Products & Services

BRISSKit offers the ability to upload, manage, combine and collaborate with datasets in-project using a secure infrastructure and then enables analytics and reuse. The use of open source tools for common tasks allows researchers to integrate their own more specialized software and analytics expertise using the platform.

We have developed a platform on which biomedical research can be done. At its heart are three main existing types of software; study management (including patient cohorts), sample management and survey or other data capture then data warehousing and querying. The relationship between the different pieces of software is illustrated in the figure below, the three central clouds representing the three different types of software.


Illustration of the relationship between the different aspects of the software stack.

The BRISSKit platform is designed to run on the cloud, so all the applications have a web front end to them, making them readily accessible without the need to install any software. Another feature of being hosted on the cloud is that it is relatively quick to create a new instance of the stack for a new user, and it can be charged on a usage basis.

Each of the applications is integrated with the other components above and below it in the diagram, this means data can move seamlessly between applications. Moreover, since the interface each application sees is standard across the stack it is relatively straightforward to add further applications to the stack.

Management layer

CiviCRM

CiviCRM is a web-based, suite of computer software for constituency relationship management, this makes it ideal for managing participants in research studies. In BRISSKit CiviCRM is used for two key purposes in the stack; participant management and study management.

In its participant management role CiviCRM can be configured to have virtually any kind of data associated with a participant record. This takes the form of a contact card where participant details may include contact details etc. It may also include data such as demographics or local identifiers. As part of BRISSKit we have added a permission to contact field that acts as a global indicator of if the participant is willing to be contacted in the future. This can be queried before any communication is carried out.

In its study management role CiviCRM will offer a way for study managers to configure their studies, setting options such as access privileges to different studies for the users.

Data collection layer

Onyx/REDCap
One of the routes that patient data can take into the data warehouse is via electronic questionnaires. Onyx and REDCap are two different solutions that allow the direct capture of this data which is then stored in a database. Both offer an intuitive GUI for designing questionnaires with all the usual questionnaire features like multiple types of input data, branching logic etc.

caTissue
caTissue is a biospecimen tracking and inventory tool produced under the US National Cancer Institute's Cancer Biomedical Informatics Grid (caBIG) (cabig.nci.nih.gov). It has a highly configurable object model, allowing complex physical storage solutions to be accurately modelled.

Local systems
In the diagram above there is an input of clinical data directly into the data warehouse. This is a feature we have developed for one of our pilot groups where we can pull pathology data from the central hospital systems directly into the data warehouse. Since all the data is tied together with unique participant IDs this pathology data is automatically linked to the data gathered with Onyx and caTissue.

Warehousing and analysis layer

i2b2 - Informatics for Integrating Biology and the Bedside

A key component to the software stack is the data warehousing software i2b2. It is developed by the US National Center for Biomedical Computing based at Partners Healthcare Systems / Harvard Medical School in Boston, USA. All the data from each source eventually ends up in the data warehouse, the BRISSKit integration means it is all linked together at this layer. This means that complex cross data-set queries can be performed for example: How many people over 50 that drink alcohol have a low red blood cell count that we have a saliva sample for in the fridge?