Using Electronic Patient Records to Discover Disease Correlations and Stratify Patient Cohorts

By Francisco S. Roque, Peter Bjødstrup Jensen, H. Schmock, M. Dalgaard, Massimo Andreatta, T. Hansen, K. Søeby, S. Bredkjær, A. Juul, Thomas Werge and 2 more. 2011

309 citations recorded by the source corpus.

Electronic patient records remain a rather unexplored, but potentially rich data source for discovering correlations between diseases. We describe a general approach for gathering phenotypic descriptions of patients from medical records in a systematic and non-cohort dependent manner. By extracting phenotype information from the free-text in such records we demonstrate that we can extend the information contained in the structured record data, and use it for producing fine-grained patient stratification and disease co-occurrence statistics. The approach uses a dictionary based on the International Classification of Disease ontology and is therefore in principle language independent. As a use case we show how records from a Danish psychiatric hospital lead to the identification of disease correlations, which subsequently can be mapped to systems biology frameworks.

Subfields: Artificial Intelligence, Health Information Management, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Health Information Management

In Molecular Biology

Citations in the corpus

Natural Language Processing for EHR-Based Computational Phenotyping (2018, background): “[114] NA Simple sentence splitter split the text into smaller units ICD-10, narrative ICD-10, small units of sentences Sauer et al.”
Computational Phenotype Discovery Using Unsupervised Feature Learning over Noisy, Sparse, and Irregular Clinical Data (2013, neutral): “…for common patterns in the distribution of related events [10], converting measurements to an interval-based temporal abstraction [11] and then using frequent itemset mining to find common patterns [12,13], or ignoring time all together and using vector-space methods for feature learning [14].”
Clinical Natural Language Processing in languages other than English: opportunities and challenges (2018, background): “extending information contained in structured data fields with information from Danish clinical free-text with dictionary-based approaches for the study of disease correlations [86] or adverse events [87].”
Predictive modeling of structured electronic health records for adverse drug event detection (2015, background): “Although the main function of EHRs is to archive and manage patient data efficiently - in comparison to paper-based health record systems - secondary use of EHR data is currently being widely explored for various medical research, such as disease discovery and patient stratification [12,13], among w”
DISEASES: Text mining and data integration of disease–gene associations (2014, background): “useful for text mining of clinical narrative in EHRs, especially because it has been translated to many languages [11].”
Mining electronic health records: towards better research applications and clinical care (2012, methodology): “An approach to investigate the underlying molecular aetiology of disease correlations that have been uncovered using EHR data is to map the diseases to known associated genes and proteins, and to investigate the resulting protein–protein interaction network for statistical overlaps 40 .”
Towards Personalized Medicine: Leveraging Patient Similarity and Drug Similarity Analytics (2014, methodology): “With the right patient similarity in place, patient similarity analytics have been used in the target patient retrieval 14 , medical prognosis 15, 16 , risk stratification 17, 18 , and clinical pathway analysis 19 tasks.”
Mining Electronic Health Records: A Survey (2017, methodology): “…et al. 2010; Verduijn et al. 2007; Batal et al. 2012; Choi et al. 2016; Perotte and Hripcsak 2013] Cohort Identiﬁcation Descritive Atemporal [Roque et al. 2011; Bauer-Mehren et al. 2013] Descritive Temporal [Gotz et al. 2014; Schulam et al. 2015; Lasko et al. 2013; Wang et al. 2015a]…”
Dose-Specific Adverse Drug Reaction Identification in Electronic Patient Records: Temporal Data Mining in an Inpatient Psychiatric Population (2014, methodology): “It has previously been shown that investigation of clinical narratives can give an up to fourfold boost in the number of unique diagnoses associated with each patient [20], when comparing to assigned International Classification of Diseases (ICD) 10 codes.”
Natural Language Processing of Clinical Notes on Chronic Diseases: Systematic Review (2019, methodology): “In the same domain of mental and behavioral disorders, comorbidity networks were derived from the patient notes at the largest Danish psychiatric hospital in order to extract disease correlations [102].”