NorKlinTekst: Natural language processing to extract knowledge from clinical notes in electronic health records


The Norwegian centre for e-health research (NSE) is a new research centre established in 2016, with a mandate that includes advanced computational and statistical analysis of health data, with a special focus on structuring data in electronic health records (EHR). Current skills in the centre are in the right general fields of computer sciences and statistics, but more specialised focus in the area of language technology (NLP) and machine learning (ML) is required.

Much of the information in EHRs are clinical notes in unstructured text form, and this form is difficult to access and summarize in clinical practice. Manually re-coding clinical text to structured forms can be expensive and is prone errors. New methods of analysing text, such as NLP, enable us to automatically convert large amounts of text into structured data. While research on NLP methods is still in its infancy, the evidence points to promising results in areas such as diagnostic surveillance, identifying cohorts, quality assessment and patient safety.

The incubator support enables the department to build capacity in this new area. This involves professional development of employees, forming new research consortium, generating relevant publications to boost our competitiveness in national and European research funding calls. Increased capacity better positions NSE to participate in local scholarly debate and consult in regional and national programs for developing new health technology.


There are three broad goals for this incubator project;

(i) education and professional development of department researchers,

(ii) hands-on experimental work to improve previously reported specificity in predicting post-operative complications. The previous report was co-authored by some of our partners in this projects. This work generates publication to increase our competitiveness for research funding.

(iii) create consortium to develop new research projects through two PhD projects and a larger NFR project.


The general strategy for achieving these objectives is to partner with leading researchers in the field, both in Norway and abroad.