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

Abstract

Objective: Unlocking the data contained within both structured and unstructured components of Electronic Health Records (EHRs) has the potential to provide a step change in data available forsecondary research use, generation of actionable medical insights, hospital management and trial recruitment. To achieve this, we implemented SemEHR -a semantic search and analytics, open source tool for EHRs.Methods: SemEHR implements a generic information extraction (IE) and retrieval infrastructure by identifying * The manuscript has been submitted to JAMIA

Natural Language Processing for Mimicking Clinical Trial Recruitment in Critical Care: A Semi-automated Simulation Based on the LeoPARDS Trial

“…Text analytics platforms such as semEHR (built on CogStack) [13, 14] and GATE [33] are increasingly being used across large document repositories, and can incorporate a range of NLP methods such as Bio-Yodie [15] (rules-based information extraction, used in this project) and machine learning metehods. UCLH is proposing to make semEHR a core component of its new clinical research data warehouse.…”

“…Stuctured and free text data from the EHR were combined into a searchable indexed repository using the CogStack [13] platform, which contains pipelines for document processing and indexing, fast text searching, and distributed analysis. We used the SemEHR [14] biomedical document processing system on CogStack, with Elasticsearch 1 for full free text search to explore text and annotations and Bio-Yodie [15] (an NLP application) to annotate text using the Unified Medical Language System (UMLS) [16]. SemEHR contextualises each mention of a UMLS concept with the experiencer (patient or other), affirmation status (affirmed, negative or hypothetical) and temporality (past or recent).…”

Risk Factors Extraction from Clinical Texts based on Linked Open Data

“…A recent review of clinical IE applications (Wang et al, 2018) notes the increasing interest to NLP but lists only 25 IE systems which were used multiple times, outside the labs where they were created. Isolated attempts exist to apply IE in the context of EHR processing in frameworks for semantic search, for instance SemEHR deployed to identify contextualized mentions of biomedical concepts within EHRs in a number of UK hospitals (Wu et al, 2018). We mention the following research prototypes as experimental developments, based on some sort of IE: (Shi et al, 2017) reports about a system extracting textual medical knowledge from heterogeneous sources in order to integrate it into knowledge graphs; (Hassanpour and Langlotz, 2016) describes a machine learning system that annotates radiology reports and extracts concepts according to a model covering most clinically significant contents in radiology; presents the information extraction and retrieval architecture CogStack, deployed in the King's College Hospital.…”

Natural language processing algorithms for mapping clinical text fragments onto ontology concepts: a systematic review and recommendations for future studies

“…In the medical domain, SNOMED CT [ 7 ] and the Human Phenotype Ontology (HPO) [ 8 ] are examples of widely used ontologies to annotate clinical data. After the data has been annotated, it can be reused by clinicians to query EHRs [ 9 , 10 ], to classify patients into different risk groups [ 11 , 12 ], to detect a patient’s eligibility for clinical trials [ 13 ], and for clinical research [ 14 ].…”