease phenotyping from electronic health records (EHRs) based on unsupervised learning.  ...  Disease phenotypes are then derived from a seed concept and its neighbors in the embedding space.  ...  Selection We used phenotypes based on embeddings to retrieve cohorts of patients for each disease.  ... 
doi:10.1101/2020.11.14.20231894 fatcat:hcdjknsc3jfnhn2wjy3nuwboui

Applying active learning to high-throughput phenotyping algorithms for electronic health records data

Yukun Chen, Robert J Carroll, Eugenia R McPeek Hinz, Anushi Shah, Anne E Eyler, Joshua C Denny, Hua Xu
2013 JAMIA Journal of the American Medical Informatics Association  
study is supported in part by grants from the National Library of Medicine (R01-LM010685), the National Cancer Institute (R01-CA141307), the National Institute of General Medical Sciences (R01-GM102282), and  ...  The synthetic derivative, from which the annotated records were derived, is supported by National Center for Research Resources UL1 RR024975, which is now at the National Center for Advancing Translational  ...  medical language system (UMLS) concept unique identifiers from clinical notes; and refined features, which included billing codes and UMLS concepts highly relevant to the specific phenotypes, as selected  ... 
doi:10.1136/amiajnl-2013-001945 pmid:23851443 pmcid:PMC3861916 fatcat:6opdsgkds5ehlfrjcgw265psp4

Integrative disease classification based on cross-platform microarray data

Chun-Chi Liu, Jianjun Hu, Mrinal Kalakrishnan, Haiyan Huang, Xianghong Zhou
2009 BMC Bioinformatics  
In addition, we systematically map textual annotations of datasets to concepts in Unified Medical Language System (UMLS), permitting quantitative analysis of the phenotype "distance" between datasets and  ...  Using the leave one dataset out cross validation, ManiSVM achieved the overall accuracy of 70.7% (68.6% precision and 76.9% recall) with many disease classes achieving the accuracy higher than 80%.  ...  Acknowledgements The authors thank Xuegong Zhang for helpful discussions and suggestions.  ... 
doi:10.1186/1471-2105-10-s1-s25 pmid:19208125 pmcid:PMC2648756 fatcat:romlcxl7xzgt3jdtzww5ks5tte

A Data-driven Framework for Learning and Visualizing Characteristics of Thrombotic Event Phenotypes from Clinical Texts [article]

Anahita Davoudi, Sy Hwang, Danielle Mowery, Audrey Yang
2021 medRxiv   pre-print
We developed a generalizable, data-driven framework for learning, characterizing, and visualizing clinical concepts from both radiology and discharge summaries predictive of thrombotic phenotypes.  ...  Although many investigators have developed targeted information extraction methods for identifying thrombotic phenotypes from radiology notes, these methods can be time consuming to train, require large  ...  We extend our gratitude to the open-source community for making their resources available. The cui embeddings can be found at:  ... 
doi:10.1101/2021.03.09.21253233 fatcat:p5fm7lzf4raa3lqfy3o3no46ce

Next generation phenotyping using narrative reports in a rare disease clinical data warehouse

Nicolas Garcelon, Antoine Neuraz, Rémi Salomon, Nadia Bahi-Buisson, Jeanne Amiel, Capucine Picard, Nizar Mahlaoui, Vincent Benoit, Anita Burgun, Bastien Rance
2018 Orphanet Journal of Rare Diseases  
We have developed a method to detect phenotypes associated with a group of patients using medical concepts extracted from free-text clinical narratives.  ...  Methods: We leveraged the frequency and TF-IDF to explore the association between clinical phenotypes and rare diseases.  ...  and agree to be accountable for all aspects of the work.  ... 
doi:10.1186/s13023-018-0830-6 pmid:29855327 pmcid:PMC5984368 fatcat:vgagwi6sabgnbcctpx2uz3rrka

Learning to Recognize Phenotype Candidates in the Auto-Immune Literature Using SVM Re-Ranking

Nigel Collier, Mai-vu Tran, Hoang-quynh Le, Quang-Thuy Ha, Anika Oellrich, Dietrich Rebholz-Schuhmann, Luis M. Rocha
2013 PLoS ONE  
We observed the advantage of using SVM-based learn-to-rank for sequence label combination over maximum entropy and a priority list approach.  ...  Using partial matching the best micro-averaged F-score for phenotypes and five other entity classes was 79.9%.  ...  We observed the advantage of using SVM learn-to-rank for hypothesis resolution and using all resources.  ... 
doi:10.1371/journal.pone.0072965 pmid:24155869 pmcid:PMC3796529 fatcat:jcy2fr7yrjgwngtfqvr4vhczdq

Identification of Similar Patients Through Medical Concept Embedding from Electronic Health Records: A Feasibility Study for Rare Disease Diagnosis [chapter]

Xiaoyi Chen, Carole Faviez, Marc Vincent, Nicolas Garcelon, Sophie Saunier, Anita Burgun
2021 Studies in Health Technology and Informatics  
To identify patients with similar clinical profiles and derive insights from the records and outcomes of similar patients can help fast and precise diagnosis and other clinical decisions for rare diseases  ...  In this paper, we introduce the methods developed in the context of rare disease screening/diagnosis from clinical data warehouse using medical concept embedding and adjusted aggregations.  ...  As we considered the task of identifying NPH-related ciliopathies from other diseases, we selected NPH patients with sufficient information in EHRs, and re used the control patients with overlapping phenotypes  ... 
doi:10.3233/shti210241 pmid:34042646 fatcat:pbviuy5szzfxxgh7v4hbhsf4ti

Integrating biomedical research and electronic health records to create knowledge-based biologically meaningful machine-readable embeddings

Charlotte A. Nelson, Atul J. Butte, Sergio E. Baranzini
2019 Nature Communications  
Then an unsupervised machine-learning algorithm creates Propagated SPOKE Entry Vectors (PSEVs) that encode the importance of each SPOKE node for any code in the EHRs.  ...  In order to advance precision medicine, detailed clinical features ought to be described in a way that leverages current knowledge.  ...  In addition, we would like to thank Achievement Rewards for College Scientists (ARCS) Scholarship and the NHI BMI Training Grant (T32 GM067547/ 4T32GM067547-14).  ... 
doi:10.1038/s41467-019-11069-0 pmid:31292438 pmcid:PMC6620318 fatcat:b7pc3osxrrgefb34a4du2or42m

Comparison of automated candidate gene prediction systems using genes implicated in type 2 diabetes by genome-wide association studies

Erdahl T Teber, Jason Y Liu, Sara Ballouz, Diane Fatkin, Merridee A Wouters
2009 BMC Bioinformatics  
Furthermore, they can be used to filter statistically less-well-supported genetic data to select more likely candidates.  ...  Automated candidate gene prediction systems allow geneticists to hone in on disease genes more rapidly by identifying the most probable candidate genes linked to the disease phenotypes under investigation  ...  Acknowledgements The authors wish to acknowledge funding from the Ronald Geoffrey Arnott Foundation.  ... 
doi:10.1186/1471-2105-10-s1-s69 pmid:19208173 pmcid:PMC2648789 fatcat:jo6drnkmvncthg5aqhdxfstxp4

Mimvec: a deep learning approach for analyzing the human phenome

Mingxin Gan, Wenran Li, Wanwen Zeng, Xiaojian Wang, Rui Jiang
2017 BMC Systems Biology  
We finally used the derived phenotype similarities with genomic data to prioritize candidate genes and demonstrated advantages of this method over existing ones.  ...  We further derived pairwise phenotype similarities between 7988 human inherited diseases using their vector presentations.  ...  Acknowledgements We thank the Institute for Data Science at Tsinghua University for providing a computer cluster that makes this research possible.  ... 
doi:10.1186/s12918-017-0451-z pmid:28950906 pmcid:PMC5615244 fatcat:uwthfufnjfhoveenkh6lsa3gfm

SoftPanel: a website for grouping diseases and related disorders for generation of customized panels

Likun Wang, Cong Zhang, Johnathan Watkins, Yan Jin, Michael McNutt, Yuxin Yin
2016 BMC Bioinformatics  
Various methods of retrieval including a keyword search, browsing of an arborized list of International Classification of Diseases, 10th revision (ICD-10) codes or using disorder phenotypic similarities  ...  Although tools to prioritize candidate disease genes have been developed, the great majority of these require prior knowledge and a set of seed genes as input, which is only possible for diseases with  ...  All of these scores are then used to construct an eight-dimensional vector, which is used as the input for performing machine learning.  ... 
doi:10.1186/s12859-016-0998-5 pmid:27044653 pmcid:PMC4820874 fatcat:uowugoekybcernuvnjifxifjke
« Previous Showing results 1 — 15 out of 17,294 results