Natural language processing to extract symptoms of severe mental illness from clinical text: the Clinical Record Interactive Search Comprehensive Data Extraction (CRIS-CODE) project

Abstract

ObjectivesWe sought to use natural language processing to develop a suite of language models to capture key symptoms of severe mental illness (SMI) from clinical text, to facilitate the secondary use of mental healthcare data in research.DesignDevelopment and validation of information extraction applications for ascertaining symptoms of SMI in routine mental health records using the Clinical Record Interactive Search (CRIS) data resource; description of their distribution in a corpus of discharge summaries.SettingElectronic records from a large mental healthcare provider serving a geographic catchment of 1.2 million residents in four boroughs of south London, UK.ParticipantsThe distribution of derived symptoms was described in 23 128 discharge summaries from 7962 patients who had received an SMI diagnosis, and 13 496 discharge summaries from 7575 patients who had received a non-SMI diagnosis.Outcome measuresFifty SMI symptoms were identified by a team of psychiatrists for extraction based on salience and linguistic consistency in records, broadly categorised under positive, negative, disorganisation, manic and catatonic subgroups. Text models for each symptom were generated using the TextHunter tool and the CRIS database.ResultsWe extracted data for 46 symptoms with a median F1 score of 0.88. Four symptom models performed poorly and were excluded. From the corpus of discharge summaries, it was possible to extract symptomatology in 87% of patients with SMI and 60% of patients with non-SMI diagnosis.ConclusionsThis work demonstrates the possibility of automatically extracting a broad range of SMI symptoms from English text discharge summaries for patients with an SMI diagnosis. Descriptive data also indicated that most symptoms cut across diagnoses, rather than being restricted to particular groups.

Automatic mining of symptom severity from psychiatric evaluation notes

“…Recent developments in the use and availability of electronic medical records (EMRs) have triggered a number of opportunities for more efficient clinical decision support and epidemiological research. EMRs contain information such as the patient's clinical history, treatments, and laboratory results (Abbe, Grouin, Zweigenbaum, & Falissard, ; Ford, Carroll, Smith, Scott, & Cassell, ; Thomas et al, ) and—when available on a larger scale—can provide a unique opportunity for clinical investigations, decision support, meta‐analysis, and observational research (Jackson et al, ; Kovalchuk, Stewart, Broadbent, Hubbard, & Dobson, ; Perlis et al, ). Specifically, psychiatric EMRs contain rich knowledge regarding the mental health status of patients and important contextual information that is often in free text.…”

“…Cunningham, Tablan, Roberts, and Bontcheva () utilized a rule‐based approach for the extraction of mini mental state examination results from both short clinical notes and free text health record correspondence between clinicians with an overall precision ranging from 85% (in short notes) to 87% (in correspondence texts). Jackson et al () sought to capture a number of key symptoms of severe mental illness from clinical discharge summaries with a median F‐sore of 88% using regular expression pattern matching.…”

Predicting mental conditions based on “history of present illness” in psychiatric notes with deep neural networks

“…Rumshisky et al [38] predict early readmits (within 30 days) to inpatient psychiatric units through topic models built with discharge summaries. Jackson et al [17] identify over 40 key symptoms (e.g., aggression, apathy, irritability, and stupor) of severe mental illness based on discharge summaries from nearly 8000 patients visiting a UK based mental healthcare provider using SVM models. Perlis et al [32] provide results of one of the first text mining applications of psychiatric notes where they apply logistic regression models (with LASSO regularization) and show that combining information from unstructured notes with coded information results in major gains in predicting patient mood state when compared with using coded information alone.…”

SemEHR: A General-purpose Semantic Search System to Surface Semantic Data from Clinical Notes for Tailored Care, Trial Recruitment and Clinical Research

“…Search system (CRIS) [8], an anonymised replica of the EHR used in the South London and Maudsley NHS Foundation Trust (SLaM) in the UK, was designed for supporting clinical and scientific studies. Since its launch in 2009, a large number of studies ( [9][10][11][12][13] to name a few) used the CRIS resource in conjunction with NLP or text mining techniques. Although these studies answered different clinical questions, the technical requirements for extracting, structuring and making sense of the data largely overlapped, and included: a) corpus-related document pre-processing and cleansing (e.g., removing misleading form headings from scanned documents); b) common medical terminology compilation and recognition (e.g., the antipsychotic medication identification problem is almost the same in [10] and [11]); and c) deriving patient-level clinical signals from document-level NLP annotations (e.g.…”