Project promotion materials:

Project Homepage:

Data collections can be seen on:

Software is available at:

Project Members:

Andrew Janke (Project Manager,

Harald Waxenegger (Senior Programmer,

Oliver Nicolini (Programmer,

Meng-Kuan Lin (Project Adviser,

Jeremy Ullmann (Project Adviser, )

Graham Galloway (Project Adviser, )

ANDS Contact:

Mingfang Wu (

Project Status:


Brain Mapping National Resource

University of Queensland

Collaborator(s): Monash University, Howard Florey Institute, University of New South Wales, National Imaging Facility NIF

Project Description:

The TissueStack project is aimed at high resolution 3D imaging with an initial focus on neurological and biological sciences. The project has equal applicability to and has been used in other fields with a need to disseminate large 3D datasets including mining, chemical engineering and digital curation.

Modern scientific imaging systems generate massive volumes of data (up to 1TB per acquisition), this has made collaborative research difficult as the data is both unwieldy to transport and difficult to view from multiple platforms without specialise software and hardware. The objective of this project was to break down this barrier to collaborative research by developing a web based viewing system for large imaging datasets akin to Google maps but with the more traditional three perpendicular views of a single 3D object.

The resulting viewing platform is usable on both desktop and mobile systems (tablets and mobile phones utilising typical mobile networks). There has also been the side benefit in that researchers without access to specialist viewing software that is typically only available at the scanning instrument can now view the whole of their data without limitations.

The projects novelty lies in the ability to rapidly view and collaborate on very large 3D datasets via a web browser. This allows researchers to collaborate and share links that direct their collaborators to a specific point in a dataset. The imaging fields and particularly neuroinformatics are beginning to share and setup large data repositories. TissueStack will allow these databases to add viewing capability to their web based portals.

Research Champion:

Prof Jozef Gecz, Paediatric Neurogeneticist, University of Adelaide

Prof George Paxinos, NHMRC Australia Fellow, UNSW

Prof Chris Goodnow, Director of the Australian Phenomics Facility, ANU

Prof David Reutens, Director of the Australian Mouse Brain Mapping Consortium, UQa

Prof Graham Galloway, Director of the National imaging Facility, UQ

Data Type:

Input Data:
18 16.4T Magnetic Resonance Imaging (MRI) 3D mouse brain images from 18 wild type C57BL/6J individuals Approximately 400 2D Histology digitised sections per individual of wild-type C57BL/6J mouse brains for 2 individuals. Text-based wild-type mouse brain anatomy (as defined by Franklin-Paxinos) Approximately 200 manually delineated 3D regions corresponding to wild-type C57Bl/6J mouse brain anatomical structures (as defined by Franklin-Paxinos) Magnetic Resonance Imaging (MRI) 3D mouse brain images from several mutant (GABBAgamma2R43Q ) samples 2D Histology images of mutant (GABBAgamma2R43Q ) mouse brains Zebrafish MRI data from 3 wild-type strains (AB, TU and TL) ~30T2* scans + 10 DWI scans. Zebrafish Histology data ~300 digitized sections per individual for 10 individuals Zebrafish immunohistochemistry data

Output Data:
The data that is displayed in the web app will be freely downloadable for offline use. No specific dataset will be created as this project is ultimately about combining multi-scale imaging data in a single accessible interface.

High Level Software Functionality:

The TissueStack project has filled the current gap in research between those in imaging centres and laboratories who acquire high resolution data and those who use this data in their own research. Typically this gap is manifest due to the lack of cross platform tools to readily view and interact with data. There are well developed tools on some platforms (notably Debian Linux) but less on others (notably Windows). In addition to this most scanners will acquire data in their own native format and as such if post-processing is required the data must be first converted by those with the technical expertise to do so. This expertise is typically limited to researchers or staff at the imaging site.

The Tissuestack project meets both of these needs by first providing a web based viewing system for users of imaging that doesn’t require any extra software installation. The conversion from the most popular formats is also handled by the TissueStack server meaning there is a central and controlled repository of code to perform the conversions.

The Tissuestack software stack makes use of multiple open-source platforms such that there are no barriers to access to the software. The server component runs on a standard Apache/Linux webserver and integrates with a PostGres SQL database for meta-data storage. The client viewing web application does not use any proprietary visualisation platforms such as Flash/Shockwave and instead makes heavy use of HTML5 elements. The use of HTML5 for the client has meant there is a reduction in the amount of effort to maintain both desktop and mobile versions of the tissuestack interface as code can be shared. The Tissuestack application has been tested on Chrome, Firefox, Safari, iPad, iPod, Android and later versions of Internet Explorer.

The specific technical challenge our centre faced with collaborators and users of our imaging equipment was that of data sharing. By running a TissueStack in our department we have massively lowered the bar to entry into the world of the interactive viewing of large 3D datasets. All that is now required is a web browser. This change has driven further collaboration and use of existing data in our centre as novel users can also engage with the data rather than just the static screenshots that are traditionally presented.

This has resulted in increased involvement of collaborators (worldwide and locally) as they can explore their own datasets. In addition to this there is now less “dead” research time wasted in getting collaborators up to speed with the use of desktop viewing software on multiple platforms with multiple image formats that were not always conducive to research.