Responsible for this page: Webmaster, webmaster@imt.liu.se
Page last updated: 2011-11-21
LiU » LiTH » IMT »
people »
~erisu
Research timeline presented Nov. 2011: online@Prezi or offline (25 MB)
Telephone, mail, office location etc
available at: http://www.imt.liu.se/staff/
If you want to get updated of the
research progress (publications etc.), just send me a mail
marked [research updates] in the subject line and you'll be
informed (at most a couple times a year).
How can patient information in an electronic health record (EHR) be displayed best in order to gain a quick overview of a patient's history and other important facts? The thesis aims to explore different facets of this question. Requirements vary depending on the user's role and current task, the overview should adjust accordingly.
The need for more structured, computer interpretable and semantically well defined EHR content has this far to a great deal been driven by the desire to use decision support systems and reuse data for follow up purposes. The thesis will also use this improved structure to enhance overview, presentation and navigation of the EHR using various visualization techniques. A hypothesis is that automated reasoning may also be used to better filter/summarize available information before presentation.
The
structure
of the information in an EHR, and the
possibilities to aggregate data for statistical purposes are
dependent
on data models, terminologies etc. used. Good overview and
navigation
of those is a related important task that is also explored in the
thesis. The first
scientific paper
thus presented a prototype (TermViz)
applying well known methods like
"focus+context" and self-organizing layouts from
the fields of
Information Visualization and Graph Drawing to terminologies like
SNOMED CT and ICD-10. The aim was to simultaneously focus on
several nodes in the terminologies and then use interactive
animated
graph navigation and semantic zooming to further explore the
terminology systems without loosing context. The prototype, based
on
Open Source Java components, demonstrated how a number of
information
visualization methods can aid the exploration of medical
terminologies
with millions of elements and can serve as a base for further
development. Status
update:
Web/Flash/Flex-based TermViz
2.0 under development, alpha version available.
In order to have "semantically well defined EHR content" (needed for advanced EHR overviews) we need a formalized EHR structure and well defined connections to terminology systems etc. Thus I am active in our department's investigation of the archetype/openEHR-approach to EHRs (see http://www.imt.liu.se/mi/ehr/). The second paper describes an Archetype Editor developed by our department that was combined with TermViz and a system for finding terminology bindings to SNOMED CT.
Display
of interactive graphical "timelines" showing
patient history is an obvious and important part of a patient
overview.
Work along these lines has already been conducted by others (e.g.
the LifeLines
project) and is
being further explored and expanded in papers and software. Our third
paper, entitled Graphical
Overview and Navigation of Electronic Health Records in a
prototyping
environment using Google Earth and openEHR Archetypes,
introduces
some possible interaction and visualisation techniques suitable
for EHR
overviews. Evaluation and enhancement of this is one of the
current
research activities.
The use of graphic languages (pictograms etc.) for quick information overview is another thing expected to be further explored in the thesis.
Finding ways of overviewing and navigating existing EHR content is a main task in the thesis, and an interesting question to discuss in relation to this is the possibility of reducing the amount of redundant information often found in EHRs. Can suitable EHR information structures and user interfaces make it easier to find and refer to already entered information instead of re-entering it? How can patient-specific goals, plans and processes be represented and visualised, and will that be useful for overview purposes and reduced redundancy? Questions like these are likely to be touched upon in the thesis too.
The continued exploration and research depends on what can be concluded by the initial findings.
A PhD thesis at our department is expected to be completed within approximately five years (four years PhD work plus 20% teaching etc). My thesis work begun in August 2004 and will be extended due to parental leaves and department work in projects that are not part of my research.
Supervisors: Håkan Örman (formerly Petersson), late Prof. Hans Åhlfeldt (until the summer 2010), Prof. Magnus Borga (from fall 2010)
I have been involved in the teaching of the courses: