Dataset Information

A semi-supervised approach for rapidly creating clinical biomarker phenotypes in the UK Biobank using different primary care EHR and clinical terminology systems.

ABSTRACT:

Objectives

The UK Biobank (UKB) is making primary care electronic health records (EHRs) for 500 000 participants available for COVID-19-related research. Data are extracted from four sources, recorded using five clinical terminologies and stored in different schemas. The aims of our research were to: (a) develop a semi-supervised approach for bootstrapping EHR phenotyping algorithms in UKB EHR, and (b) to evaluate our approach by implementing and evaluating phenotypes for 31 common biomarkers.

Materials and methods

We describe an algorithmic approach to phenotyping biomarkers in primary care EHR involving (a) bootstrapping definitions using existing phenotypes, (b) excluding generic, rare, or semantically distant terms, (c) forward-mapping terminology terms, (d) expert review, and (e) data extraction. We evaluated the phenotypes by assessing the ability to reproduce known epidemiological associations with all-cause mortality using Cox proportional hazards models.

Results

We created and evaluated phenotyping algorithms for 31 biomarkers many of which are directly related to COVID-19 complications, for example diabetes, cardiovascular disease, respiratory disease. Our algorithm identified 1651 Read v2 and Clinical Terms Version 3 terms and automatically excluded 1228 terms. Clinical review excluded 103 terms and included 44 terms, resulting in 364 terms for data extraction (sensitivity 0.89, specificity 0.92). We extracted 38 190 682 events and identified 220 978 participants with at least one biomarker measured.

Discussion and conclusion

Bootstrapping phenotyping algorithms from similar EHR can potentially address pre-existing methodological concerns that undermine the outputs of biomarker discovery pipelines and provide research-quality phenotyping algorithms.

SUBMITTER: Denaxas S

PROVIDER: S-EPMC7717266 | biostudies-literature | 2020 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A semi-supervised approach for rapidly creating clinical biomarker phenotypes in the UK Biobank using different primary care EHR and clinical terminology systems.

Denaxas Spiros S Shah Anoop D AD Mateen Bilal A BA Kuan Valerie V Quint Jennifer K JK Fitzpatrick Natalie N Torralbo Ana A Fatemifar Ghazaleh G Hemingway Harry H

JAMIA open 20201205 4

<h4>Objectives</h4>The UK Biobank (UKB) is making primary care electronic health records (EHRs) for 500 000 participants available for COVID-19-related research. Data are extracted from four sources, recorded using five clinical terminologies and stored in different schemas. The aims of our research were to: (a) develop a semi-supervised approach for bootstrapping EHR phenotyping algorithms in UKB EHR, and (b) to evaluate our approach by implementing and evaluating phenotypes for 31 common bioma ...[more]

PMID: 33619467

Similar Datasets

Project description:BackgroundModern molecular profiling techniques are yielding vast amounts of data from patient samples that could be utilized with machine learning methods to provide important biological insights and improvements in patient outcomes. Unsupervised methods have been successfully used to identify molecularly-defined disease subtypes. However, these approaches do not take advantage of potential additional clinical outcome information. Supervised methods can be implemented when training classes are apparent (e.g., responders or non-responders to treatment). However, training classes can be difficult to define when assessing relative benefit of one therapy over another using gold standard clinical endpoints, since it is often not clear how much benefit each individual patient receives.ResultsWe introduce an iterative approach to binary classification tasks based on the simultaneous refinement of training class labels and classifiers towards self-consistency. As training labels are refined during the process, the method is well suited to cases where training class definitions are not obvious or noisy. Clinical data, including time-to-event endpoints, can be incorporated into the approach to enable the iterative refinement to identify molecular phenotypes associated with a particular clinical variable. Using synthetic data, we show how this approach can be used to increase the accuracy of identification of outcome-related phenotypes and their associated molecular attributes. Further, we demonstrate that the advantages of the method persist in real world genomic datasets, allowing the reliable identification of molecular phenotypes and estimation of their association with outcome that generalizes to validation datasets. We show that at convergence of the iterative refinement, there is a consistent incorporation of the molecular data into the classifier yielding the molecular phenotype and that this allows a robust identification of associated attributes and the underlying biological processes.ConclusionsThe consistent incorporation of the structure of the molecular data into the classifier helps to minimize overfitting and facilitates not only good generalization of classification and molecular phenotypes, but also reliable identification of biologically relevant features and elucidation of underlying biological processes.

Project description:BackgroundClinically, behavior, cognitive, and mental functions are affected during the neurodegenerative disease progression. To date, the molecular pathogenesis of these complex disease is still unclear. With the rapid development of sequencing technologies, it is possible to delicately decode the molecular mechanisms corresponding to different clinical phenotypes at the genome-wide transcriptomic level using computational methods. Our previous studies have shown that it is difficult to distinguish disease genes from non-disease genes. Therefore, to precisely explore the molecular pathogenesis under complex clinical phenotypes, it is better to identify biomarkers corresponding to different disease stages or clinical phenotypes. So, in this study, we designed a label propagation-based semi-supervised feature selection approach (LPFS) to prioritize disease-associated genes corresponding to different disease stages or clinical phenotypes.MethodsIn this study, we pioneering put label propagation clustering and feature selection into one framework and proposed label propagation-based semi-supervised feature selection approach. LPFS prioritizes disease genes related to different disease stages or phenotypes through the alternative iteration of label propagation clustering based on sample network and feature selection with gene expression profiles. Then the GO and KEGG pathway enrichment analysis were carried as well as the gene functional analysis to explore molecular mechanisms of specific disease phenotypes, thus to decode the changes in individual behavioral and mental characteristics during neurodegenerative disease progression.ResultsLarge amounts of experiments were conducted to verify the performance of LPFS with Huntington's gene expression data. Experimental results shown that LPFS performs better in comparison with the-state-of-art methods. GO and KEGG enrichment analysis of key gene sets shown that TGF-beta signaling pathway, cytokine-cytokine receptor interaction, immune response, and inflammatory response were gradually affected during the Huntington's disease progression. In addition, we found that the expression of SLC4A11, ZFP474, AMBP, TOP2A, PBK, CCDC33, APSL, DLGAP5, and Al662270 changed seriously by the development of the disease.ConclusionsIn this study, we designed a label propagation-based semi-supervised feature selection model to precisely selected key genes of different disease phenotypes. We conducted experiments using the model with Huntington's disease mice gene expression data to decode the mechanisms of it. We found many cell types, including astrocyte, microglia, and GABAergic neuron, could be involved in the pathological process.

Project description:Background and objectiveElectronic Health Record (EHR) data often include observation records that are unlikely to represent the "truth" about a patient at a given clinical encounter. Due to their high throughput, examples of such implausible observations are frequent in records of laboratory test results and vital signs. Outlier detection methods can offer low-cost solutions to flagging implausible EHR observations. This article evaluates the utility of a semi-supervised encoding approach (super-encoding) for constructing non-linear exemplar data distributions from EHR observation data and detecting non-conforming observations as outliers.MethodsTwo hypotheses are tested using experimental design and non-parametric hypothesis testing procedures: (1) adding demographic features (e.g., age, gender, race/ethnicity) can increase precision in outlier detection, (2) sampling small subsets of the large EHR data can increase outlier detection by reducing noise-to-signal ratio. The experiments involved applying 492 encoder configurations (involving different input features, architectures, sampling ratios, and error margins) to a set of 30 datasets EHR observations including laboratory tests and vital sign records extracted from the Research Patient Data Registry (RPDR) from Partners HealthCare.ResultsResults are obtained from (30 × 492) 14,760 encoders. The semi-supervised encoding approach (super-encoding) outperformed conventional autoencoders in outlier detection. Adding age of the patient at the observation (encounter) to the baseline encoder that only included observation value as the input feature slightly improved outlier detection. Top-nine performing encoders are introduced. The best outlier detection performance was from a semi-supervised encoder, with observation value as the single feature and a single hidden layer, built on one percent of the data and one percent reconstruction error. At least one encoder configurations had a Youden's J index higher than 0.9999 for all 30 observation types.ConclusionGiven the multiplicity of distributions for a single observation in EHR data (i.e., same observation represented with different names or units), as well as non-linearity of human observations, encoding offers huge promises for outlier detection in large-scale data repositories. https://github.com/hestiri/superencoder.

Dataset Information

A semi-supervised approach for rapidly creating clinical biomarker phenotypes in the UK Biobank using different primary care EHR and clinical terminology systems.

Objectives

Materials and methods

Results

Discussion and conclusion

Publications

A semi-supervised approach for rapidly creating clinical biomarker phenotypes in the UK Biobank using different primary care EHR and clinical terminology systems.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets