Case Studies of Biomedical Informatics Applications

Chairperson: Johan van der Lei
Erasmus University Medical Center
The Netherlands

José Luis Oliveira
Universidade de Aveiro
Aveiro, Portugal
"On the Integration of Biomedical Databases: a Naive Approach"
State of the art methods on bioinformatics include the use of public databases to publish the scientific breakthroughs. These databases provide valuable knowledge for the medical practice. But, given their specificity and heterogeneity, we cannot expect the medical practitioners to include their use in routine investigations. To obtain a real benefic from them, the clinician needs integrated views over the vast amount of knowledge sources, enabling a seamless querying and navigation. Diseasecard (www.diseasecard.org) is a public web portal system that provides a single entry point to access relevant medical and genetic information available in the Internet about rare diseases. By navigating in the entire alphabetic list, or by searching directly a disease name, a gene symbol or an OMIM code, the user can retrieve information about near 2000 rare diseases.

Francesca Incardona
Informa S.r.l.
Rome, Italy
"Multiple Modelling Techniques"
In order to provide an integrated system for clinical management of antiretroviral drug resistance, several modelling techniques are unders study within the EuResist project. The objective is to provide the clinicians with a prediction of response to antiretroviral treatment in HIV patients, thus helping them to choose the best drugs and drug combinations for any given HIV genetic variant. To this end, a large European integrated data set has been created by merging three of the largest existing resistance databases: ARCA, AREVIR and Karolinska. The first results of the prediction engines are encouraging.

Dimitris Kafetzopoulos, George Potamias, Manolis Tsiknakis
FORTH-ICS, Heraklion
Crete, Greece
"Protein Profiling and Diagnosis with Lab-on-Chip and Proteomics Technology: The LOCCANDIA Integrated Clinico-Proteomics Environment"
The human plasma proteome holds the promise of a revolution in disease diagnosis and therapeutic monitoring provided that major technological challenges can be addressed. The plasma protein analysis aims to characterize the proteomic environment of cells and in particular to detect the expression of their disorder. This is in particular highly relevant to study cancer disease. One major breakthrough should come from the utilization of multiprotein disease markers instead of single protein analytes and the detection of all the isoforms of the selected proteins. Gastric cancers as pancreatic cancers are among the most severe diseases in western countries. The only effective therapy is an early resection of the tumors. Those cancers are detected by expensive diagnostic imaging methods at a late stage with a poor prognosis and a high lethality. The most important factor determining the outcome is an early detection of tumors at a resectable stage. It should be based on a routine screening before clinical symptoms arise. Meeting the aforementioned problems and challenges, the LOCCANDIA (Lab – On – Chip based protein profiling for CANcer DIAgnosis) FP6-IST project aims to validate the application of a new integrated lab-on-chip devise (developed by CEA) to plasma protein profiling for early pancreatic cancer diagnosis. Detection of protein panels and protein isoforms is achieved by advanced mass spectrometry approaches. The overall approach aims to deliver and validate an integrated full proteomics analysis chain from blood sample acquisition and preparation to diagnosis, combining bio-, nano- and information-related technologies. In this context, and in order to ease the full analysis chain an Integrated Clinico-Proteomics Environment (ICPE) is designed and build (already in development phase) around three major modules and respective software components and modules for each of these: (i) Proteomic Information management System (PIS) – to enable data access, storage, transformation, and processing of mass-spectrometry data, based on respective information and data standards (e.g., mzXML); the module system handles all significant details about sample processing and the acquired mass spectroscopy analysis raw data, facilitates embedding of plug-in modules for further data processing; (ii) Clinical Information System (CIS) – to store and manage the extended patient information including the outcome of other modalities, in order to allow statistical characterisation of the LOCCANDIA analysis chain, such as sensitivity and specificity assessment; and (iii) Information and data Mediation infrastructure - to facilitate homogenization, integration and sharing of information and data. Furthermore, the LOCCANDIA mediation layer is enriched with established and novel data processing and analysis modules for (a) Preprocessing - for protein quantification including data standardization and noise filtering, along with a profile reconstruction module to recover the protein profile; (b) Protein/peptide identification – for the identification of those proteins/peptides which are not in the initial targeted panel but might be present in the measurement based on the principles; (c) Data mining and knowledge discovery - for data interpretation and clinical decision support aiming to the definition of a range window to discriminate between healthy and non-healthy patient, and the classification of patient; and (d) Visualization – to support a common way to visualize and compare the output of the various MS devices. The current talk will present the essentials of LOCCANDIA and will report on its recent demonstrated achievements.

Bruno G. Loos
University of Amsterdam
Amsterdam, The Netherlands
"Biomedical Informatics in Chronic Infectious and Inflammatory Disease Research: Periodontitis as a Model"
Chronic inflammatory illnesses impose a major drain on society. They are thus priority targets for current biomedical informatics (BMI) research. Periodontitis is a chronic inflammatory disease of the supporting tissues of the teeth. If left untreated, teeth will become mobile and migrate, and will eventually exfoliate. Chronic inflammatory diseases like periodontitis have a complex pathogenesis and a multifactorial etiology, involving complex interactions between multiple genetic loci, infectious agents and environmental (behavioral) factors. A Periodontitis Data Warehouse (PDW) was built including genotypic, phenotypic, infectious and environmental data, as well as phenomics (extent and severity of the disease). Anonimysation and privacy issues have been addressed. A dental imaging analysis (DIA) tool has been constructed for establishing phenomics. The PDW contains currently cross-sectional data for over 800 patients and controls. The preliminary data mining (decision tree learning and association rule mining) has modelled a genetic susceptibility index for periodontitis and showed that genomic information together with microbial data gives the best accuracy in distinguishing patients from controls. Conclusion: The PDW has been structured and allows an integrative research approach, where BMI will contribute to new insights in chronic inflammatory and infectious diseases. Supported by the INFOBIOMED network, 6th R&D Framework, EC (IST 2002 507585).

Inge Bernstein
Hvidovre Hospital
Copenhagen, Denmark
"INFOBIOMED: the Danish Pilot – Genomics and Colon Cancer"
The Danish HNPCC-register was established in 1991 as a public national central database including epidemiological and genomic data on all families with a high risk of colorectal cancer due to the dominantly inherited cancer syndrome (HNPCC) in order to improve their prognosis by identifying high-risk family-members and establish screening. HNPCC-families are diagnosed in many de-partments all over Denmark, and data is today sent as paper copy to the register and here typed into the database. In INFOBIOMED the Danish HNPCC-register is used as a model for electronic ex-change of epidemiological and genomic data on high-risk families between treating departments and the central database. The aim is to obtain knowledge useful for planning and organization of screening in high-risk families and generate tools usable and generic enough to be implemented in other countries and for other oncogenetic diseases, if wanted. The experiences are presented.

George Potamias, Manolis Tsiknakis, Dimitris Kafetzopoulos
FORTH-ICS, Heraklion
Crete, Greece
"Cancer, Clinico-Genomics and Biomedical Informatics: The ACGT Approach"
Recent advances in research methods and technologies have resulted in an explosion of information and knowledge about cancers and their treatment. Exciting new research on the molecular mechanisms that control cell growth and differentiation has resulted in a quantum leap in our understanding of the fundamental nature of tumor genesis and cancer cells and has suggested valuable new approaches to cancer diagnosis, prognosis and treatment. The breadth and depth of available information present an enormous opportunity for improving our ability to reduce mortality from cancer, improve therapies and meet the demanding individualization of care needs. While these opportunities exist, the lack of a common infrastructure has prevented clinical research institutions from being able to mine and analyze disparate data sources. As a result, very few cross-site studies and multi-centric clinical trials are performed and in most cases it isn’t possible to seamlessly integrate multi-level data (from the molecular to the organ and individual levels). Moreover, clinical researchers or molecular biologists often find it hard to exploit each other’s expertise due to the absence of a cooperative environment which enables the sharing of data, resources or tools for comparing results and experiments, and a uniform platform supporting the seamless integration and analysis of disease-related data at all levels. The implementation of discovery driven translational research, however, will not only require co-ordination of basic research activities, facilities and infrastructures, but also the creation of an integrated and multidisciplinary environment with the participation of dedicated teams of clinicians, oncologists, pathologists, epidemiologists, molecular biologists, as well as a variety of disciplines from the domain of information technology and biomedical informatics. In this context, the ACGT: Advancing Clinico-Genomic Trials on Cancer FP6-IST integrated project integrated project (launched on January 2006) aims to contribute to the resolution of the aforementioned problems by developing a semantic grid infrastructure in support of multi-centric, post-genomic clinical trials, and thus enabling for discoveries in the laboratory to be quickly transferred to the clinical management and treatment of patients.In achieving its objectives, ACGT has formulated a coherent, integrated workplan for the design, development, integration and validation of all technologically challenging areas of work. Namely: (I) GRID: delivery of a European Biomedical GRID infrastructure offering seamless mediation services for sharing data and data-processing methods and tools, and advanced security; (II) Integration: semantic, ontology based integration of clinical and genomic/proteomic data - taking into account standard clinical and genomic ontologies and metadata, and enabled by a flexible scientific workflow editing and enactment infrastructure; (III) Knowledge Discovery: delivery of data-mining GRID-enabled services in order to support and improve complex knowledge discovery processes. The ACGT environment and technological platform will be validated in concrete setting of advanced (research) clinical trials on Cancer. Pilot research trials have been selected based on the presence of clear research objectives: (a) The ACGT TOP breast-cancer study: aims to identify biological markers associated with pathological complete response to anthracycline (epirubicin) pre-surgical therapy; (b) The ACGT Wilms' tumor study: aims to reduce therapy for children with low-risk tumors based on the identification of novel molecular, histological and clinical risk factors for stratification of treatment intensity; and (c) The ACGT in-silico modelling and simulation: aims to provide clinicians with a decision support tool able to simulate, the response (tumor growth/shrink) of a solid tumor to therapeutic interventions based on the individual patient’s multi-level data. The current talk will present the essentials of ACGT and will report on its recent demonstrated solutions.

Magi Lluch
MicroArt
Barcelona, Spain
"HealthAgents: Distributed Decision Support System for Brain Tumour Diagnosis and Prognosis using Agent Technology."
HealthAgents is a Specific Targeted Research Project (STREP) funded by the Framework Programme 6 of the European Commission. HealthAgents plans to create a multi-agent distributed Decision Support System, to help in the early diagnosis and prognosis of brain tumours. HealthAgents is not only developing new pattern recognition methods for a distributed classification and analysis of in-vivo magnetic resonance spectroscopy (MRS) and ex-vivo/in-vitro high resolution magic angle spinning nuclear magnetic resonance (HR-MAS) and DNA data, but also defining a method to assess the quality and usability of new cases. HealthAgents is to apply agent technology to securely connect user sites with a distributed database, employing agent negotiation and argumentation mechanisms. Further details of the project can be gathered at http://www.healthagents.net/.

Vassilis Moustakis
FORTH-ICS, Heraklion
Crete, Greece
"Excellence in Biomedical Informatics Development and Implementation"
The presentation integrates biomedical informatics (BMI) and industrial excellence (IE). The demonstration field of the integration is Drug Discovery. We review the drug life-cycle, place it in context with product life-cycle, and insert BMI as an enable technology. We use few indicative case studies to motivate the need to harness BMI potential in drug discovery. The presentation closes with an overview of IE leading to a discussion on how IE models specific to BMI endeavors can be developed. The motivation for the presentation lies in that implementation of BMI often comes with high cost and one should know where effort is expected to wind-up. Thus there is a need to develop self-assessment models to enable continuous monitoring and steering of effort on course.

close window