Tuesday 27 May 2014

ANDS AP24 BMRI Application - Final Product Report
May 28th, 2014
The AP24 project (Brain  and Mind Research Institute Application) has reached end of development and we are happy to announce the BMRI Application is available for the general public.
Introductory Product Information
One of the main problems in Brain and Mind Research Institute is that there is no such a tool that can efficiently and effectively collaborate the data captured from multiple tests.  The researchers need to collect, collate, and compile the data and the images together manually in order to perform analysis.  Another is the pre – processing of the MRI images. In order to read the image, research needs to pre – process the raw image from the scanner. Professor Jim Lagopoulos spends hours and hours a week just on helping the pre- processing because he is the only person who has the expertise to do it quickly.

These manual processes consume significant time and effort, replacing these by an automated process will be a huge success in the research area.

The system will improve the process efficiency by replacing the manual pre – processing to an automated pre- processing based on the researcher’s requirement.  Moreover, system integrates multiple sources of records and presents the summery of linked records according to user’s query, giving the user an overall view of related records. This automated functionality is previously not existed in this research filed of BMRI the University of Sydney.

The new process and functionality that has been brought by the application is highly welcomed by the researchers in BMRI, including:

  • Integrate, manage and store a diverse range of clinical research data.
  • Run MRI Scan pre-process through the process
  • Ingest existing research data
  • Enable users to manage access to research data.
  • Allow user to define and queries on research data.
  • Provide RIF-CS 1.4, Service, Party, Input Collection and Output Collection records to Research Data Australia.
  • Meets security standards.
  • Meets minimum accessibility requirements.

Project Aim

The aim of the project was to integrate various (a) imaging, (b) neurological & electrophysiological testing data and (c) clinical data for the purpose of providing a refined answer to complex research questions that was not previously possible using the separately existing (and expanding) data sets across the clinical research programs at BMRI making use of brain scans.

The project provided an integration tool for the multiple data sets and data points within those data sets and developed an analytical tool for interrogating the data in response to specific research questions. For example, researchers at the BMRI are trying to establish if there is a characteristic pattern that emerges from MRI structural imaging data and a specific electrophysiological component that predicts with some degree of accuracy a clinical phenotype, such as depression in young people.

The system developed through this project provides an integrated research data set to be expanded and used for future research projects. An additional objective was to integrate contributions to the data set from other clinical researchers and to make the larger data set available through research collaborations which was achieved by linking to the National Imaging Facility (NIF) via the DARIS application and has also enabled a University of Sydney wide imaging platform.

The outcomes of the project will provide data and an analysis platform for the development of information based diagnostic and treatment response algorithms for young people with major mental health problems like depression and psychotic disorders, as well as ageing subjects.

AP24 has provided a software solution covering the following functionality:
1.       Integrate large and sometimes disparate data sets held in SPSS and provide a platform to enable data transformation and analysis by authorised parties.
2.       Provide an analytical tool for interrogating the data in response to research questions (of the type discussed above).
3.       In cases where a research subject is under clinical care, and research findings may provide additional diagnostic support, provide mechanisms for communicating findings to the clinician.

Project outcomes

Prior to this project, researchers would manually collect, collate, and combine the data from the three information sources i.e. MRI, Neuropsychological, and EEG data by querying the various datasets for the specific requirements. The project has implemented an automated system that not integrates the data in one place, it also enables queries via metadata and attributes down to the field level to assist researchers in collecting data for analysis.


Instructional Product Information

The University of Sydney has chosen Daris as the solution software. The DARIS application utilises MediaFlux software and provides functions required by University of Sydney researchers.

The key functions required that Daris will be able to produces are:

  • Storage of MRI data
  • Storage of Neuropsychological data
  • Enable uses to manage access to specific data sets
  • Enable users to describe research data collections
  • Enable users to query the systems research data collection
  • Enable users to download research data
  • Enable the visualisation of research data
  • Make available RIF-CS 1.4 metadata available for harvesting via OAI-PMH.
  • Enable the creation of output collections
  •  
Solution components are broken down into the following:
  • Access control for authentication and authorization purposes.
  • Admin functionality is controlled by roles based permissions. Functional permissions management is a permissions controlled function that allows users to grant and revoke data access and to manage user roles.
  • Other function is to implement business functions. For example building form that allows users to add and update Neuropsychological data.
  • Reporting is a permissions controlled function that packages, compresses and makes datasets available for download by authenticated users.
  • File system storage will store the pre-processed data and its extended information like index.
  • Relational database will be used for system management configuration or user management information only.
  • Email Connector is the components that send appropriate emails to application users.
  • The OAI harvester makes RIF-CS 1.4 compliant metadata records .xml files public available for harvest Research Data Australia is a RIF-CS 1.4 .xml harvester.

Getting started

Please refer to the Developers Guide on instructions on how to download or build the tools.

Documentation

There are a number of manual and document resources available:
·         User Manual
·         Github repositories: Daris Wiki Home Page
·         Blog
·         Technical documentation:
Developers Guide- includes links to deployment and developer guides

Overview diagram

The following diagram outlines where the different tools fit within the research paths.




Product re-usability information

Imaging and clinical testing is not limited to neuroscience and mental health research, but extends to the University's research priority areas of cardiovascular disease, diabetes and obesity clinical research. As such the system will be developed to be disease independent and re-deployable for other systems (this will include the University of Sydney Charles Perkins Centre (formerly Centre for Obesity, Disease and Cardiovascular Disease (CODCD) or other research programs as required)  

The system has been designed to be able to be redeployed to other sites outside the University of Sydney. The source code is documented and packaged within http://nsp.nectar.org.au/wiki-its-r/doku.php?id=data_management:daris , enabling download and installation elsewhere.

The custom code that relates to the AP24 BMRI Application is code shared by a number of other institutions and has been sponsored by the Australian National Data Service (ANDS) http://ands.org.au/

During the project we have communicated our developments and progress with other Universities with a view gauging interest in redeployment and collaboration e.g. ANDS Distribution list and other meetings.

Contextual Product Information

All code is licensed under the GNU GPL v3 license - see LICENSE.txt in each code repository for license text. Documentation (contained in the Github wiki) is licensed under Creative Commons Attribution-Share Alike
Due to the extensive testing by the research stakeholders, the software is now robust, mature and fit for purpose. The ongoing maintenance of the software may occur in a number of ways, as appropriate:
  • Daris Software support is provided by the DARIS team under http://nsp.nectar.org.au/wiki-its-r/doku.php?id=data_management:daris:about 
  • MediaFlux support is provided by Arcitecta
  • The team at the BMRI and the University of Sydney may continue to enhance the software
  • Further enhancements and fixes may be done by the DARIS team or Arcitecta, under the support and maintenance agreement between the University of Sydney and Arcitecta. The sustainability of the product has been considered throughout the project and the software has been designed to maximise future maintainability:
  • The software has an extensive suite of automated unit tests that clearly describe the expected behaviour of the code.
  • This is augmented by a suite of integration tests created by the research stakeholders.
  • The code is open source so that other groups can contribute code back to the project.
The system has extensive documentation to explain the design and modules of the system and can be found at http://nsp.nectar.org.au/wiki-its-r/doku.php?id=data_management:daris
  •  
Posted by Neal Anderson May 14th, 2014


No comments:

Post a Comment