Katholieke Universiteit Leuven, Belgium
KU Leuven, founded in 1425, is a comprehensive university with 15 faculties organised in 3 groups: Humanities & Social Sciences; Science, Technology & Engineering; and Biomedical Sciences. It is Belgium's largest university with about 40.000 students, offering a wide variety of academic programmes in Dutch and English, nurtured by high quality interdisciplinary research. KU Leuven accommodates about 6.600 researchers and has a research budget of about 347 million euro (2010). It is ranked 5th university in FP7 with up to now more than 350 accepted projects representing a budget of more than 135 Meuro (4th interim FP7 report, August 2011). Research directly related to ICT accounts for over 20% of these projects, biomedical research for nearly 1/3.
From an ITFoM perspective following research groups have been active in roadmapping activities so far:
Department of Electrical Engineering (ESAT) – SCD/SISTA/BioI & SCD/COSIC
SCD's main research objective is to design and build advanced methods for crucial problems in information processing. SCD has build up world recognised expertise in bioinformatics, biomedical data processing, signal (audio, communications) processing, cryptography, embedded systems, data mining, neural networks, identification, control, ... These topics are bundled into three research units: COSIC, SISTA and DocArch of which the first two are relevant for ITFoM.
The Bioinformatics (BioI) research group of SCD/SISTA focuses on the development of computational methods to tackle two important scientific challenges in Systems Biology: (1) System Biomedicine, where it develops probabilistic and statistical methods for the analysis, integration, and mining of high-throughput and clinical data (including through visual analytics), to contribute to the identification of key genes and proteins for the diagnosis, prognosis, or targeting of disease. And (2) Top down Systems Biology, focusing on the reconstruction of networks from high throughput molecular data. The group represents 27 researchers among which there are 3 professors.
The SCD/COSIC group conducts basic, applied and contract research focusing on Computer Security and Industrial Cryptography. The research concentrates on the design, evaluation, and implementation of cryptographic algorithms and protocols, on the development of security architectures for information and communication systems, on the development of security mechanisms for embedded systems and on Privacy enhancing technologies. The group has about 58 researchers among which there are 4 professors.
Department of Computer Sciences – DTAI/ML
The DTAI/ML group focuses on machine learning and data mining research involving structured data, symbolic, logical and probabilistic representations & background knowledge and applies its techniques to challenging domains in the life sciences and action- and activity learning. It is one of the largest European labs in this area. It currently consists of 41 researchers among which there are five faculty members.
Department of Electrical Engineering (ESAT): SCD/COSIC, SCD/SISTA
SCD is a division of the department of Electrical Engineering (ESAT) of the Faculty of Engineering of the K.U.Leuven and is headed by Prof. Joos Vandewalle. SCD's major research objective is to design and build advanced methods for crucial problems in information processing. It builds on the enormous growth in computer power, communication bandwidth and available data and the various needs in society for effective use of these opportunities. SCD has a number of mutually reinforcing groups that build expertise from the more conceptual and mathematical level downstream until the validation with our partners in practical applications in the field. A major driving force of the research is the diverse use of generic information processing methods based on applied mathematics like linear and multilinear algebra, statistics, discrete mathematics, differential geometry, and optimization. SCD has built up world recognized expertise in bioinformatics, biomedical data processing, signal (audio, communications) processing, cryptography, embedded systems, data mining, neural networks, identification, control, ... These topics are bundled into 3 research units: SISTA, COSIC and DocArch
COSIC was established in 1978 as part of the Dept. of Electrical Engineering at KUL. COSIC has started immediately conducting basic, applied and contract research focusing on Computer Security and Industrial Cryptography.
SISTA (prof. Bart De Moor, Yves Moreau and Jan Aerts active in bioinformatics) has as research objective to translate concepts of data mining, machine learning and linear algebra into algorithmic implementations, focussing on bio- and medical informatics, biomedical data processing and digital signal processing. The methodology is applied in several fields such as eHealth, personal genomics, modeling for improved cancer diagnosis, hearing aids, brain monitoring etc.
Department of Computer Science: DTAI/ML
The Machine Learning group in K.U.Leuven's Department of Computer Science is one of the largest European labs in this area. It currently consists of five faculty members, a research expert, approximately 10 post-docs and approximately 25 Ph.D students. The group is very internationally oriented and has around 20 non-Belgian members.
COSIC has played an influential role at an international level in the development of cryptography as a scientific and engineering discipline. The Advanced Encryption Standard (AES), selected by the US government after an open competition, was designed by COSIC members; the hash function RIPEMD-160 is a joint design of COSIC and BSI (DE). In the final eSTREAM portfolio, 2 of the 8 ciphers have been designed by COSIC members. Half of the 14 Round-2 candidates in NIST's SHA-3 competition have a designer who has been or is a member of COSIC. During the past 15 years, COSIC has participated in 30 European research projects, and has acted as the coordinator of the FP5 Strategic Roadmap for Crypto (STORK); the FP5 New European Schemes for Signatures, Integrity, and Encryption (NESSIE); the FP6 ECRYPT Network of Excellence and its successor FP7 ECRYPT-II; and the FP7 Integrated Project TAS3 (Trusted Architecture for Securely Shared Services); and the 18-month EU study MODINIS Lot 3. Short- and the medium-term contract research were carried out for companies like Banksys, Hitachi, Mastercard, etc.
COSIC gained thorough experience in privacy-enhancing technologies, identity management, and the design and analysis of cryptographic algorithms, protocols and architectures, through a long history of participation in collaborative projects (e.g., PRIME, PrimeLife, FIDIS, MODINIS, TAS3). COSIC is a member of IBBT (Interdisciplinary Center for Broadband Research).
SCD/SISTA (Yves Moreau, Bart De Moor, Jan Aerts) has developed strong expertise in applications of probabilistic graphical models (Bayesian networks, biclustering) and kernel methods in computational biology (Thijs et al., 2001; Sheng et al., 2003; Gevaert et al., 2006; Antal et al., 2004; Yu et al., 2010). It has extensive expertise in the analysis and visualization of next-generation DNA sequencing data for the discovery of causative variants in rare diseases and the visualization of complex data sets in the search for structural variation in the human genome. It is a leading team on the topic of genomic data fusion for candidate gene prioritization. Its platform Endeavour (Aerts et al., 2006, De Bie et al., 2007) is currently the most extensively biologically validated platform for candidate disease gene prioritization in rare genetic disorders. Through a strong collaboration with the Center for Human Genetics of the University of Leuven, it has developed computational solutions for arrayCGH analysis of rare genetic disorders. These solutions have been transferred to Cartagenia, a spin-off of the K.U.Leuven that develops IT solutions for routine clinical genetic diagnosis (http://www.cartagenia.com). Current research in this area focuses on addressing similar questions with NGS, in particular exome sequencing. SCD/SISTA is part of the IBBT Future Health Department. IBBT is the Interdisciplinary Research Institute for Broad Band Technology, one of the four strategic research centers in Flanders (http://www.ibbt.be). IBBT Future Health positions itself between data and decision making by developing ICT-based Health Decision Support Systems to enable better health care. Focus is on three main areas: clinical, patient and policy decision support.
The group actively investigates all types of machine learning and data mining problems and techniques. It primarily focuses on learning from structured data sources (such as relational databases, graphs, trees and sequences), symbolic, logical and relational representations, and the use of knowledge and constraints. The group is particularly active in the subfields of statistical relational learning, probabilistic programming languages, relational learning (e.g., inductive logic programming), graph mining, constraint-based mining, inductive databases and decision tree learning.
The group also emphasizes important applications in the life sciences such as medical informatics, bioinformatics and chemoinformatics. For example, the group is working on applying machine learning and data mining techniques to clinical data. The group has worked on three tasks in this domain. In the first task, the goal is to use a patient's clinical history to predict, at prescription time, whether the patient will have an adverse reaction to a medicine. The second task involves learning a predictive model from structured reports about mammograms. The goal is to predictive whether an abnormality on a mammogram is benign or malignant. The final task involves analyzing data about patients in the intensive care unit. We have tackled a variety of predictive tasks in this domain.
Besides KULeuven could bring in additional ICT related expertise on:
- Human-information interaction.
- Distributed and secure software and middleware.
- Legal aspects of ICT.
- (Molecular) imaging.
Within the Biomedical Faculties and University Hospitals we also have additional medical experience available on e.g. hereditary diseases, cancer, neurological disorders.
Profile of staff members
Following staff members are currently involved in ITFoM activities:
Prof. Bart Preneel is a full professor at the KU Leuven. He is Vice president of the International Association for Cryptologic Research (IACR) and a Member of the Editorial Board of J. Cryptology, IEEE Trans. Information and Forensics Security and of the ACM Trans. on Information Security. He is also president of L-SEC vzw (Leuven Security Excellence Consortium).In 2003 he received the European Information Security Award in the area of academic research. His main interests are in cryptography, network security and wireless communications.
Prof. Dr. Ir. Claudia Diaz is an assistant professor in Privacy Technologies. Her research is broadly focused on the topic of Privacy Enhancing Technologies. She is a member of the advisory board of the Privacy Enhancing Technologies Symposium (PETS) and the scientific committee of CPDP. She has organised multiple workshops (among which PETS 2008) and served as a program chair for ESORICS'11 and chair of the PET award in 2011 and 2012.
Prof. Yves Moreau's research focuses on computational methods for systems biology and computational infrastructure necessary to bring new technologies to routine clinical application in medical genetics. He is co-developer of Endeavour, one of the main platforms for the prioritization of candidate disease genes and coordinates the KULeuven center of excellence SymBioSys "from variome to phenome". He is a director of the Master in Bioinformatics program and cofounder together with Bart De Moor of Cartagenia, a spin-off of KULeuven distributing IT solutions for clinical genetic diagnosis.
Prof. Bart De Moor's bioinformatics research focuses on data mining, text mining, medical diagnosis systems, and Bayesian learning and decision. He is currently vice rector International Policy of the KU Leuven and scientific director of the IBBT-KU Leuven Future Health Department. He is in the board of the Flemish Interuniversity Institute for Biotechnology (VIB) and is chairman of the Industrial Research Fund of the KU Leuven and of the Hercules foundation for heavy equipment funding in Flanders and responsible for the Flemish High Performance Computing Centre.
Prof. Jan Aerts focuses on the analysis and visualization of next-generation DNA sequencing data for the discovery of causative variants in rare diseases and the visualization of complex data sets in the search for structural variation in the human genome. In addition, he investigates methods to process the massive amounts of data generated within this field. He also has an interest in preventive genomics.
Prof.Jesse Davis holds a tenure-track assistant professorship on data mining for personalized medicine at the KU Leuven. His main research interests are on machine learning, data mining and artificial intelligence and the application of these techniques to important medical problems. He serves on the editorial board for the Machine Learning Journal.
Being a professor at the KU Leuven, Prof. Hendrik Blockeel's primary research interests include machine learning, data mining, artificial intelligence and bioinformatics and the biological and medical applications of these techniques. He serves on the editorial board for Applied Artificial Intelligence, Machine Learning Journal and Intelligent Data Analysis. He has organized multiple workshops and has been the program chair for the European Conference on Machine Learning.
Five recent publications relevant to the project
1) Drac: An Architecture for Anonymous Low-Volume Communications. George Danezis, Claudia Diaz, Carmela Troncoso, and Ben Laurie. In Proceedings of the 10th Privacy Enhancing Technologies Symposium (PETS 2010), M. Atallah and N. Hopper (Eds), Springer LNCS 6205, pp. 202-219, 2010.
2) The wisdom of Crowds: attacks and optimal constructions. George Danezis, Claudia Diaz, Emilia Käsper, and Carmela Troncoso. In Procceedings of the 14th European Symposium on Research in Computer Security (ESORICS 2009), M. Backes and P. Ning (Eds), Springer LNCS 5789, pp. 406-423, 2009.
3) Origins and functional impact of copy number variation in the human genome., Conrad DF, Pinto D, Redon R, Feuk L, Gokcumen O, Zhang Y, Aerts J, Andrews TD, Barnes C, Campbell P, Fitzgerald T, Hu M, Ihm CH, Kristiansson K, Macarthur DG, Macdonald JR, Onyiah I, Pang AW, Robson S, Stirrups K, Valsesia A, Walter K, Wei J; Wellcome Trust Case Control Consortium, Tyler-Smith C, Carter NP, Lee C, Scherer SW, Hurles ME. Nature. 2010 Apr 1;464(7289):704-12. Epub 2009 Oct 7.
4) Bioinformatics software for the Ruby programming language., Goto N, Prins P, Nakao M, Bonnal R, Aerts J, Katayama T., BioRuby., Bioinformatics, 2010 Oct 15;26(20):2617-9. Epub 2010 Aug 25.
5) A Probabilistic Computer Model Developed from Clinical Data in the National Mammography Database Format to Classify Mammography Findings., E. Burnside, J. Davis, J. Chhatwal, O. Alagoz, M. Lindstrom, B. Geller, B. Littenberg, K. Shaffer, C. Kahn, D. Page, Radiology, volume 251, number 3, pages 663-672, 2009.