ACCUMULATE - ACquiring CrUcial Medical information Using LAnguage TEchnology

From 01-01-2016 to 30-06-2020

Description

General goal:

ACCUMULATE will automatically recognize crucial information in the free text of clinical reports1

written in English and Dutch by designing, developing and evaluating advanced language technology (LT)

for deep semantic processing of the texts which are often morpho-syntactically not well-formed. An extra

focus is on easy portability of the technology across domains and languages.

ACquiring CrUcial Medical information Using LAnguage Technology (ACCUMULATE) is an important

and necessary step in medical knowledge and healthcare management sectors and has specific applications

in the drug development cycle and patient recruitment. A large amount of valuable medical knowledge

is now buried in free text and we absolutely need advanced natural language processing tools to mine this

knowledge. Studies have demonstrated the large economic value of text analytics, especially in medical

sciences and healthcare management (e.g., McKinsey Report 2011, JISC Report 2012). These studies

pinpoint the need for research to further deepen the semantic extractions using advanced language

technologies so as to substantially enlarge the uptake of the technologies.

The ACCUMULATE technology will have a large impact as multiple companies in the industrial

advisory board (IAB) of the project have recognized the machine understanding of free texts in clinical

reports as the missing link in medical knowledge management and hospital document management

systems. The companies Agfa Healthcare, CompuGroup, CrossLang, Custodix, IMS, InterSystems,

Ontoforce, SmartSpoken and Textgain are eager to integrate the results of the language technology in their

systems. In addition, the end users, which are the pharmaceutical companies (Janssen Pharmaceutica) and

hospitals have also shown a large interest in adopting the ACCUMULATE technologies. Last but not least,

a substantial impact is to be expected on language technology specialized in content understanding (on non

well-formed texts in multiple language) in Flanders. This influence might go as far as content

understanding of speech (e.g., on the processing of clinical speech data). Beyond uptake of the

technologies, social impact and benefits can be expected in the management of healthcare by governments,

health insurers, and other stakeholders.

Team

• Tim Van den Bulcke, Co-promoter (External)
• Elien Bellon, Co-promoter (External)
• Walter Daelemans, Co-promoter (External)
• Jan Aerts, Co-promoter
• Bart Van den Bosch, Co-promoter (External)
• Patrick Van Eecke, External Partner (External)
• Maarten Laga, External Partner (External)
• Lieven Annemans, External Partner (External)
• Griet Jappens, External Partner (External)
• Marie-Francine Moens, Promoter (External)

Financing

Funding: IWT - Agentschap voor Innovatie door Wetenschap en Technologie 

Program/Grant Type: IWT SBO - IWT Projects Strategic Basic Research