HIPPA-Hospital Intelligence for better patient security
(In Swedish: HIPPA-Hospital Intelligence för bättre patientsäkerhet)

Project manager: Hercules Dalianis
Participants from DSV:  Erik Perjons, Claudia Ehrentraut, Paul Johannesson, Monica Winge
Participants from Karolinska University Hospital: Ann-Britt Bolin, Gunnar Ekeving, Elda Sparrelid
Participants from Capish Knowledge: Staffan Gestrelius, Eva Kelty och Jerker Ringström
Other Participants: Martin Williamsson, TakeCare founder
Project duration: May 2013 - January 2014
Funding: Vinnova

Summary

Patient safety is a serious challenge for Swedish health care where adverse events cause great human suffering as well as significant economic costs. It is, therefore, important that the Swedish healthcare builds knowledge that can be used to improve patient safety. Such knowledge can be obtained by health care providers that collect and analyze health data for monitoring, evaluation and comparison. This kind type of knowledge management is a complex activity that needs to address a number of issues. Data from different sources need to be integrated and reused. The quality of collected data needs to be secured. Data in different systems need to be unlocked and made available to external actors as patients and companies. Data need to be described and explained so that it can be interpreted in a simple and consistent way. Furthermore, data collection requires today often extensive manual work, as it is difficult to retrieve relevant data from existing systems.
The solution proposed by the project to these problems is to develop a Hospital Intelligence system consisting of models, methods and tools to collect, analyze, and visualize information about patient safety. The solution is based on techniques for semantic interoperability, data warehousing, text and data mining and visualization.
The project results provide a basis for common information structures that make it possible to improve patient safety, reduce societal costs of preventable adverse events, enable companies to develop innovations for patient safety, as well as support the individual patient to select caregivers also based on information about patient safety.

Electronic patient records and language technology

Digital or electronic patient records is a vast source for reuse since a large part of the health care process is documented in free text. As language technology has been developed and matured in recent years one can use automated tools to structure and extract important information. This can be used to improve the health care process in general, diminish adverse events and specifically hospital acquired infections.

Published paper

Ehrentraut, C, H. Tanushi, H. Dalianis and J. Tiedemann. 2012.
Detection of Hospital Acquired Infections in sparse and noisy Swedish
patient records. A machine learning approach using Naïve Bayes, Support
Vector Machines and C4.5. In the proceedings of the Sixth
Workshop on Analytics for Noisy Unstructured Text Data, AND, December 9,
2012 held in conjunction with Coling 2012, Bombay, PDF.