NLP for real-time data capture in EHR to improve clinical care and operational efficiency

The context

Electronic health records (EHR) and the data stored in them can play an important role in improving the quality of patient care. Much of the information in EHR is recorded as “free text”, that is ordinary language without any restriction on format, since this is the natural way in which people communicate. Free text often contains words, abbreviations and sentences that may be interpreted in more than one way – for example, “HR” can mean “Hour” or “Heart Rate”, or there might be a sentence where a diagnosis is mentioned but to the effect that is was ruled out (that is, a negation). As a consequence, although computers can be used to interpret free text, they cannot always get it right, and clinicians will always have to check the results to ensure patient safety. Expressing information in a structured way can avoid this problem, but has a big disadvantage - it can be time-consuming for clinicians to enter it. This can mean that information is incomplete, or clinicians are so busy entering notes on the computer that they do not have time to listen to their patients.

Cutting-edge research

Funded by the National Institute for Health and Care Research (NIHR) and led by Drs Anoop Shah (UCL IHI and UCLH) and Wai Keong Wong (UCLH), MiADE aims to develop a system to support automatic conversion of the clinician’s free text notes into a structured format. The system will allow the clinician to record diagnoses, medications and allergies in a structured way, using NHS-endorsed clinical data standards. The clinician can then check the structured data immediately, before making it a formal part of a patient’s record.

The system will use Natural Language Processing (NLP), a technique typically applied by research teams to extract information from existing EHRs, but rarely used to improve the way information is entered in the first place. The aspiration is for the NLP system to continually learn and improve as more text is analysed and checked by clinicians.

The researchers will first test the system in a simulation environment, and then will instal it for testing at University College London Hospitals, where a new EHR system called Epic is in place. They will study how effective the system is, and how clinicians and patients find it when it is used in consultations. Based on feedback, improvements will be implemented and the system will be installed for further testing at a second site (Great Ormond Street Hospital).

For more information, please visit the project's webpage.