Abstract—In this paper we present the POCEMON
platform, a platform aiming to the early prognosis and
diagnosis of autoimmune diseases at any point of care, even the
primary. The objective of the POCEMON platform is the
development of a diagnostic lab-on-chip device based on
genomic microarrays of HLA-typing. The POCEMON is going
to advance and promote the primary health care across Europe
by supporting a) point-of-care diagnostics, b) monitoring of
immune system status and c) management of the chronic
multiple sclerosis (MS) and rheumatoid arthritis (RA)
autoimmune diseases. The platform combines high-end
Information and Communication Technologies based on
microfluidics, microelectronics, microarrays and intelligent
Personalized diagnosis and treatment is driven by the rapid
development of information
technologies concerning advances on both hardware and
Manuscript received July 5, 2008.
This work is funded by the European Commission (project POCEMON:
Point-Of-Care MONitoring and Diagnostics for Autoimmune Diseases FP7-
F. G. Kalatzis is with the Unit of Medical Technology and Intelligent
Information Systems, Dept. of Computer Science, University of Ioannina,
Ioannina, Greece (e-mail: firstname.lastname@example.org ).
T. P. Exarchos is with the Unit of Medical Technology and Intelligent
Information Systems, Dept. of Computer Science, University of Ioannina,
Ioannina, Greece, with the Dept. of Medical Physics, Medical School,
University of Ioannina, Ioannina, Greece and with the Institute of
Biomedical Technology, CERETETH,
N. Giannakeas is with the Laboratory of Biological Chemistry, Medical
School, University of Ioannina, Ioannina, Greece and with the Unit of
Medical Technology and Intelligent Information Systems, Dept. of
Computer Science, University of Ioannina, Ioannina, Greece (e-mail:
S. Markoula is with the Medical School of University of Ioannina,
Greece (e-mail: email@example.com).
E. Hatzi is with the Laboratory of Genetics and IVF, University Hospital
of Ioannina, Greece (e-mail: firstname.lastname@example.org).
P. Rizos is with the Unit of Medical Technology and Intelligent
Information Systems, Dept. of Computer Science, University of Ioannina,
Ioannina, Greece (e-mail: email@example.com)
I. Georgiou is with the Genetics Unit, Dept of Obstetrics and
Gynaecology, Medical School, GR 45110, Ioannina, Greece. (e-mail:
D.I. Fotiadis is with the Unit of Medical Technology and Intelligent
Information Systems, Dept. of Computer Science, University of Ioannina,
Ioannina, Greece, GR 45110 and with the Institute of Biomedical
Technology, CERETETH, Larissa, Greece. (Corresponding author; phone:
0030-26510-98803; fax: 0030-26510-97092; e-mail: firstname.lastname@example.org).
Larissa, Greece. (e-mail:
software development. These efforts promote new
diagnostic procedures based on Lab-on-a-chip (LOC)
technologies . The LOC technologies are capable of
performing a wide range of proteomic and genomic tests by
using a sample of blood or other body fluid such as saliva.
These tests aim to facilitating healthcare at preferred
environments and point of care disease diagnosis at primary
healthcare level. The LOC technologies expose a
tremendous potential to improve the health of people
worldwide. Ever since the modern inception of LOC and
microfluidic technologies around 1990 their use in remote
settings has been perceived as potentially one of the most
powerful applications of the technology by taking advantage
of its small size, low volume requirement for samples, and
rapid analysis. Indeed, portable LOC devices are now
beginning to be used in remote settings as a result of
developments in integrating fluid actuation, sample pre-
treatment, sample separation, signal amplification and signal
detection into a single device. These advances place the field
of LOC research in a prime position to tackle the profound
issue of global health where the challenges in device design
are arguably the most demanding and the need for new
health technologies the greatest.
Autoimmune disorders  develop when the immune
system destroys normal body tissues. This is caused by a
hypersensitivity reaction similar to allergies, where the
immune system reacts to a substance that normally would
ignore. The disorder may affect only one organ or tissue
type or may affect multiple organs and tissues. Organs and
tissues commonly affected by autoimmune disorders include
blood components such as red blood cells, blood vessels,
connective tissues, endocrine glands such as the thyroid or
pancreas, muscles, joints, and skin.
There are many diseases which are believed to be
autoimmune. These include Multiple Sclerosis (MS) ,
Rheumatoid Arthritis (RA) , Addison's Disease,
Ankylosing Spondylitis, Autoimmune Haemolytic Anaemia,
Autoimmune Hepatitis, Coeliac Disease, Chronic Fatigue
Syndrome (CFS, CFIDS),
Polyneuropathy, Crohn's Disease, Dermatomyositis, Grave's
Disease, Hashimoto's Thyroiditis, Type 1 Diabetes,
Polymyalgia Rheumatica, Systemic Lupus Erythematosus
(SLE) and many others. More specific, there are over 80
The HLA (human leukocyte antigens) are proteins found
Point-Of-Care Monitoring and Diagnostics for Autoimmune
Fanis G. Kalatzis, Themis P. Exarchos, Student Member, IEEE, Nikolaos Giannakeas, Student
Member, IEEE, Sofia Markoula, Elisavet Hatzi, Panagiotis Rizos, Ioannis Georgiou and
Dimitrios I. Fotiadis, Senior Member, IEEE
in the membranes (outer coating) of nearly every cell in the
body (all cells that have a nucleus) (Fig. 1). HLA antigens
are the major determinants used by the body's immune
system for recognition and differentiation of self from non-
self (foreign substances). There are many different major
histocompatibility (HLA) proteins and individuals possess
only a small, relatively unique set that is inherited by their
parents. It is unlikely that 2 unrelated people will have the
same HLA make-up. Many HLA molecules exist but some
are of special interest because they are more common in
certain autoimmune diseases. For example, HLA-B27
antigen is found in 80-90% of people with ankylosing
spondylitis and Reiter's syndrome and can aid in the
diagnosis of these diseases. HLA-B27 is also present in 5-
7% of people without autoimmune disease. Thus, the mere
presence of this HLA molecule is not indicative of a disease.
HLAs detectable in blood provide clues to immune
There have been some reports on HLA genotyping by
microarray [5,6]. Most of these studies focused on tissue
transplantation and association between HLA genotypes and
disease or pathological process [7,8]. A genotyping
microarray approach can analyze polymorphisms at
multiple-sites in a single gene or in multiple genes rapidly
and in a parallel way . The genomic HLA microarray
scanning data can also be computerized and shared with a
Personal Digital Assistant (PDA) device which will be
hosted in the diagnostic POCEMON platform. Although, the
feasibility of microcantilever based DNA biosensors has
been demonstrated [10,11], as well as, the integration of a
electronic interface realized with a CMOS technology
[12,13], a complete system including the detection module
and the microfluidic structures for the sample handling is
still not available in the market for specific applications in
the genomic field.
In this paper, we describe the methodology which is
applied to the huge amount of data produced by the gene
discovery phase (a Whole Genome Association Study) of
the POCEMON platform to select the most informative
SNPs which are directly related with the RA & MS
autoimmune disorders. Finally, we present the lab-on-chip
technology which will be used to develop the innovative
diagnostic LOC platform based on HLA-typing microarrays.
II. MATERIAL AND METHODS
A major milestone of the POCEMON platform is the
development of specific HLA-typing microarrays capable to
diagnose the MS & RA autoimmune disorders. To complete
this task the following stepwise procedure is applied:
a) Genome-wide association study. Genotyping of 750
control and case DNA samples both from MS & RA
patients and production of a huge amount of raw data.
This process is based on the iLLumina technology
b) Selection of the most informative SNPs by using an
approach based on data mining methods and genetic
c) Development of a diagnostic lab-on-chip device based
on genomic microarrays of HLA-typing to allow the
early prognosis of diseases such as multiple sclerosis
(MS) and rheumatoid arthritis (RA).
Fig. 1. Gene map of the human leukocyte antigen (HLA) region. The HLA region spans 4 Χ 106 nucleotides on chromosome 6p21.1 to p21.3, with
class II, class III and class I genes located from the centromeric (Cen) to the telomeric (Tel) end. HLA class I molecules restrict CD8+ cytotoxic T
lymphocyte function and mediate immune responses against ‘endogenous’ antigens and virally infected targets, whereas HLA class II molecules are
involved in the presentation of ‘exogenous’ antigens to T helper cells. The HLA class III region contains many genes encoding proteins which are
unrelated to cell-mediated immunity but that nevertheless modulate or regulate immune responses in some way, including tumour necrosis factor
(TNF), heat shock proteins (Hsps) and complement proteins (C2, C4).
A. Genome-Wide Association Study
It is estimated that more than 10 million SNPs exist in the
human genome. SNPs are known to contribute to population
diversity and phenotypic differences between individuals,
and cause predispositions to diseases. Whole-genome
association studies hold the promise of identifying SNPs
associated with a certain phenotype of interest as well as
those that can serve as diagnostic markers. Genes in the
vicinity of SNPs that appear to cause the phenotype could be
qualified as new drug targets for pharmaceutical companies.
To successfully identify candidate SNPs using whole
genome association analysis, the researcher needs to
consider sample size, multiple testing correction, SNP
selection and genotyping quality.
A genome-wide association study (GWA study) - also
known as whole genome association study (WGA study) - is
an examination of genetic variation across the human
genome, designed to identify genetic associations with
observable traits, such as blood pressure or weight, or why
some people get a disease or condition.
These studies require two groups of participants: people
with the disease (cases) and sex- and age-matched
unaffected individuals (controls). After obtaining samples
from an individual, the set of markers such as SNPs are
scanned into computers. The computers survey each
participant's genome for markers of genetic variation.
If genetic variations are more frequent in people with the
disease, the variations are said to be "associated" with the
disease. The associated genetic variations are then
considered pointers to the region of the human genome
where the disease-causing problem resides. Since the entire
genome is analysed for the genetic associations of a
particular disease, this technique allows the genetics of a
disease to be investigated in a non-hypothesis-driven
A genome-wide association study is an approach that
involves rapidly scanning markers across the complete sets
of DNA, or genomes, of many people to find genetic
variations associated with a particular disease. Once new
genetic associations are identified, researchers can use the
information to develop better strategies to detect, treat and
prevent the disease. Such studies are particularly useful in
finding genetic variations that contribute to common,
complex diseases, such as asthma, cancer, diabetes, heart
disease and mental illnesses.
Illumina Infinium II technology allows the fast and
reliable genotyping of up to 1.000.000 SNPs per individual
at once. Whole Genome Genotyping (WGG) is based on the
Sentrix BeadChip platform  on which 13 million DNA-
immobilized Beads are randomly dispersed and assembled
into wells created on a single slide. The location and
identification of each bead is obtained through a decoding
process. On average a redundancy of around 20 beads per
bead type is obtained. The first step of Infinium WGG is a
Whole Genome Amplification of genomic DNA (WGA).
WGA is based on Multiple Displacement Amplification
(MDA) which uses the φ29 DNA polymerase and random
primers to amplify the entire genome generating hundreds of
micrograms of amplified DNA starting from an initial input
of several hundred nanograms (currently, 750 ng). WGA
gives a uniform locus representation that enables the access
to almost all the SNPs in the genome. The amplified DNA is
then fragmented to an average size of around 300 bp using
an enzymatic fragmentation protocol and locus-specifically
hybridized to each individual target on the beads. The
specificity of the hybridization is assured by the use of long
full-length 50-mer oligonucleotides. After hybridization,
each SNP is characterized through a single-base enzymatic
extension assay using two-color labeled nucleotides. After
extension, the labels are visualized by staining with a
sandwich-based immunohistochemistry assay that increases
the overall sensitivity of the assay. BeadChips are imaged
using a two-color confocal laser system with 0.8µm
resolution. The Bead intensities are extracted and genotypes
are calculated with the BeadStudio 3.2.29 software using a
“cluster” file, based on a set of reference samples supplied
B. Selection of the most informative SNPs
Genetic profiles of the MS & RA individuals are constructed
using the data produced by the gene discovery phase using
the iLLumina technology. The analysis of the produced
huge amount of data, leads to genes/SNP patterns which are
responsible for the RA & MS diseases as well as genetic
risks. In order to handle such a dataset there is a need to
select the most informative genes/SNPs for further analysis.
While the target of POCEMON aims to the early prognosis
and diagnosis of the RA & MS autoimmune diseases the
selection of the most informative SNPs is necessitated. The
approach which is used in the POCEMON platform is based
on data mining and genetic algorithms. Artificial intelligent
methods such as global search and weighted decision trees
are employed for the selection of the most significant SNPs.
The SNPs selection approach is applied to datasets to
provide the early prognosis and diagnosis of the RA & MS
autoimmune disorders. It is expected that the number of the
features is significantly reduced while the quality of
knowledge is enhanced. The selection approach is going to
provide the most significant SNPs.
Data mining methods support all the processes of
discovering interesting and previously unknown patterns in
the genotyped MS & RA data sets. A main advantage of the
data mining concerns the study of an individual rather than
the population, providing routes for personalization in early
prognosis and diagnosis. The goal of the selection is to
identify the minimum set of non-redundant features (e.g.,
SNPs, genes) which are useful in the classification phase.
C. LOC platform development and integration
The integration of LOC technology with PDA - using
microelectronics - is going to provide new diagnostic tools
at the primary care level (Fig. 2). The implementation and
the development of the portable POCEMON diagnostic
LOC is achieved by following a modular approach of the
various sub-parts of the integrated platform. The different
components are developed according to the following steps:
• Identification of the major relationship between HLA
alleles and the specific multiple sclerosis and
rheumatoid arthritis autoimmune diseases.
• Design of the HLA microarray probes based on the
technology of oligonucleotide
• Implementation of the HLA genotyping microarrays on
• Design of the microelectronics LOC with mechanisms
to handle blood and/or saliva samples.
• Incorporation of LOC techniques to automatically
hybridization inside the diagnostic platform using
• Employment of portable and mobile technology in the
diagnostic LOC using PDAs.
• Development of a large data repository (database)
which is hosted on a central laboratory or medical
centre. The PDA device
interchanges data with the database server by tracking
and updating medical records concerning the patient
• Development of the oligonucleotide Microarray
computerized scanning mechanism in the PDA of the
• Implementation of appropriate desktop PDA software
for signal intensity analysis of the microarray.
• Extraction of HLA probe’s results in the PDA screen.
In this phase the LOC platform provides a preliminary
diagnosis for autoimmune diseases.
• Wireless communication with centralized database to
provide detailed diagnostic analysis of scanned
microarray probes using comprehensive and intelligent
• Integration of all of the above in the final diagnostic
product, an autonomous portable diagnostic LOC
D. Development of the Diagnostic HLA Lab-on-Chip
Starting from the state-of-the-art in literature a suitable
configuration for the LOC is studied in order to provide the
handling of DNA samples, PCR amplification and HLA
typing with a labelfree approach based on a microcantilever
array of appropriate size functionalized with oligonucleotide
probes (Fig. 3). The microfluidics includes sample ports,
channels for sample delivering and reservoirs with
temperature control for PCR amplification. The selected
technology for the realisation of the LOC fluidics involves
bulk micromachining techniques (wet and dry etching steps)
and wafer-to-wafer bonding to create channels and
reservoirs, deposition of thin film metallization to implement
heaters and thermometers for the thermal control of the PCR
reactor. The label-free detector is realised with a
microcantilever array working in the bending mode (i.e. in
the stress-detection mode) which is more suitable for
operating in liquid phase. The beam surface is provided with
a gold electrode in order to address and bond the specific
DNA probes based on thiol chemistry. The signal
conditioning electronics, including a switching matrix,
signal amplification and ADC is integrated on-chip with a
CMOS technology. The technologies involved deals with
both ad hoc Silicon-On-Insulator (SOI) approaches for the
realisation of thin suspended Silicon beams with tight
thickness control and the implementation of thin film
cantilever compatible with CMOS processes. In the
framework of this system implementation, the design of
single system modules (microfluidic module for sample
handling and PCR amplification, detector module and signal
conditioning electronic module) is studied as well as a
technological approach suitable for the realisation of a single
module and system integration. At first, each module is
Fig. 2. The integration of all POCEMON parts via wireless communication
developed as stand-alone device for validation of the
selected technological processes and preliminary efficiency
testing. In a second step a definitive design is realized in
order to implement the final modules in a monolithic chip.
The signal conditioning interface and device specifications
will be addressed to the development on a PDA platform of
the software tools needed to control the LOC operation and
data acquisition and analysis.
Summarizing, the approach aimed to develop a diagnostic
HLA Lab-on-Chip is based on:
• The development of the microfluidics components for
DNA sample handling, PCR amplification and
• The development of detector modules based on
cantilevers array for preliminary studies of the
sensitivity to specific DNA chains.
• The development of optimized detector modules based
on a high density cantilevers array on CMOS
technology. Related to the ultra-high expected
sensitivity of the sensors, the PCR amplification
module could be substituted with a less sophisticated
• The development of an integrated control unit
(electronic circuit) for sensor read out and signals
The integration of LOC technology with genomic
microarrays of HLA-typing is going to provide new
diagnostic tools at the primary care level. Also, the most
informative selected SNPs are used to design and produce
new microarray chips containing the appropriate probes to
diagnose the RA & MS autoimmune diseases. The
application of the algorithms and methods to datasets
produced by other similar genome wide association studies
is going to enhance the scientific and research community
with innovative tools for the early prognosis and diagnosis
of almost all autoimmune disorders.
 P. Yager, T. Edwards, E. Fu, K. Helton, K. Nelson, M. R. Tam and B.
H. Weigl, “Microfluidic diagnostic technologies for global public
health,” Nature, vol. 442, no 7101, pp. 412–418, 2006.
 P. Parham, “The immune system,” New York: Garland Science, 344,
 G. Rosati, “The prevalence of multiple sclerosis in the world: an
update,” Neurol. Sci., vol. 22, no 2, pp. 117–39, 2001.
 V. Majithia and S.A. Geraci, “Rheumatoid arthritis: diagnosis and
management,” Am. J. Med. vol. 120, no 11, pp. 936–9, 2007.
 5. F. Zhang, S.W. Hu, J. Huang, H. Wang, W. Zhang, Y.Y. Geng, et
al., “Oligonucleotide microarray for HLA-DRB1 genotyping:
preparation and clinical evaluation,” Tissue Antigens, vol. 65, pp.
 C. Consolandi, E. Busti, C. Pera, L. Delfino, G.B. Ferrara, R. Bordoni,
B. Castiglioni, L. Rossi Bernardi, C. Battaglia and G. De Bellis,
“Detection of HLA polymorphisms by ligase detection reaction and a
universal array format: a pilot study for low resolution genotyping,”
Hum Immunol. vol. 64, no. 1, pp. 168-78, 2003.
 N. Dhiman, R.G. Bonilla, R.M. Jacobson, D. O'Kane and A. Poland,
“Differential HLA gene expression in measles vaccine seropositive
and seronegative subjects: A pilot study,” Scandinavian Journal of
Infectious Diseases, vol. 35, pp. 332-6, 2003.
 L.M. Rimsza, R.A. Roberts, T.P. Miller, J.M. Unger, M.LeBlanc,
R.M. Braziel, D.D. Weisenberger, W.C. Chan, H.K. Muller-
Hermelink, E.S. Jaffe, R.D. Gascoyne, E. Campo, D.A. Fuchs, C.M.
Spier, R.I. Fisher, J. Delabie, A. Rosenwald, L.M. Staudt and T.M.
Grogan, “ Loss of MHC class II gene and protein expression in diffuse
large B-cell lymphoma is
immunosurveillance and poor patient survival regardless of other
prognostic factors: a follow-up study from the Leukemia,” Blood, 103
vol. 11, pp. 4251-8, 2004.
 S. Wang, “Development of microarray and its applications,” Progress
in Biotechnology, vol 19, pp. 45, 1999.
 J. Fritz, M.K. Baller, H.P. Lang, H. Rothuizen, P. Vettiger, E. Meyer,
H.-J. Güntherodt, Ch. Gerber, and J.K. Gimzewski, “Translating
Biomolecular Recognition into Nanomechanics,” Science, vol. 288,
pp. 316, 2000.
 K.M. Hansen, H.F. Ji, G.H. Wu, R. Datar, R. Cote, A. Majumdar, and
T. Thundat, “Cantilever-based optical deflection assay for
discrimination of DNA single-nucleotide mismatches,” Anal. Chem.,
vol. 73, no. 7, pp. 1567–1571, 2001.
 J. Verd, G. Abadal, J. Teva, M.V. Gaudó, A. Uranga, X. Borrisé, F.
Campabadal, J. Esteve, E.F. Costa, F. Pérez-Murano, Z. J. Davis, E.
Forsén, A. Boisen, and N. Barniol, “Design, Fabrication, and
Characterization of a Submicroelectromechanical Resonator With
Monolithically Integrated CMOS Readout Circuit,” Journal of
Microelectromechanical Systems, vol. 14, pp. 508, 2005.
 X. Yu, Y. Tang, H. Zhang, T. Li, and W. Wang, “Design of High-
Sensitivity Cantilever and Its Monolithic Integration With CMOS
Circuits,” IEEE Sensors Journal, vol. 7, pp. 489, 2007.
 F.J. Steemers, K.L. Gunderson, “Illumina, Inc.”, Pharmacogenomics,
vol. 6(7), pp. 777-782, 2005.
related to decreased tumor
Fig. 3. Example of a Lab-on-Chip device. Obtained from Burns, et al., Science, vol 282, pp. 484-487, 1998 .
 K.L. Gunderson, F.J. Steemers, G. Lee, L.G. Mendoza, and M.S.
Chee, “A genome-wide scalable SNP genotyping assay using
microarray technology,” Nat.Genet. vol. 37, pp. 549−554, 2005.
 M.A. Burns, B.N. Johnson, S.N. Brahmasandra, K. Handique, J.R.
Webster, M. Krishnan, T.S. Sammarco, P.M. Man, D. Jones, D.
Heldsinger, C.H. Mastrangelo, D.T. Burke, “An Integrated Nanoliter
DNA Analysis Device,” Science, vol. 282 (5388), pp. 484 – 487,