For many years, foundations and
non-governmental organizations
have focused their efforts in Africa
by donating cash, electronic
devices, or even whole labs.
However, a fundamental gap for
creating a solid scientific infra-
structure is the lack of trained staff
and academic professionals. In this
context, we coordinate the AFRICA
BUILD project to build the infra-
structures needed to increase
learning, research and collabora-
tive health activities in Africa. We
have created the first social net-
work for African biomedical re-
searchers through the AFRICA
BUILD Portal —a “facebook for
medical professionals in Africa”.
This facility includes many free and
open technological and educational
resources for training and support
of African students and profession-
als. Two pilot projects related to
training in HIV-AIDS and reproduc-
tive health were designed as a
proof of the AFRICA BUILD con-
cept. With such an approach, we
are building a network of virtual
communities in various biomedical
topics, fostering new collaborative
South-South biomedical initiatives.
References:
Ramirez-Robles, M., Jimenez-Castellanos,
A., Khalifa, A., Anne, A., KAMGA, Y.,
Afagbedzi, S. Maojo, V.: AFRICA BUILD
Portal: Developing A Social Network of
African Health Researchers and Educa-
tors. Proceedings of IST-Africa 2013
Jimenez-Castellanos, A., de la Calle, G.,
Alonso-Calvo, R., Hussein, R., Maojo, V.
Accessing advanced computational
resources in Africa through Cloud Compu-
ting. In: Proceedings of IEEE CBMS 2012
Jimenez-Castellanos
,
A., Maximo Ramirez-
Robles
,
M., and Maojo
,
V. Creating an
African biomedical research community
through a social network. Proceedings of
AMIA Annual Symposum 2013
Jimenez-Castellanos A, Ramirez-Robles M,
Shousha A, Bagayoko CO, Perrin C, Zolfo
M, Cuzin A, Roland A, Aryeetey R, Maojo
V. Enhancing Research Capacity of African
Institutions through Social Networking.
Proceedings of Medinfo 2013
AFRICA BUILD
The objective of our EC-funded
DICODE project is to facilitate and
augment collaboration and deci-
sion making support in data-
intensive and cognitively-complex
disparate research disciplines. The
DICODE project aims to develop
innovative big data methodologies
by providing seamless integration
and interoperability among existing
and new applications under a
unique web-based platform. This
platform will enable users to work
collaboratively, sharing applica-
tions and data, to facilitate the
decision making tasks. The DICO-
DE approach and platform have
been evaluated by experts in three
different domains: bioinformatics,
medical informatics and social
media, with the participation of
various leading companies in this
area. Text and opinion mining
techniques were applied to analyze
‘
big data’ coming from specialized
literature and the unstructured
Web 2.0. Information in the social
networks can facilitate access to
population trends and attitudes,
which must be analyzed and
filtered using cutting-edge techni-
ques and approaches.
References:
De la Calle G, García-Remesal M, Tzaga-
rakis M, Christodolou S, Tsiliki G, Karaca-
pilidis N. On a Meaningful Integration of
Web Services in Data-Intensive Biomedi-
cal Environments: The DICODE Approach.
In Proceedings of the 25th IEEE CBMS
2012
De la Calle G, Alonso-Martínez E, Tzagara-
kis M, Karacapilidis N. The Dicode
Workbench: A Flexible Framework for the
Integration of Information and Web
Services. In Proceedings of IIWAS 2012
Cases M, Furlong LI, Albanell J, Altman
RB, Bellazzi R, Boyer S, Brand A,Brookes
AJ, Brunak S, Clark TW, Gea J, Ghazal P,
Graf N, Guigó R, Klein TE,López-Bigas N,
Maojo V, Mons B, Musen M, Oliveira JL,
Rowe A, Ruch P, Shabo A, Shortliffe EH,
Valencia A, van der Lei J, Mayer MA, Sanz
F. Improving data and knowledge
management to better integrate health
care and research. Journal of Internal
Medicine 2013
Since 2004, we have been working
on developing models and tools to
integrate clinical trials databases,
following semantic approaches.
Years after working on ACGT
(
advanced clinic-genomic trials on
cancer), the objective of our FP7
INTEGRATE and EURECA projects
is to advance research in oncology
through a unique accessible
biomedical infrastructure integrat-
ing diverse datasets, building
predictive bionetworks and offer-
ing advanced tools to guide diag-
nosis and therapeutics. Based on
multi-centric clinical trials pro-
grammes on breast cancer and
other oncology domains, INTE-
GRATE and EURECA exploit a
collaborative environment to
combine multi-scale biomarkers
(
from genetic level to tissue level
including imaging biomarkers) to
define a methodology to improve
the prognostic power of practices
for assessing modern therapies in
cancer treatment. Working togeth-
er with partners such as Philips
and various leading oncology
centers from Europe, we aim to
develop a new framework for
future clinical trials.
References:
D. Perez-Rey, A. Jimenez-Castellanos, M.
Garcia-Remesal, J. Crespo, V. Maojo.
CDAPubMed: a browser extension to
retrieve EHR-based biomedical literature.
BMC Medical Informatics and Decision
Making 2012
Maojo V, García-Remesal M, Billhardt H,
Alonso-Calvo R, Pérez-Rey D, Martín-
Sánchez F. Designing New Methodologies
for Integrating Biomedical Information in
Clinical Trials. Methods Inf Med 2006
Martin L, Anguita A, Graf N, Tsiknakis M,
Brochhausen M, Rüping S, Bucur A et al
ACGT: advancing clinico-genomic trials
on cancer - four years of experience. Stud
Health Technol Inform 2011
Aso S, Perez-Rey D, Alonso-Calvo R, Rico-
Diez A, Bucur A, Claerhout B, Maojo V.
Analyzing SNOMED CT and HL7 Termino-
logy Binding for Semantic Interoperability
on Post-Genomic Clinical Trials. Proceedi-
ngs of Medinfo 2013
Clinical Trials
and
Cancer Research
Applications of digital imaging
include the enhancement and
filtering of noisy images, the
segmentation of regions of inter-
est, the extraction of measure-
ments, and shape processing. The
main areas of our work have been
the following:
(
a) Theoretical and practical
aspects of morphological connect-
ed filtering (which can preserve
the shapes and forms in input
images), including the so-called
“
levelings”.
(
b) Shape interpolation methods
that allow to impose shape inclu-
sion restrictions that can preserve,
if desired, certain homotopy
properties of the interpolated
images.
(
c) Segmentation techniques, such
as variants of the morphological
watershed that include shape
constraints, and region merging
methods. Some application do-
mains have been the segmentation
of internal structures of the brain
and the extraction of particles in
pathology.
References:
Jose Crespo, Jean Serra, and Ronald W.
Schafer. Theoretical aspects of
morphological filters by reconstruction.
Signal Processing 1995
Jose Crespo, Ronald W. Schafer, Jean
Serra, C. Gratin, and F. Meyer.
The flat zone approach: A general low-
level region merging
segmentation method. Signal Processing
1997
Jose Crespo and Victor Maojo. New
results on the theory of
morphological filters by reconstruction.
Pattern Recognition 1998
Javier Vidal, Jose Crespo, and Victor
Maojo. A shape interpolation
technique based on inclusion relationships
and median sets. Image and
Vision Computing 2007
Jose Crespo and Victor Maojo. The strong
property of morphological
connected alternated filters. Journal of
Mathematical Imaging and
Vision 2008
Big Data and
Opinion Mining
Image Analysis
and
Processing
Biomedical
Applications in
Imaging
Doctors need tools to use and to
manage volumetric radiological
data (three-dimensional imaging
data, such as TC and MRI). We
have worked on applications of 3D
visualization of radiological data to
navigate inner parts of the body
and to model inner structures
(
using image segmentation tech-
niques as well). We aimed to
utilize relatively inexpensive
equipment, such as PCs with
specialized volumetric
visualization hardware, for surgical
planning purposes in virtual
endoscopies.
We have also worked on medical
imaging databases and PACS that
are scalable and that can be used
in both department-wide applica-
tions and in isolated workstation
settings. Such applications benefit
from an easy-to-use medical image
explorer to interact with image
databases, allowing, if desired,
remote collaboration sessions
among doctors.
References:
Alberto Muñoz, Joaquín De Vergas, José
Crespo. Imaging and Clinical
Findings in Patients with Aberrant Course
of the Cervical Internal Carotid Arteries.
The Open Neuroimaging Journal 2010
Raúl Alonso-Calvo, José Crespo, José
Crespo, Victor Maojo, Alberto
Muñoz, Miguel García-Remesal, David
Pérez- Rey. Cloud Computing
Service for Managing Large Medical
Image Data-Sets Using Balanced
Collaborative Agents, Advances on
Practical Applications of Agents and
Multiagent Systems 2011
Alberto Muñoz, Isidro Mateo, Valentina
Lorenzo, Jeronimo Martinez, and
Jose Crespo. MR cisternogra-
phy/myelography of post-traumatic spinal
CSF fistulae and meningeal lesions in
small animals. Acta Radiologica 2013
Vargas-Vázquez, D., Crespo, J., Gabriel
Ríos-Moreno, J, Trejo-Perea,
M, and Maojo, V. Reconstruction with
criterion from labeled markers:
new approach based on the morphologi-
cal watershed. J. Electron. Imaging 2010
