Healtex: UK Healthcare Text Analytics Research Network

Lead Research Organisation: University of Manchester
Department Name: Computer Science

Abstract

Healthcare is a prime example of "big data science" with a number of challenges and successful stories where actionable information extracted from data has improved and saved lives [1]. The majority of concerted efforts focused on real-time processing and integration of structured data streams coming from clinical coding, diagnostic tests, sensor measurements, questionnaires, etc. to support timely clinical interventions and facilitate patients' self-management. Nonetheless, natural language remains the main means of communication within healthcare with its written accounts becoming increasingly available in an electronic form, thus giving rise to big text data. Prominent examples include text data embedded within electronic health records (e.g. referral letters, case notes, pathology reports, hospital discharge summaries, etc.), patient-reported outcome measures (e.g. questionnaires, diaries, etc.) or unsolicited informal feedback shared openly on the Web 2.0 (e.g. social media, fora, etc.). Unfortunately, the capacity to effectively utilise information from unstructured text data on a big scale is lagging behind its structured counterpart. The fact that the majority of actionable information in healthcare is contained within text data (some estimates shows as much as 85%) clearly indicates a potential to dramatically transform community health and care by the ability to process and integrate such information in real time. However, automated and large-scale "understanding" of diverse healthcare sublanguages is still largely unsolved research challenge due to their dynamics, idiosyncrasy, ambiguity and variability.

The aim of this proposal is to build a UK-wide multi-disciplinary research network in order to explore the barriers to effectively utilising healthcare narrative text data, road-map research efforts and principles for sharing text data and text analytics methods between academia, NHS and industry. The network will directly address the "Transforming Community Health and Care" grand challenge by enabling research that will deploy healthcare narratives as real-time sensors and integrate them with the structured data streams into a patient-focused collaborative ecosystem, which will involve healthcare professionals, patients, carers and researchers. Such systemic network of healthcare activities will facilitate informed decision making, timely interventions, deeper digital phenotyping for clinical epidemiology and population-based modelling. On the other hand, by processing patient-generated narratives, which are often a preferred and likely means to provide patient responses (e.g. text messages) to complement structured healthcare data (e.g. signals from wearable devices), we will "use real-time information to support self-management of health and wellbeing".

The main outcome of the network will be a strong, sustainable community that will continue its mission after the initial 3 years of support. Other outcomes will include (1) reports describing the state-of-the-art and challenges for key barriers in harnessing text narratives and making sense from them; (2) a research roadmap for healthcare text analytics; (3) an enlarged membership and expanded collaborations within the network, in particular with early career researchers and internationally; (4) a series of focused pilot/feasibility projects that will inform further developments and kick-start collaborative projects; (5) a collection of research papers at conferences and journals, improving the UK competitiveness in this growing area; (6) several project proposals scoped during the project and prepared for submission; (7) proposals for discipline-bridging personal fellowships, and (8) an interactive registry of healthcare text analytics expertise, resources and tools so that the users and collaborators can identify existing resources and initiate new collaboration.

Planned Impact

The network will create societal impact in:

a. Health impact. Research into improving our health relies mostly on structured variables that are collected routing in clinical practice. Clinical narratives convey other valuable diagnostic and contextual information that is predictive of the prognosis and biological behaviour of a disease process, but are rarely used for secondary purposes, e.g. to improve our understanding of patients clinical pathways. Currently 85% of actionable information remains locked within such narratives, i.e. available only for sporadic manual inspection and therefore often overlooked. Using such data to understand and improve the health of our nation is therefore a key area that will benefit from our network. Building on a range of previous studies that have proven the feasibility of NLP for structuring clinical narratives, we will pilot new systems that focus specifically on the types of narratives used within the NHS (e.g. referral letters, pathology and imaging reports, accident reports, etc.) in order to extract accurate data on patient clinical journey and service provision. This will facilitate actionable healthcare analytics, which aims to improve quality and efficiency of care, thus bringing direct health benefits both at the patient level (timely interventions) and to the entire population. For the latter, information extracted from clinical narratives will lend itself to large-scale epidemiology studies describing fine-grained findings. This will help identify previously unknown risk factors, drug interactions, drug's unintended targets, etc. Feeding these research results back into the healthcare service will open another avenue for improving patient health outcomes, public health and quality of life.

b. Impact on healthcare services: With patient groups becoming more adept at lobbying for their consumer "rights" on social media, we will be be able to identify globally unmet patient needs and concerns by mining patient narratives on the web. Information extracted from clinical narratives can be used to ensure that patient pathways are designed to optimise quality, patient outcomes and cost effectiveness. When integrated with other clinical data, narrative can be then used to support decision making for individual patients, and deeper digital phenotyping for clinical epidemiology and population-based modelling. Such findings can impact professional standards within the NHS and inform NICE guidance, which will in turn improve the quality of healthcare, and consequently patient health outcomes, public health and quality of life.

c. Impact on public policy. While healthcare stands to benefit from automated processing of clinical narratives, there are ethical and legal issues associated with their use as they contain private and confidential information. Patient data cannot be used for text analytics purposes without consent. In the EU, Data Protection Directive (95/46/EC) specifies rules on using information about individuals. Unfortunately, even when all reasonable measures to protect privacy and confidentiality are taken (e.g. de-identification and data use agreements), in most cases only researchers with local affiliation are allowed access. This remains the main bottleneck for progress in healthcare applications of text analytics. We will provide research evidence that will stimulate and inform data privacy policy debate with the ultimate aim of adjusting legislation, regulations and guidelines for data sharing that fully protects personal privacy whilst maximising benefits to society.

d. Impact on industry. The outcomes from the network will also provide benefits to pharmaceutical and healthcare businesses, buy making free-text data part of their data space. For example, in the world of drug safety, much of the rich information about the plausibility of a causal association between drug and side effect is held within the narrative, and this can be used by both sectors.
 
Description We are developing a multi-disciplinary community of computer scientists, clinicians, patients, industry and data controllers to support the use of routinely collected free text data. The network has been successful in mobilising the key actors to make the case for use of such data taking into account all the complexity of such processing and all privacy concerns that public might have.
Exploitation Route This will be considered in the final year.
Sectors Agriculture, Food and Drink,Healthcare,Pharmaceuticals and Medical Biotechnology

 
Description The work of Healtex has made impact to industry and NHS by linking the key players that operate in this area to identify opportunities and work on challenges in processing healthcare text. We have organised an industry forum and a number of demo session where industry representatives presented their solution to a wider audience, and where NHS and research organisation where able to point to their needs. Another area of impact was our work on using free-text data for research, via discussion on the Citizens' jury and Text governance projects we have organised, bringing both wider public and specialists views on whether and under what conditions free-text data should be used for research. These findings have been used to influence the policy through discussions with the National Data Guardian Office, NHS England, The Alan Turing Institute and other organisations, including Understanding Patient Data, Wellcome and Use MY Data.
First Year Of Impact 2018
Sector Healthcare
Impact Types Societal,Economic

 
Description Annual Conference - British Society for Rheumatology 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Goran Nenadic from Healtex presented on Text mining from the EHRs & patient generated data. The talk was part of the Annual Conference of the British Society for Rheumatology, and was widely tweeted.
Year(s) Of Engagement Activity 2018
 
Description Artificial Intelligence and Cognitive Healthcare at the Kings Fund 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact This was an event organised by the Kings Fund on Artificial Intelligence and Cognitive Healthcare. Goran Nenadic talk part in a panel that discussed the role of AI (and text mining in particular) for delivering healthcare.
Year(s) Of Engagement Activity 2018
URL https://www.eventbrite.co.uk/e/artificial-intelligence-and-cognitive-healthcare-conference-tickets-3...
 
Description Citizens' Jury on Using Clinical Free Text for Research 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact Healtex, Brighton and Sussex Medical School and Citizens' Juries c.i.c. organised a three-day "citizens' jury" between 6th and 8th June 2018 to understand whether, and under what conditions, the public would accept medical free text data being used for research. Citizens' juries are a recognised methodology for deliberative research with the public, established over 40 years ago by the Jefferson Center in the USA.

This was the first study to ask the public what they think about de-identified clinical narrative being used for secondary purposes. The citizens' jury with 18 people, comprising a cross-section of the public, have learnt from expert witnesses over three days about structured and unstructured health data, and the challenges, needs and opportunities in processing free text data. The jury deliberated together, exploring the complexities and potential trade-offs between privacy and the public good, and reached reasoned conclusions about whether and under what circumstances use of free-text data for medical research can be justified.
Year(s) Of Engagement Activity 2018
URL http://healtex.org/jury/
 
Description Clinical Text De-identification Hackathon 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact A focused 2-day hackathon on clinical text de-identification was held in Manchester, UK on the 17-18th of November. Building on a set of established algorithms such as ChristieDEID and ElasticGazetteer we aimed to extend current requirements, design, and development. This community-driven event worked towards releasing a robust and scalable open source software for clinical text de-identification.

The follow-up activities followed this hachkathon.
Year(s) Of Engagement Activity 2016
URL http://healtex.org/event/text-de-identification-hackathon/
 
Description Data-Intensive Neuroscience Workshop 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact This workshop aimed to showcase the research in applying and developing data-intensive methodologies to support clinical practice and research in the area of neuroscience. A series of talks on different projects, specifically focused on healthcare text analytics, were presented by speakers from the University of Manchester, University of Sydney, Chinese University of Hong Kong, and Salford Royal NHS Trust. In addition to these talks the workshop provided an opportunity for networking and brief flash talks (3 min) from the audience showcasing their research and/or ideas for collaboration.

The event was organised by Healtex and supported through a collaboration grant on 'Text Mining for Integrative Medicine Research in Neurological Disorders' between the University of Manchester and the Chinese University of Hong Kong, and by Healtex.
Year(s) Of Engagement Activity 2016
URL http://healtex.org/event/data-intensive-neuroscience-workshop/
 
Description De-identification working group 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Automated methods for de-identification of medical free-text is a key topic in healthcare text analytics. There are several solutions implemented at local sites, and this working group has been established to bring together expertise and identify the challenges that still need to be addressed.

The group includes researchers from Manchester, Swansea, Sheffield, King's College, UCL, MHRA, Brighton, Sussex.
Year(s) Of Engagement Activity 2016
 
Description Developing data governance standards for working with free-text data in healthcare 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Healtex organised a workshop on "Developing data governance standards for working with free-text data in healthcare" in London on January 16, 2019. Free-text data held in healthcare records represent a vast, untapped source of rich information to guide research and clinical care. Generally, clinical data need to be de-identified before they can be used for secondary purposes such as audit and research, but there are major challenges in finding effective methods that do not damage free-text data utility as a by-product.

TexGov is a Healtex-funded project to work towards the creation of data governance standards to enable free-text data to be used safely for research for public benefit. We run a dynamic and interactive workshop with a combination of presentations, a panel discussion and group activities to take forward the work in this exciting area. It had over 50 participants from a wide range of organisations, including data providers (CRIS, SAIL), patient engagement groups, legal representatives, UK data anonymisation network, etc.
Year(s) Of Engagement Activity 2018
 
Description Governance working group 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Access to healthcare free-text data has been identified as one of the main barriers in this area. We have therefore established the Governance working group (GWG) for medical free text use in research. We have identified two streams of work for the group:

1) Public engagement and attitudes:
2) Developing a governance strategy for moving the free text agenda forward with regulators

We want to identify examples where navigation of governance issues has worked well (e.g. CRIS and CLEF) and examples where governance is a challenge or is currently holding back progress (e.g. CPRD, Christie?). Ideally we would develop a strategy and governance plan (drawing on technical working groups expertise, and the examples which have worked well), for one or two examples as test cases. We would present our governance plan to a regulator to achieve new approvals for free text work, this may then form a framework for other groups or institutions to follow.

The group is interdisciplinary (text miners, legal experts, PPI, patients) and involve universities of Brighton, Sussex, Manchester, Swansea, King's College London, UCL and is open to new members.
Year(s) Of Engagement Activity 2016
 
Description HealTAC - UK Healthcare Text Analytics Conference 2018 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact HealTAC 2018 was the first UK healthcare text analytics conference organised by Healtex. It was a huge success - we had over 100 attendees gathered for a busy 2-day event at the Manchester Conference Centre. The conference featured two excellent keynotes from leading experts in healthcare text analytics (Prof Wendy Chapman, University of Utah and Prof Pierre Zweigenbaum, LIMSI-CNRS), nine research paper presentations, 15 posters, two panels (gaining public trust in healthcare text analytics and mining veterinary clinical records), an industry forum (with key players from the UK and internationally) with seven demo sessions for various software solutions from industry and NHS. We also had a PhD forum where early career researchers presented their projects and received feedback from an expert panel and the audience. The forum was followed by an excellent career talk by Prof Wendy Chapman.

HealTAC 2018 brought the academic, clinical, industrial and patient communities together for the first time in the UK to discuss the current state of the art in processing healthcare free text and share experience, results and challenges. The conference's main sponsor was EPSRC via the Healtex network (EPSRC's NetworkPlus programme for Healthcare Technologies Grand Challenges). It is also supported by industry: SciBite, Linguamatics, DeepCognito, IMO Intelligent Medical Objects and Health Unlocked, and the Connected Health Cities.

HealTAC will become an annual community event.
Year(s) Of Engagement Activity 2018
URL http://healtex.org/healtac-2018/
 
Description Healtac Veterinary text-mining panel 2018 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact An expert panel convened at the Healtac Conference in April 2018
Year(s) Of Engagement Activity 2018
 
Description Healtex Impact video 
Form Of Engagement Activity A broadcast e.g. TV/radio/film/podcast (other than news/press)
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact This is a Healtex promotional video, aimed at general public. It presents aims and objectives of the network and discusses the value of using free-text data in healthcare research.
Year(s) Of Engagement Activity 2018
URL https://www.youtube.com/watch?v=NhBJSZ8yxAo
 
Description Healtex Launch Event 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact This was the official launch of the network that gathered over 70 people for a two day event. It provided an opportunity for establishing links and networking.

Agenda (Monday, November 14th):

12:30-2:00 Registration and Lunch
2:00-2:15 Goran Nenadic: "The UK healthcare text analytics network"
2:15-2:45 Iain Buchan: "Mining text for health discovery and actionable analytics"
2:45-3:15 Robert Stewart: "Unmet needs and immediate challenges in healthcare text processing - the CRIS experience"

3:15-4:00 Break, posters, networking

4:00-4:30 Nigel Collier: "NLP capabilities and challenges in the health arena"
4:30-5:00 Elizabeth Ford: "Issues of Privacy and Governance when Using Medical Free Text for Research"

5:00-6:30 Posters, networking, wine/nibbles reception
Year(s) Of Engagement Activity 2016
URL http://healtex.org/event/healtex-launch/
 
Description Healtex datathon on ADR identification from Social Media (Sept 2018) 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Healtex and our partner HealthUnlocked co-organised a datathon on adverse drug reaction mining from social media post, which took place in London on September 28th, 2018. A team of 17 participants from the universities of Manchester, Sheffield, Brighton and Sussex along with the hosts from HealthUnlocked spent the full day analysing and discussing the current state-of-the-art and challenges in the identification of key elements of automated pharmacovigilance using social media.

The participants, including clinicians, text and data miners, qualitative researchers and user engagement experts, worked with an annotated dataset of 200 posts from five communities present in HealthUnlocked to establish the feasibility of automated extraction of the core Yellow Card data (e.g. treatment indication, drug, side effect, severity, outcome etc.). Yellow Card is a scheme used by the Medicines and Healthcare products Regulatory Agency (MHRA) to monitor the safety of all healthcare products in the UK. After an exciting and exhausting day of coding and discussions, we were able to identify a number of challenges that would require additional work to make the results of text mining of patient generated data applicable in real-world applications. The participants also looked at a wider context of how to support better and more efficient reporting of ADRs to regulatory bodies.
Year(s) Of Engagement Activity 2018
URL http://healtex.org/news/
 
Description Healtex emailing list 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact JiSC-hosted emailing list with 200+ members. The list is used to communicate all relevant information to the community.
Year(s) Of Engagement Activity 2016
 
Description Healtex leaflet 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact This is a leaflet that describes the main objectives and aims of the Healtext network. It has been distributed to many conferences and events.
Year(s) Of Engagement Activity 2016
 
Description Healtex newsletter - November 2018 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact A new Healtex newsletter features the following stories:

- HealTAC 2019 - keynote speakers and call for contributions
- Call for dissemination and outreach events (14th January 2019)
- New Healtex feasibility studies
- Past events
- Healtex early career research network - supporting future leaders
Year(s) Of Engagement Activity 2018
URL http://healtex.org/wp-content/uploads/2018/11/newsletter-November-2018.pdf
 
Description Healtex newsletter April 2018 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact This was the first Healtex newsletter, which features the following stories:

- Second call for feasibility studies (deadline: 1st June 2018)
- Healtex Governance Working Group - call for governance case studies
- Healtex early career research network - supporting future leaders
- Sharing clinical text analytics methods and algorithms: a GATE-based hackathon (May 16-17, 2018)
- Datathon on Identification of Adverse Drug Reactions from Social Media (July 11-12, 2018 - tbc)
Year(s) Of Engagement Activity 2018
URL https://www.dropbox.com/s/vgra2tqarxt2g1n/Healtex-newsletter-April-2018.pdf?dl=0
 
Description Healtex twitter feed 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact Healtex Twitter feed, used to announce events, papers, distribute calls and information.
Year(s) Of Engagement Activity 2016
URL https://twitter.com/UK_healtex
 
Description Healtex web site 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact The web site is the hub of the network that presents information about the activities, news and calls relevant to the community. We have also introduced information on relevant conferences, jobs, scholarships.
Year(s) Of Engagement Activity 2016
URL http://healtex.org/
 
Description LOUHI 2018: The Ninth International Workshop on Health Text Mining and Information Analysis 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Goran Nenadic gave a keynote presentation " Distributed text mining in healthcare: linking data, methods and people", which introduced Healtex to the international community, as well as the main challenges and opportunities for clinical NLP.
Year(s) Of Engagement Activity 2018
URL https://louhi2018.fbk.eu/program
 
Description Leaflet: What access should researchers have to free-text data in health records? 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact This is a leaflet that summarises finding of the Citizens' jury on 'What access should researchers have to free-text data in health records?'
Year(s) Of Engagement Activity 2018
 
Description Medical Confidentiality: When is it OK to use your Patient Records in Research 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Patients, carers and/or patient groups
Results and Impact Patients have contributed to research in the NHS since it began, often taking part in trials of new drugs or procedures. But increasingly medical researchers are seeing the value of using patient data for research. Patient data is the information collected and recorded by the NHS, each time someone comes in to see the doctor.
This was a discussion event hosted by the Brighton and Sussex Medical School and the UK Healthcare Text Analytics Network. The event aimed to discuss the following questions with the members of public: Do you know what's in your medical record? Have you ever thought how information about your care and outcomes might help others? Where should we draw the lines between privacy and making use of data for better care?

Some data is harder to make anonymous such as letters written between GPs and specialists, scan reports, and doctor's notes about patients' symptoms, and so is often not included in a data set used by researchers.

We wanted to know what the public thinks about medical data that is harder to make anonymous being used for research.

The event involved learning more about research using NHS patient data through presentations from medical researchers, doctors and law specialists, as well as roundtable discussions.
Year(s) Of Engagement Activity 2018
URL http://healtex.org/event/medical-confidentiality-ok-use-patient-records-research/
 
Description Mental Health NLP working group 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Free text data is a key source of information in mental health clinical practice. In the UK we have several databases of routinely collected clinical narrative from mental health trusts that are anonymised and can be used to support research (e.g. CRIS, UK CRIS, etc.). There has been work in different groups to build NLP solutions to extract information from such data and this working group was established to bring together these groups to discuss the opportunities and challenges, and to come up with a research road map. The group was formed from King's College London, University of Oxford, University of Manchester, Nottingham Uni, Nottingham Trent University, University of Warwick and EPFL (Switzerland).
Year(s) Of Engagement Activity 2018
URL http://healtex.org/mental-health-nlp-working-group/
 
Description Mental Health NLP working group road-map meeting 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact A research road-mapping meeting was organised at King's College London on January 15th, with participants from KCL, Nottingham, Warwick, Manchester, Nottingham Trent and Oxford. Each group presented their current activities and a discussion was held on where the group can make an impact. This will be followed up with a document with open research questions.
Year(s) Of Engagement Activity 2019
 
Description Mental health NLP emailing list 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact A specialist focused emailing list to support the work of the Mental Health Working group.
Year(s) Of Engagement Activity 2018
 
Description Mining Radiology Reports - Plenary Meeting (Edinburgh) 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact A plenary meeting of the Working group was organised in Edinburgh on March 4th, with participants from the University of Edinburgh, Cancer Research, NHS Lothian, University of Manchester, Salford NHS FT. The group presented the current state of the art and discussed where the challenges are. This will be followed up with a report.
Year(s) Of Engagement Activity 2019
 
Description Neurodegenerative diseases research roadmap 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Healtex has presented its objectives and opportunities to a multi-disciplinary audience of neurodegenerative disease researchers. This has generated interests and potential collaborations with the NeuroD Platform technology network (EPSRC funded).
Year(s) Of Engagement Activity 2017
 
Description Pharmacovigilance using social media text mining - Manchester Summer School in Digital Epidemiology 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Healtex participated in a summer school organised by our partner the Arthritis Research UK Centre for Epidemiology. The school was on Digital Epidemiology, aiming to explore and understand the opportunities, challenges and methods for capturing and using digital data to support high-quality epidemiological research. The Healtex team presented on 'Pharmacovigilance using social media text mining'.
Year(s) Of Engagement Activity 2018
URL http://documents.manchester.ac.uk/display.aspx?DocID=36944
 
Description Presentation "Evidence-based medicine using healthcare free-text data" 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Third sector organisations
Results and Impact This invitated talk was part of a seminar at Analytic and Clinical Cooperative Laboratory of Integrative Medicine (ACCLAIM), a Joint laboratory between Hong Kong Institute of Integrative Medicine (HKIIM), The Chinese University of Hong Kong, and School of Information Technologies, The University of Sydney
Summary: "Evidence-based medicine using clinical free-text data"
Recent developments in the availability of Electronic Medical Records (EMRs) have triggered a number of new opportunities for more efficient epidemiological research and clinical decision support. EMRs contain information that - when available on a larger scale - can provide a unique opportunity for clinical investigations, decision support, meta-analysis and observational research. In many disciplines, free-text narrative is a key source of information but its unstructured nature is a key barrier for use in evidence-based medicine.
In this talk we will review recent developments in processing of healthcare narrative, including automated harvesting of important clinical concepts and events, clinical coding and improving the accuracy and scope of other clinical data. We will also outline main challenges.
Year(s) Of Engagement Activity 2017
 
Description Radiology Report NLP working group 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Free text radiology reports are a key source of information for many clinical applications. In the UK we have several local databases of routinely collected radiology reports but these are not often used to support research. There has been work in different groups to build NLP solutions to extract information from such data and this working group was established to bring together these groups to discuss the opportunities and challenges, and to come up with a research road map. The group was formed from University of Edinburgh, Cardiff University, The Alan Turing Institute, Leeds University.
Year(s) Of Engagement Activity 2018
 
Description Radiology reports emailing list 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Emailing list to support the work of the Radiology Reports Working group.
Year(s) Of Engagement Activity 2018
 
Description SAVSNET 10th birthday 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact To mark ten years of The Small Animal Veterinary Surveillance Network (SAVSNET, which is a Healtex partner), an event was organised to showcase the research being done using electronic health data from veterinary practices and laboratories, and how it is informing primary veterinary practice. A lot of data that SAVSNET collects is in free text - that's why text analytics is an integral part of their processing workflow. On behalf of Healtex, Goran Nenadic presented the current state of the art and challenges in free-text data processing. The event had around 40 delegates including from the USA and Portugal.
Year(s) Of Engagement Activity 2018
URL https://www.liverpool.ac.uk/savsnet/happybirthday/
 
Description SLTC 2018 - Seventh Swedish Language Technology Conference, Stockholm, 7-9 November 2018 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Goran Nenadic took a part in a panel "Clinical text mining - Challenges and future visions" with Aurélie Névéol (LIMSI, France), Lilja Øvrelid (University of Oslo) and Hercules Dalianis (Stockholm University). Healtex was introduced.
Year(s) Of Engagement Activity 2018
URL http://sltc2018.su.se/program/
 
Description Sharing clinical text analytics methods and algorithms: a GATE-based hackathon 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Healtex is focused on sharing clinical text analytics methods and algorithms. To support this aim, we hosted an evening and a day in Sheffield, UK on 16-17 May 2018, where participants worked on adaptation and wrapping clinical text analytics software for sharing with GATE, the widely used open source natural language processing architecture.

A small number of short presentations introduced the GATE API; wrapping software for GATE; new document formats for GATE; considerations for multi-threading, cloud, and web applications with GATE. These were combined with hackathon sessions where participants worked under the guidance of GATE team members to adapt and wrap their own software.

The hackathon was for those with non-GATE based software that they wish to adapt to GATE, and those with software already adapted to GATE for which they wish to extend or improve adaptation.
Year(s) Of Engagement Activity 2018
URL http://healtex.org/event/sharing-clinical-text-analytics-methods-algorithms-gate-based-hackathon/
 
Description The Yellow Card datathon with Healtex and HealthUnlocked 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact A video interview was conducting after the Yellow Card datathon with Healtex and HealthUnlocked, with Prof William Dixon (University of Manchester, project PI), Matt Jameson Evans (Chief Medical Officer, HealthUnlocked.com) and Prof Goran Nenadic (Healtex). The interview explored how machine learning can help us understand and improve the experiences of patients on the HealthUnlocked platform.
Year(s) Of Engagement Activity 2018
URL https://youtu.be/jIyjYKwzao4
 
Description Turing Health Programme 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Healtex successfully presented a case to establish a theme on 'Text mining' at the Turing Health Programme conference in Manchester on March 7th and 8th 2019. This theme will work together with Healtex, the Alan Turing Institute and HDR-UK to coordinate activities in this area.
Year(s) Of Engagement Activity 2019
 
Description Tutorial: "Healthcare Text Analytics - Analysing Free-Text Health Data" 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact A group of 20 clinicians, NLP researchers and epidemiologists attended the tutorial, organised as a pre-conference event of the Informatics for Health 2017 conference in Manchester. The tutorial explained the main steps in text mining and discussed the challenges and opportunities for healthcare text analytics.
This was a full-day event that looked at several case studies such as automated coding to a standard vocabulary (e.g. SNOMED CT or ICD-10) or identification of adverse events as reported in social media. The topics included: Introduction to basics concepts and steps in text analytics; Demonstration of 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 of existing platforms for clinical text analytics; Discussion of unmet needs and opportunities.
The audience was extremely multi-disciplinary, including healthcare data scientists, health informaticians, IT and EPR specialists, epidemiologists, NLP researchers and clinicians.
Year(s) Of Engagement Activity 2017
URL http://healtex.org/event/tutorial-healthcare-text-analytics-analysing-free-text-health-data/
 
Description Unmet needs and challenges in using clinical narrative 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Unmet needs and challenges in using narrative data for clinical/epidemiological research

This workshop provided an opportunity to consider text fields across clinical specialties and platforms, focusing on the potential opportunities to open up new data for analysis and to support novel healthcare e-interventions. Particular priority was given to the views of clinical researchers who are most familiar with the ways in which information is recorded in their services. There are likely to be important applications which may not always generalise between specialties and which may not have been adequately considered to date because of the tendency for health records research to focus on the data which are most readily available/accessible.

The workshop is part of the activities organised by the network to identify unmet needs and research challenges at an early stage in order to direct its development work most effectively.

Agenda (Tuesday, November 15th)

09:30-11:00 Making sense from clinical narrative: stakeholder and clinical needs

Panelists:

Dr Anoop Shah (UCL and Farr Institute)
Prof Rob Stewart (KCL)
Dr Kate Button (Cardiff and Vale University Health Board)
Prof Jackie Cassell (Brighton and Sussex Medical School)
Divya Srivastava, NHS England
Dr Adrian Parry-Jones, Salford Royal FT

11:00-11:30 Break and networking

11:30-13:00 Challenges and barriers in processing clinical narrative

Panelists:

Prof Richard Dobson, KCL
Dr Jeremy Rogers, NHS digital
Clive Stringer, KCL Systems Delivery Manger
Dr Angus Roberts, University of Sheffield
Dr Claire Grover, University of Edinburgh
Tom Liptrot, The Christie NHS FT

13:00 - 14:30 Lunch

14:30 Close
Year(s) Of Engagement Activity 2016
URL http://healtex.org/event/workshop-clinical-narrative-and-text-mining/
 
Description Veterinary Text Mining Workshop 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact This workshop had two aims:
1) present the state of the art in processing free-text in veterinary electronic health records, and
2) establish a special interest group that will work together on improving the capabilities for making sense of veterinary free-text.

The event followed the two veterinary text mining bootcamp events. The workshop was held in Nottingham and was co-organised by the Centre for Evidence-based Veterinary Medicine @ Nottingham, SAVSNET and Healtex.
Year(s) Of Engagement Activity 2016
URL http://healtex.org/event/veterinary-text-mining-workshop/
 
Description Veterinary text-mining bootcamp 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact This was a hands on informal event aimed at increasing the familiarity with the potential of text mining in their datasets, and equipping a cross-section of researchers with the fundamental skills to extract information from narrative data. The participants have been given tasks to develop clinical text analytics methods to extract specific information from veterinary records. By the end of the day most participants had got a practical skill to extract information from narrative.

Venue/date: Leahurst on the 15th September 2016.
Year(s) Of Engagement Activity 2016
URL http://healtex.org/event/second-veterinary-text-mining-bootcamp/
 
Description Workshop: "Extracting evidence from clinical free text: opportunities and challenges" 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact We collocated this workshop with the Informatics for Health 2017 conference and asked a couple of international colleagues (Germany, Sweden) to present their experience in building the capacity to effectively utilise free text to extract evidence to support clinical practice and epidemiological research. The workshop brought together clinicians and researchers who then discussed key opportunities and challenges in clinical text analytics, focusing on technical, ethical and legal barriers and unmet needs. The event started with three talks presenting the state of the art in information extraction from clinical narrative, mental healthcare, automatic surveillance of healthcare-associated infections and diagnosis code assignment. This was followed by four break-out sessions that were coordinated by Healtex stream-leads to discuss the main research challenges (2 groups), clinical needs (1 group) and governance issues (1 group). Finally, a plenary session was convened to sum-up the main discussion points.
Year(s) Of Engagement Activity 2017
URL http://healtex.org/event/workshop-extracting-evidence-clinical-free-text-opportunities-challenges/
 
Description Workshop: "Using patient-reported data for research and to improve health outcomes and services" 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Patients, carers and/or patient groups
Results and Impact This event was co-organised with the Farr-Institute of Health Informatics Research. The main aim was to identify the key health informatics opportunities and challenges with regard to capturing and using patient-reported data for research and to improve health outcomes and services. One of the major sources of patient-generated data are social media networks, which have created an unprecedented opportunity for acquiring health data directly from patients. Healthcare text analytics can contribute to unlocking this potential, but only if we understand the main opportunities and challenges of patient-reported data. The event started with four talks presenting case studies on collecting and using patient-report data, also discussing governance issues. This was followed by a break-out session where participants (many of them members of the Farr's Patient and Public Involvement (PPI) group) discussed their views on using and collecting patient and carer reported data. Finally, a plenary session was convened to sum-up the main discussion points.
Year(s) Of Engagement Activity 2017
URL http://healtex.org/event/informatics-health-2017/