Tutorial: “Healthcare Text Analytics – Analysing Free-Text Health Data”

Overview
Vast amounts of healthcare data is in free text (clinical notes, letters, prescription extraction, healthcare social media): harnessing that data space has potential to change, streamline and personalise service delivery. Typical examples include automated coding to a standard vocabulary (e.g. SNOMED CT or ICD-10) or identification of adverse events as reported in social media. However, free text brings a number of challenges in identification of key clinical variables, including intense terminological variability and ambiguity. Healthcare text analytics provides technologies for automated, large-scale extraction of information from healthcare free-text.

Aims & Objectives
The main aim of this tutorial is to explain and discuss challenges and opportunities for healthcare text analytics. Specifically, we will:

• Introduce the basics concepts and steps in text analytics
• Demonstrate example applications, such as identification of symptoms and signs, automated coding, medication extraction, extraction of social factors (smoking, alcohol), adverse events and outcomes of treatments, effects on quality of life etc.
• Review existing platforms for clinical text analytics
• Discuss unmet needs and opportunities

Intended Audience
This tutorial will aim to introduce text analytics to a multi-disciplinary audience, including healthcare data scientists, health informaticians, IT and EPR specialists, epidemiologists and clinicians.

Instructor Biography
Prof Goran Nenadic is a Professor at the University of Manchester and The Farr Institute’s Health eResearch Centre (HeRC). His research focus is on large-scale extraction and linking of clinical/epidemiological findings from electronic health records, healthcare social media and biomedical literature. He leads the UK healthcare text analytics network (http://www.healtex.org).