Health informatics covers a large number of research areas: Health care analytics and modeling, data and text mining, system development, modelling and information access for systems within the health care sector. More than twenty researchers at DSV work within the field. Many of the research projects are carried out as a partnership between organisations and industry. Research within health care analytics and modeling is applied to real-life problems and results are brought back to the university to research and teaching.

  • Text mining for better health. In collaboration with Karolinska University Hospital over one million patient records, Stockholm EPR Corpus, from the years 2006-2010 encompassing over 500 clinical units and containing both structured information as well as unstructured information (free text) have been analyzed automatically. The aim of the analysis is to find both new and hidden relations, between, symptoms and diagnoses (comorbidities), age, gender, social situation of the patients, (see demonstrator to the right).  We have also used this abundant data base to develop new tools that can to detect hospital acquired infections (HAI) indicated in clinical text but also to extract the most relevant part of the patient records, as a summary or an overview, see Press release (in Swedish) for more examples on such tools.
  • High-performance data mining for drug effect detection.  In addition to Karolinska University Hospital is also University of Borås, AstraZeneca and WHO-Uppsala Monitoring Centre involved in developing techniques and tools to support decision making and discovery of drug effects by analyzing patient records, drug registries, case safety reports and chemical compound data in the form of both structured and unstructured (free text) data. The research contributes with novel approaches to data mining and clinical text mining and platforms for large-scale analysis of massive, heterogeneous and continuously growing data sets. These tools can be used by clinical researchers in their daily research as well as for hospital management to obtain real time monitoring in health care, so called medical business intelligence.

  • Mobility for better health. Elderly and handicapped people benefit greatly from being able to live at home. However, the social care sector is extensively controlled by public authorities, and services that make life easier, such as alarm telephones and attendant carers, require decisions by local officials. We are studying how “open social services” should be formulated and introduced. Through e-services, customers can increase their insight and reduce the impact of public authority. We are also studying how professional user group adopts mobile and handheld technology. The research area includes researchers from the fields of computer and system sciences, law and social work.
  • Health communitiesE-health communities have contributed to a shift of focus: from health care and health care givers being the only source of help for most patients, to a situation where patients and citizens take active and daily part in the health and recovery of each other. The e-health communities exist in different health areas, from severe physical and mental disorders to aftercare and lifestyle problems, such as bad eating habits, smoking and stressful lives. We study ways of designing and using these e-health communities for communication and learning in groups. We also look for holistic solutions where the community of peers is related to other sources, and with the interaction with expertise. Flexibility issues are also being targeted, and we explore the combination of different technology tools. How to systematically evaluate these systems constitutes another important research topic.
  • Information access for better health
    Today patients are actively trying to find answers on Internet on their complaints before visiting the physician. We have developed information access tools to improve the quality of the search results, by correcting misspelled search words, adapting the search words to the morphology of the words in the text, but also by automatically translating the search query to other languages, cross lingual information access. We have also created tools that can automatically answer queries in emails. 
  • Visualisation of business models within medical care. Medical and general nursing care requires collaboration between many players within fields such as primary care and specialist care. One way of describing and analysing this complex interaction is via business models. A business model can graphically describe the actors that are working together, the values that are being generated for the patient and the way in which the medical and general care is financed and managed.
  • Simulation in health. Many healtcare areas are complex and traditional care and diagnose methods does not work as good as they should. In such cases, computer based Simulation and Visualization systems might improve the clinical situation. At DSV we are performing research on simuated environments for treatment, rehabilitation and diagnostics. In those projects we are developing and studying the effects of for example Virtual Encounters and similar similated environments for clinical use. Examples of projects are Reactions-on-Display - RoD/Forensic Psyhciatry and RoD/Intimate Partner Violence.