Publications by authors named "Luke T Slater"

5 Publications

  • Page 1 of 1

Improved characterisation of clinical text through ontology-based vocabulary expansion.

J Biomed Semantics 2021 04 12;12(1). Epub 2021 Apr 12.

Institute of Cancer and Genomic Sciences, College of Medical and Dental Sciences, University of Birmingham, Birmingham, B15 2TT, UK.

Background: Biomedical ontologies contain a wealth of metadata that constitutes a fundamental infrastructural resource for text mining. For several reasons, redundancies exist in the ontology ecosystem, which lead to the same entities being described by several concepts in the same or similar contexts across several ontologies. While these concepts describe the same entities, they contain different sets of complementary metadata. Linking these definitions to make use of their combined metadata could lead to improved performance in ontology-based information retrieval, extraction, and analysis tasks.

Results: We develop and present an algorithm that expands the set of labels associated with an ontology class using a combination of strict lexical matching and cross-ontology reasoner-enabled equivalency queries. Across all disease terms in the Disease Ontology, the approach found 51,362 additional labels, more than tripling the number defined by the ontology itself. Manual validation by a clinical expert on a random sampling of expanded synonyms over the Human Phenotype Ontology yielded a precision of 0.912. Furthermore, we found that annotating patient visits in MIMIC-III with an extended set of Disease Ontology labels led to semantic similarity score derived from those labels being a significantly better predictor of matching first diagnosis, with a mean average precision of 0.88 for the unexpanded set of annotations, and 0.913 for the expanded set.

Conclusions: Inter-ontology synonym expansion can lead to a vast increase in the scale of vocabulary available for text mining applications. While the accuracy of the extended vocabulary is not perfect, it nevertheless led to a significantly improved ontology-based characterisation of patients from text in one setting. Furthermore, where run-on error is not acceptable, the technique can be used to provide candidate synonyms which can be checked by a domain expert.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s13326-021-00241-5DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8042947PMC
April 2021

Towards similarity-based differential diagnostics for common diseases.

Comput Biol Med 2021 Apr 1;133:104360. Epub 2021 Apr 1.

College of Medical and Dental Sciences, Institute of Cancer and Genomic Sciences, University of Birmingham, UK; Institute of Translational Medicine, University Hospitals Birmingham, NHS Foundation Trust, UK; NIHR Experimental Cancer Medicine Centre, UK; NIHR Surgical Reconstruction and Microbiology Research Centre, UK; NIHR Biomedical Research Centre, UK; MRC Health Data Research UK (HDR UK) Midlands, UK; University Hospitals Birmingham NHS Foundation Trust, Edgbaston, Birmingham, UK.

Ontology-based phenotype profiles have been utilised for the purpose of differential diagnosis of rare genetic diseases, and for decision support in specific disease domains. Particularly, semantic similarity facilitates diagnostic hypothesis generation through comparison with disease phenotype profiles. However, the approach has not been applied for differential diagnosis of common diseases, or generalised clinical diagnostics from uncurated text-derived phenotypes. In this work, we describe the development of an approach for deriving patient phenotype profiles from clinical narrative text, and apply this to text associated with MIMIC-III patient visits. We then explore the use of semantic similarity with those text-derived phenotypes to classify primary patient diagnosis, comparing the use of patient-patient similarity and patient-disease similarity using phenotype-disease profiles previously mined from literature. We also consider a combined approach, in which literature-derived phenotypes are extended with the content of text-derived phenotypes we mined from 500 patients. The results reveal a powerful approach, showing that in one setting, uncurated text phenotypes can be used for differential diagnosis of common diseases, making use of information both inside and outside the setting. While the methods themselves should be explored for further optimisation, they could be applied to a variety of clinical tasks, such as differential diagnosis, cohort discovery, document and text classification, and outcome prediction.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.compbiomed.2021.104360DOI Listing
April 2021

A fast, accurate, and generalisable heuristic-based negation detection algorithm for clinical text.

Comput Biol Med 2021 Mar 16;130:104216. Epub 2021 Jan 16.

College of Medical and Dental Sciences, Institute of Cancer and Genomic Sciences, University of Birmingham, UK; Institute of Translational Medicine, University Hospitals Birmingham, NHS Foundation Trust, UK; NIHR Experimental Cancer Medicine Centre, UK; NIHR Surgical Reconstruction and Microbiology Research Centre, UK; NIHR Biomedical Research Centre, UK; MRC Health Data Research UK (HDR UK) Midlands, UK; University Hospitals Birmingham NHS Foundation Trust, Edgbaston, Birmingham, UK.

Negation detection is an important task in biomedical text mining. Particularly in clinical settings, it is of critical importance to determine whether findings mentioned in text are present or absent. Rule-based negation detection algorithms are a common approach to the task, and more recent investigations have resulted in the development of rule-based systems utilising the rich grammatical information afforded by typed dependency graphs. However, interacting with these complex representations inevitably necessitates complex rules, which are time-consuming to develop and do not generalise well. We hypothesise that a heuristic approach to determining negation via dependency graphs could offer a powerful alternative. We describe and implement an algorithm for negation detection based on grammatical distance from a negatory construct in a typed dependency graph. To evaluate the algorithm, we develop two testing corpora comprised of sentences of clinical text extracted from the MIMIC-III database and documents related to hypertrophic cardiomyopathy patients routinely collected at University Hospitals Birmingham NHS trust. Gold-standard validation datasets were built by a combination of human annotation and examination of algorithm error. Finally, we compare the performance of our approach with four other rule-based algorithms on both gold-standard corpora. The presented algorithm exhibits the best performance by f-measure over the MIMIC-III dataset, and a similar performance to the syntactic negation detection systems over the HCM dataset. It is also the fastest of the dependency-based negation systems explored in this study. Our results show that while a single heuristic approach to dependency-based negation detection is ignorant to certain advanced cases, it nevertheless forms a powerful and stable method, requiring minimal training and adaptation between datasets. As such, it could present a drop-in replacement or augmentation for many-rule negation approaches in clinical text-mining pipelines, particularly for cases where adaptation and rule development is not required or possible.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.compbiomed.2021.104216DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7910278PMC
March 2021

Towards semantic interoperability: finding and repairing hidden contradictions in biomedical ontologies.

BMC Med Inform Decis Mak 2020 12 15;20(Suppl 10):311. Epub 2020 Dec 15.

Computer, Electrical and Mathematical Sciences and Engineering Division, Computational Bioscience Research Center, King Abdullah University of Science and Technology, Thuwal, 23955, Saudi Arabia.

Background: Ontologies are widely used throughout the biomedical domain. These ontologies formally represent the classes and relations assumed to exist within a domain. As scientific domains are deeply interlinked, so too are their representations. While individual ontologies can be tested for consistency and coherency using automated reasoning methods, systematically combining ontologies of multiple domains together may reveal previously hidden contradictions.

Methods: We developed a method that tests for hidden unsatisfiabilities in an ontology that arise when combined with other ontologies. For this purpose, we combined sets of ontologies and use automated reasoning to determine whether unsatisfiable classes are present. In addition, we designed and implemented a novel algorithm that can determine justifications for contradictions across extremely large and complicated ontologies, and use these justifications to semi-automatically repair ontologies by identifying a small set of axioms that, when removed, result in a consistent and coherent set of ontologies.

Results: We tested the mutual consistency of the OBO Foundry and the OBO ontologies and find that the combined OBO Foundry gives rise to at least 636 unsatisfiable classes, while the OBO ontologies give rise to more than 300,000 unsatisfiable classes. We also applied our semi-automatic repair algorithm to each combination of OBO ontologies that resulted in unsatisfiable classes, finding that only 117 axioms could be removed to account for all cases of unsatisfiability across all OBO ontologies.

Conclusions: We identified a large set of hidden unsatisfiability across a broad range of biomedical ontologies, and we find that this large set of unsatisfiable classes is the result of a relatively small amount of axiomatic disagreements. Our results show that hidden unsatisfiability is a serious problem in ontology interoperability; however, our results also provide a way towards more consistent ontologies by addressing the issues we identified.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1186/s12911-020-01336-2DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7736131PMC
December 2020

A Practical Guide to Assess the Reproducibility of Echocardiographic Measurements.

J Am Soc Echocardiogr 2019 12 22;32(12):1505-1515. Epub 2019 Oct 22.

Institute of Cardiovascular Sciences, University of Birmingham, Birmingham, United Kingdom; University Hospitals Birmingham NHS Foundation Trust, Birmingham, United Kingdom. Electronic address:

Echocardiography plays an essential role in the diagnosis and assessment of cardiovascular disease. Measurements derived from echocardiography are also used to determine the severity of disease, its progression over time, and to aid in the choice of optimal therapy. It is therefore clinically important that echocardiographic measurements be reproducible, repeatable, and reliable. There are a variety of statistical tests available to assess these parameters, and in this article the authors summarize those available for use by echocardiographers to improve their clinical practice. Correlation coefficients, linear regression, Bland-Altman plots, and the coefficient of variation are explored, along with their limitations. The authors also provide an online tool for the easy calculation of these statistics in the clinical environment (www.birmingham.ac.uk/echo). Quantifying and enhancing the reproducibility of echocardiography has important potential to improve the value of echocardiography as the basis for good clinical decision-making.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.echo.2019.08.015DOI Listing
December 2019