Capturing biomarker metadata for mental health research
Melbourne-based data management systems specialist Arcitecta is working with the Cooperative Research Centre (CRC) for Mental Health to build a web-based repository for clinical observation studies that will open up huge amounts of research data to the 19 different organisations involved in the CRC.
The repository will be used to capture, access and query clinical observation data from longitudinal studies of biomarkers, which form the basis of the research being done by the CRC for Mental Health to identify indicators of the early onset of Alzheimer’s, Parkinson’s, schizophrenia and other mood disorders and psychoses.
Data sources include the CRC's bio-banks of blood and tissue samples, complex MRI data, gene sequences and the results of proteomic studies, and data from clinical observations. Arcitecta will use its Mediaflux metadata management platform, which can handle both structured and unstructured data, as the basis of the repository.
Arcitecta’s software allows researchers to build their own interfaces to access the repository and present data to clinicians, lead investigators and study coordinators in a way that is intuitive and easy for them to use, the company says. It also has a mobile application to enable clinicians such as nurses and other members of a study team to capture data at the bedside.
This will be able to be entered directly through the app if the nurse is visiting study volunteers at home, or collected manually and entered into the repository at the conclusion of the home visits.
However, the web-based repository will be the primary source of data from the CRC for Mental Health’s clinical observation studies, the CRC's lead bioinformatician, Noel Faux, said.
“The focus of this project is to make the collection, access and management of clinical observation study data easier for the users involved in these studies, to reduce the time between the data being gathered and the insights gained from it,” Dr Faux said.
“For example, we previously considered schizophrenia to be a single disease. As a result of being able to assess the information gained from longitudinal studies into this condition, we now know there is a spectrum of schizophrenia that people experience.
“It is easier to uncover this sort of information from bigger data points, which is something the development of the web-based repository and mobile app will make possible.”
Data currently stored in various programs, including legacy databases and spreadsheets, will also be able be stored in the repository, Dr Faux said.
“It basically enables users to pull all sorts of different data together in one place, making it easier to analyse, share and manage,” he said.
“Users will be able to enter data directly into this web-based repository, and have masses of proteomic data, MRI data and other complex and large data types automatically ingested and made immediately available for analysis.
“Users can then either continue using their individual clinical studies software, or enter data directly into the central repository using the web-based interface.”
Dr Faux said one of the great advantages of using a web-based interface and mobile application over individual applications and programs is that the data is available from anywhere and is more easily shared with collaborators.
The interface will be used to develop specialised versions of the repository, based on a generic template created by Arcitecta and underpinned by Mediaflux. Arcitecta CTO Jason Lohrey said the Mediaflux platform will work as a repository, a database and a clearinghouse, with the data 'cleaned’ on entry into Mediaflux.
“It provides one point of access for all data, based on an application that sits on the web and connects to the Mediaflux server, so data can be accessed from anywhere,” Mr Lohrey said.
“It will contain a database of all bio-sampling information, something that is not currently available at the CRC for Mental Health, and it will provide a single repository of structured and unstructured data.
“It will also provide a clearinghouse of quality assured data, as data will automatically be checked for errors as it is ingested into Mediaflux, before it is made available to users via the repository.”
Dr Faux said the data will be presented to researchers and clinicians according to their role in the clinical observation study and the type of information they need to analyse.
“For example, researchers can draw out data on the results of a study in a suitable file format that can be readily loaded into a specialist statistical program, or extract metadata from MRI scans to analyse in another specialist program. Alternatively, these researchers can query and analyse the data directly from the repository, using the web-based interface.
“By bringing all this data together in one place, to a consistent quality standard, and making it immediately discoverable means it will be easier to identify where errors occur in a calculation or results from a study. This means it will be much easier to go back into the data and change the error for all studies using the same calculation.”
A practical example includes checking the dosage of levodopa to treat Parkinson’s disease. As Parkinson's affects each person differently, it is important to tailor medication regimens accordingly.
“Using the central web-based repository, researchers will be able to compare drug dosages to a standardised value," Dr Faux said. "This will allow researchers to adjust for the effect of medication, and so more effectively use the data for biomarker discovery projects."
Arcitecta is currently in the early stages of building the product to the CRC for Mental Health’s specifications for delivery and implementation within 12 months. It will be promoted through the CRC for Mental Health consortium – which includes Austin Health, Barwon Health, the CSIRO, three universities, the Florey Institute for Neuroscience and Mental Health, several biotech and pharmaceutical companies as well as two aged care providers and philanthropic organisations – and its global alliance with high performance computing firm SGI.
The new platform will be testing in multiple sites in Melbourne and Perth, and multiple clinical settings. It is being funded by the Victorian Department of State Development, Business and Innovation with in-kind contributions from Arcitecta and the CRC for Mental Health.
Posted in Australian eHealth