PosterPDF Available

Web-cloud platform for storing, processing and analyzing multi-modal neuroimaging data



In this work, we present one of our first designs for the QMENTA web-based cloud system used to store, process, analyze, and visualize aspects of the neuroimaging muli-modal data.
Nikola Lazovski1, Marc Ramos1, and Paulo Rodrigues1
1Mint Labs, S.L., Barcelona, Spain
Data organization and solid data analytics tools
are essential for generating good and reliable
results. Moreover, any system providing these
features has to provide means of collaboration
between users and relatively fast data accessi-
bility in order to reduce the research cost and
time, and to boost the quality of the outcome.
This also applies to neuroscience. However, as
far as the practice suggests, the neuroscien-
tists often use general purpose storage tools
(such as DropBox) to share data, which are of-
ten unorganized and unsearchable. In this work
we have designed and developed a web-based
cloud system used to store, process, analyse,
and visualize aspects of the neuroimaging muli-
modal data.
The system involves two main physical and logical parts: cli-
ent side and server side. The client side is represented by the
client platform loaded into a web-browser which means there
is no need of installing an additional software or plug-ins. The
functionality provided to the user is divided in four main cli-
ent modules: project management module, neuroimaging data
management module, subject metadata management module,
and analysis management module.
The project management module (See Figure 2) embraces the
concept of a project as a central entity in the system - each
project is dedicated to one type of research that the users want
to do. Within our platform, each project contains disjoint set of
neuroimaging data, subject definitions and performed analysis.
Each user can create many project as needed and can include
other users within a project.
The neuroimaging data management module (See Figure 3)
allows uploading scan session data sets (i.e., T1, T2, Diffusion
Weighted Imaging, fMRI, among others) and uploading other
session-related data. Every scan session data-set is attached to
only one subject-under-investigation (e.g., human patients or
animals) identified by a unique subject ID.
When uploading scan session images, the user uploads only one
ZIP file over which the system performs classification , modality
recognition, and anonymization of each file within the ZIP file.
At the end of this scanning process, the user is presented with a
report containing the results of this recognition and classifica-
tion process. This way the user does not lose time in searching
and extracting the important neuroimaging data this process
is performed by the system itself.
The subject metadata management module (See Figure 4)
provides tools to manage additional data related to the sub-
jects-under-investigation. Aside default CRUD (i.e., Create-
Read-Update-Delete) operations, the user can define additional
descriptive and quantitative parameters and assign tags to sub-
jects making them easily searched and clustered by using the
advanced search engine. Also, this module generates a profile
page for each subject where all information, scan-session data,
and the analysis related to the chosen subject are presented in
one place (See Figure 5). Import and export of these data are
also provided as a feature of this module.
The analysis management module (See Figure 6) provides a
list of predefined projects or data processing templates (e.g.,
scalar maps derived from DWI data, tractography maps, con-
nectome construction and visualization) executed over the
scan-session data stored in the neuroimaging data management
module. When the data processing finishes, the user can see
the results and download them in a certain format.
With respect to its quality, the platform provides good perform-
ance, solid data protection and data availability. The platform,
regarding the cloud resources, relies on the Google Cloud serv-
ices: the data uploaded by the user are stored and secured in
the Google Storage and the system can be instantly extended
with more resources (e.g. server nodes), if more processing
power is needed.
This system has already showed its value in
practice. Dozens of users have stored, shared
and analyzed their data. However there is more
to come in the near future, such as: more ad-
vanced tools in the analysis module, and more
sophisticated classification and recognition
The academia and the market offer various,
but not numerous, software systems and tools
that can be used in the neuroimaging analysis
domain. However, neuroscientists need an easy
to use, well supported and highly integrated
system covering many aspects of the neuroim-
aging data, where the technical details are
simplified and the data-flow can be focused
on hypothesis experimentation and validation.
That is exactly the goal we tend to achieve
with this system.
Figure 3. Main form used for listing and managing
scan session files. In the background, the main data
workspace is shown.
Figure 4. The workspace of the subject manage-
ment module. The subject can be easily searched by
using the advanced search at the right side of the
Figure 5. For each subject, the user can
open its profile where all data related to
that subject are presented in one place.
Figure 2. Starting welcome page presented after
logging into the platform. All projects available to
the user are listed at the top right side of this page.
The user can easily activate and use the project
needed at the moment.
Each project can be shared among any number of us-
Figure 1. Abtract level diagram representing the
platform and its ecosystem.
Figure 6. The platform provides variety
of tools that can be applied over the up-
loaded data. The user can either download
the result files or can visualize them in a
certain format.
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.