Towards similarity-based differential diagnostics for common diseases

Research output: Contribution to journalArticlepeer-review

Standard

Towards similarity-based differential diagnostics for common diseases. / Slater, Luke T.; Karwath, Andreas; Williams, John A.; Russell, Sophie; Makepeace, Silver; Carberry, Alexander; Hoehndorf, Robert; Gkoutos, Georgios V.

In: Computers in Biology and Medicine, Vol. 133, 104360, 06.2021.

Research output: Contribution to journalArticlepeer-review

Harvard

APA

Vancouver

Author

Slater, Luke T. ; Karwath, Andreas ; Williams, John A. ; Russell, Sophie ; Makepeace, Silver ; Carberry, Alexander ; Hoehndorf, Robert ; Gkoutos, Georgios V. / Towards similarity-based differential diagnostics for common diseases. In: Computers in Biology and Medicine. 2021 ; Vol. 133.

Bibtex

@article{5384390ebe1749eb8ef167034fafb56d,
title = "Towards similarity-based differential diagnostics for common diseases",
abstract = "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.",
keywords = "Differential diagnosis, Mimic-iii, Ontology, Semantic similarity, Semantic web",
author = "Slater, {Luke T.} and Andreas Karwath and Williams, {John A.} and Sophie Russell and Silver Makepeace and Alexander Carberry and Robert Hoehndorf and Gkoutos, {Georgios V.}",
note = "Publisher Copyright: {\textcopyright} 2021 The Author(s)",
year = "2021",
month = jun,
doi = "10.1016/j.compbiomed.2021.104360",
language = "English",
volume = "133",
journal = "Computers in biology and medicine",
issn = "0010-4825",
publisher = "Elsevier",

}

RIS

TY - JOUR

T1 - Towards similarity-based differential diagnostics for common diseases

AU - Slater, Luke T.

AU - Karwath, Andreas

AU - Williams, John A.

AU - Russell, Sophie

AU - Makepeace, Silver

AU - Carberry, Alexander

AU - Hoehndorf, Robert

AU - Gkoutos, Georgios V.

N1 - Publisher Copyright: © 2021 The Author(s)

PY - 2021/6

Y1 - 2021/6

N2 - 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.

AB - 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.

KW - Differential diagnosis

KW - Mimic-iii

KW - Ontology

KW - Semantic similarity

KW - Semantic web

UR - http://www.scopus.com/inward/record.url?scp=85103691545&partnerID=8YFLogxK

U2 - 10.1016/j.compbiomed.2021.104360

DO - 10.1016/j.compbiomed.2021.104360

M3 - Article

VL - 133

JO - Computers in biology and medicine

JF - Computers in biology and medicine

SN - 0010-4825

M1 - 104360

ER -