Clinical Phenotypic Spectrum of 4095 Individuals with Down Syndrome from Text Mining of Electronic Health Records

Abstract

Human genetic disorders, such as Down syndrome, have a wide variety of clinical phenotypic presentations, and characterizing each nuanced phenotype and subtype can be difficult. In this study, we examined the electronic health records of 4095 individuals with Down syndrome at the Children’s Hospital of Philadelphia to create a method to characterize the phenotypic spectrum digitally. We extracted Human Phenotype Ontology (HPO) terms from quality-filtered patient notes using a natural language processing (NLP) approach MetaMap. We catalogued the most common HPO terms related to Down syndrome patients and compared the terms with those from a baseline population. We characterized the top 100 HPO terms by their frequencies at different ages of clinical visits and highlighted selected terms that have time-dependent distributions. We also discovered phenotypic terms that have not been significantly associated with Down syndrome, such as “Proptosis”, “Downslanted palpebral fissures”, and “Microtia”. In summary, our study demonstrated that the clinical phenotypic spectrum of individual with Mendelian diseases can be characterized through NLP-based digital phenotyping on population-scale electronic health records (EHRs).

Phenotype-Driven Molecular Genetic Test Recommendation for Diagnosing Pediatric Rare Disorders

“…The model achieved its highest performance when utilizing features derived from structured billing codes (converted to Phecodes) along with basic demographic information. While various studies 26,53,54 have explored the extraction of clinical phenotypes from clinical narratives for genetic data analysis, our study did not exhibit any noticeable performance improvement when using phenotypes extracted from clinical notes. Additionally, we also explored the use of phenotypes based on another popular ontology, HPO, by extracting HPO terms from the notes, as described in a previous study 29 , and we found no statistical improvement when using either highlevel HPO phenotypic abnormality or fine-grained HPO terms.…”

A Systematic Review of the Application of Computational Technology in Microtia

“…The medical records or the clinical information of patients are also an indispensable resource for data mining. A study exploring the clinical phenotype spectrum of 4095 patients with Down syndrome based on their electronic health records with the help of the MetaMap, a natural language processing tool, revealed that microtia is not significantly related to Down syndrome 51 . Another data mining based on the NSQIP-P database revealed that after reviewing the clinical information of 593 patients, rib grafts were found to be the most common surgical method for auricle reconstruction.…”

“…A study exploring the clinical phenotype spectrum of 4095 patients with Down syndrome based on their electronic health records with the help of the MetaMap, a natural language processing tool, revealed that microtia is not significantly related to Down syndrome. 51 Another data mining based on the NSQIP-P database revealed that after reviewing the clinical information of 593 patients, rib grafts were found to be the most common surgical method for auricle reconstruction. The alloplastic implants have a relatively higher risk of postoperative bleeding complications rate and longer surgical time, but the reliability of these results still needs to be verified by more multicentered data with high reliability.…”

Klarigi: Characteristic Explanations for Semantic Data

“…This is an area of research that is of recent interest, especially with advancements in structured representation of phenotype profiles through phenopackets, and the automated creation of phenotypic profiles through text mining. For example, one study performed a statistical analysis of text-mined phenotype profiles for 4,095 individuals with Down Syndrome [14], descriptively reporting on phenotypes and their frequency of appearance. This analysis, however, only used a measure of patient frequency to stratify phenotypes.…”