CSIRO researchers are developing novel applications for processing and understanding health and medical data.
Information captured and generated during clinical treatment is valuable for patient treatment, and for clinical and medical research. However, accessing and querying patient data can be complicated by multiple storage locations and formats, and also raises concerns about confidentiality.
A group of projects under the heading 'Health Information Environment' tackle problems associated with the life cycle of clinical data, including storage, analysis, integration and visualisation.
The projects are contributing to the:
-
extraction of information about patient health from primary data, such as physiological monitors, pathology reports, and doctors' notes
-
integration of all the data into a single framework to provide information for decision support and research.
There are two main issues in bringing this data together:
Projects in this research area aim to build information tools that help in the identification of clinical features, in the structural and semantic integration of the data for further analysis, and in novel analytical methods.
Health Data Integration (HDI)
The HDI software tool allows sharing of clinical data from many databases, by creating a virtual dataset without revealing patient identities.
This enables research and analysis on a larger scale than would be possible on the individual databases alone. This project is part of CSIRO's Preventative Health National Research Flagship.
Read more about HDI - Opening up access to health data while protecting privacy.
Following long complex data trails
Health Information Environment projects tackle problems associated with the life cycle of clinical data, from storage, analysis, integration and visualisation.
When patients live with chronic disease for lengthy periods, much of their data is recorded in a myriad of formats and databases.
Consistent and reliable data access is difficult.
We have therefore developed a mapping tool to convert these scattered and diverse data sets into standard terminology as part of the CSIRO Preventative Health National Research Flagship.
The standard we are using is SNOMED CT™. This allows consistent access to large volumes of data.
Mapping to SNOMED CT™
We are building tools to describe data that already exists in databases, using codes from the SNOMED CT ™ clinical terminology.
This enables the building of new applications that use relationships in the terminology to extract information from datasets with a mixture of terminologies.
SNOMED CT™ has been endorsed in Australia and worldwide as the preferred clinical terminology for collection, management and sharing of health-related data.
snorocket™
snorocket processes the SNOMED CT™ clinical terminology to prepare it for fast and efficient querying.
It is much faster than comparable tools, and will assist CSIRO's development of data management tools that use SNOMED CT™.
Displaying and analysing the data
Displaying such a large amount of data collected over time can also be difficult. To allow visualisation of this data we have developed a simple timeline approach, which shows different events during the patient journey on a timeline.
The patient timeline allows clinicians to drill down into individual time points to reveal more details.
We are also developing novel analysis techniques to better understand the progression of disease over time.
Querying of large complex datasets
Making best use of the large amount of monitoring data (such as EEG and ECG) not only requires new ways of processing the data, but new and inventive ways of using it.
Information obtained during clinical treatment can provide valuable knowledge for future treatment, for the individual patient and for other patients.
With such a flood of data being recorded, novel techniques are required to process it and retrieve information.
This project is part CSIRO's Preventative Health National Research Flagship.
Statistical modelling and pattern recognition
Our health data and information group is working with clinicians across Australia through CSIRO's Preventative Health National Research Flagship, using advanced statistical methods to extract knowledge from clinical data.
We also work closely with CSIRO's Division of Mathematical and Information Sciences. Our goal is to provide information about health service delivery, and the initiation and progression of particular diseases.
These projects are carried out by the Australian e-Health Research Centre, a partnership between CSIRO and the State of Queensland.
Read more about the Australian e-Health Research Centre.
Site Map | Legal Notice and Disclaimer | Privacy | Copyright | Last updated: 18 July 2008
