Privacy - Preserving Data Exchange and Aggregation in Healthcare

Medical data are often scattered among multiple clinics, hospitals, insurance companies, pharmacies, and research institutions that store and process personal healthcare information. The use of information and communication technologies for health (eHealth) provides us with the means to share healthcare data between authorized parties in an efficient manner. In this thesis, we address some of the challenges of implementing eHealth in practice: to achieve interoperability between data sources, and to ensure privacy for patients. Achieving both of these guarantees is our goal but they seem conflictual, hence the challenge. Once interoperability is achieved and a patientâs data are shared, it becomes evenmore difficult to ensure the patientâs privacy i.e., to provide to a patient control over his data and to guarantee the data anonymity in medical research. We address the aforementioned challenges by studying requirements from medical and legal perspectives, and by developing algorithms and frameworks to support privacy-preserving dynamic data-sharing, exchange, and aggregation from multiple data sources.

In the first part of the thesis, we address certain privacy challenges. We present a framework based on the blockchain technology for ensuring traceability and accountability when sharing, exchanging, and aggregating medical data. Our framework ensures privacy, security, availability, and fine-grained access control over highly sensitive patient-data. We also analyze the potential of applying blockchain technology in different eHealth settings: primary care, medical-data research, and connected health. Our second contribution is a framework for privacy-preserving data aggregation: an algorithm for constructing the anonymized database and a protocol that improves the utility of the anonymized database as the database grows.

In the second part of the thesis, we focus on achieving interoperability. We design an interface specification that defines communication protocols andmessages supporting integration of a new software tool in clinical practice. Then, we develop a multi-agent system (MAS) for the dynamic aggregation of the data collected and generated by this software tool for the purpose of clinical research. This MAS takes into account the objectives of the research study, the availability of data, and could employ our proposed algorithm for privacy-preserving data aggregation. The negotiation protocol in the framework of theMAS achieves a precise definition of database characteristics, such as schema, content, and privacy parameters, therefore increasing the efficiency of data collection for medical research and ensuring the privacy of patients.

A Multiagent System for Dynamic Data Aggregation in Medical Research

Karl Aberer, Alevtina Dubovitskaya, Michael Ignaz Schumacher

The collection of medical data for research purposes is a challenging and long-lasting process. In an effort to accelerate and facilitate this process we propose a new framework for dynamic aggregation of medical data from distributed sources. We use agent-based coordination between medical and research institutions. Our system employs principles of peer-to-peer network organization and coordination models to search over already constructed distributed databases and to identify the potential contributors when a new database has to be built. Our framework takes into account both the requirements of a research study and current data availability. This leads to better definition of database characteristics such as schema, content, and privacy parameters. We show that this approach enables a more efficient way to collect data for medical research.

Hindawi Publishing Corporation2016

Sharing of probabilistically correlated data in peer-to-peer networks

Roman Schmidt

The impact of Peer-to-Peer (P2P) networks on the Internet landscape is undisputed. It has led to a series of new applications, e.g., as part of the so-called Web 2.0. The shift from the classical client-server based paradigm of the Internet, with a clear distinction between information providers and consumers, towards consumers sharing information among each other led to the rise of the P2P paradigm. The distributed setting enables users to share their content autonomously and locally, i.e, information remains at computers at the edge of the Internet and is not gathered and organized at central servers. Structured overlay networks were designed to organize the huge quantity of data shared in P2P networks by building a global, though distributed, index of shared information. Whereas the initial aim of these systems was to provide efficient lookup operations for single keyword operations, the need for more complex operations emerged very quickly. Peer Data Management Systems (PDMS) is one such example of application that enables data integration and complex query processing similar to (distributed) database systems on top of structured overlays. Complex query operators usually require joint access to multiple data entries whereas single key lookups usually only affect a single data entry. The partitioning of the distributed index of standard structured overlays is optimized towards single key lookups and joint data access as required by PDMS was neglected so far. (Distributed) databases have already shown that (index-)data organization supporting correlated data access is necessary and crucial for efficient processing, as network usage is minimized. We aim at applying this insight to structured overlays by clustering correlated data frequently accessed jointly by applications, including PDMS but also other types of applications. Data correlations can be derived from different sources, data properties, processing properties, users and applications. We study and present solutions for three different types of correlations in the context of different applications: (i) range queries where exploiting the order relationship of data is a fundamental basis for any database-like system; (ii) distributed probabilistic inference where exploiting user-defined complex data correlations gains importance through the Web 2.0; (iii) multi-term queries where exploiting data correlations derived from data access statistics enables simple but powerful keyword search to users. Our approach exploits properties of structured overlays, such as order-preserving hashing, to realize efficient range query processing. We further introduce a distributed clustering algorithm based on the spring relaxation technique to cluster strongly correlated data entries at one node respectively in its proximity. Joint data access is thus performed on a single or few nodes to reduce network bandwidth consumption and therefore to increase system performance. Our approaches are realized on top of the structured overlay network P-Grid although they are generic enough to be applied to other P2P networks with similar properties. This thesis presents details about the Java implementation of P-Grid, its architectural design and its evaluation on PlanetLab, today's standard testbed for P2P systems. The implementation of P-Grid in a Java application finally enabled us to build a PDMS system on top relying on P-Grid's efficient query processing. We present the UniStore system, a distributed data management system aiming at public data management and support for database-like query operators on heterogeneous data, and its evaluation on PlanetLab.

EPFL2008

Open-Ended Learning of Visual and Multi-Modal Patterns

Jie Luo

A common trend in machine learning and pattern classification research is the exploitation of massive amounts of information in order to achieve an increase in performance. In particular, learning from huge collections of data obtained from the web, and using multiple features generated from different sources, have led to significantly boost of performance on problems that have been considered very hard for several years. In this thesis, we present two ways of using these information to build learning systems with robust performance and some degrees of autonomy. These ways are Cue Integration and Cue Exploitation, and constitute the two building blocks of this thesis. In the first block, we introduce several algorithms to answer the research question on how to integrate optimally multiple features. We first present a simple online learning framework which is a wrapper algorithm based on the high-level integration approach in the cue integration literature. It can be implemented with existing online learning algorithms, and preserves the theoretical properties of the algorithms being used. We then extend the Multiple Kernel Learning (MKL) framework, where each feature is converted into a kernel and the system learns the cue integration classifier by solving a joint optimization problem. To make the problem practical, We have designed two new regularization functions making it possible to optimize the problem efficiently. This results in the first online method for MKL. We also show two algorithms to solve the batch problem of MKL. Both of them have a guaranteed convergence rate. These approaches achieve state-of-the-art performance on several standard benchmark datasets, and are order of magnitude faster than other MKL solvers. In the second block, We present two examples on how to exploit information between different sources, in order to reduce the effort of labeling a large amount of training data. The first example is an algorithm to learn from partially annotated data, where each data point is tagged with a few possible labels. We show that it is possible to train a face classification system from data gathered from Internet, without any human labeling, but generating in an automatic way possible lists of labels from the captions of the images. Another example is under the transfer learning setting. The system uses existing models from potentially correlated tasks as experts, and transfers their outputs over the new incoming samples, of a new learning task where very few labeled data are available, to boost the performance.