1,721,079 research outputs found
Privacy in characterizing and recruiting patients for IoHT-aided digital clinical trials
Full text linkNowadays there is a tremendous amount of smart and connected devices that produce data. The so-called IoT is so pervasive that its devices (in particular the ones that we take with us during all the day - wearables, smartphones...) often provide some insights on our lives to third parties. People habitually exchange some of their private data in order to obtain services, discounts and advantages. Sharing personal data is commonly accepted in contexts like social networks but individuals suddenly become more than concerned if a third party is interested in accessing personal health data. The healthcare systems worldwide, however, begun to take advantage of the data produced by eHealth solutions. It is clear that while on one hand the technology proved to be a great ally in the modern medicine and can lead to notable benefits, on the other hand these processes pose serious threats to our privacy. The process of testing, validating and putting on the market a new drug or medical treatment is called clinical trial. These trials are deeply impacted by the technological advancements and greatly benefit from the use of eHealth solutions. The clinical research institutes are the entities in charge of leading the trials and need to access as much health data of the patients as possible. However, at any phase of a clinical trial, the personal information of the participants should be preserved and maintained private as long as possible. During this thesis, we will introduce an architecture that protects the privacy of personal data during the first phases of digital clinical trials (namely the characterization phase and the recruiting phase), allowing potential participants to freely join trials without disclosing their personal health information without a proper reward and/or prior agreement. We will illustrate what is the trusted environment that is the most used approach in eHealth and, later, we will dig into the untrusted environment where the concept of privacy is more challenging to protect while maintaining usability of data. Our architecture maintains the individuals in full control over the flow of their personal health data. Moreover, the architecture allows the clinical research institutes to characterize the population of potentiant users without direct access to their personal data. We validated our architecture with a proof of concept that includes all the involved entities from the low level hardware up to the end application. We designed and realized the hardware capable of sensing, processing and transmitting personal health data in a privacy preserving fashion that requires little to none maintenance
Conceptual Constraints for Data Quality in Data Lakes
Full text linkA data lake is a loosely-structured collection of data at scale built for analysis purposes that is initially fed with almost no requirement of data quality. This approach aims at eliminating any effort before the actual exploitation of data, but the problem is only delayed since robust and defensible data analysis can only be performed after very complex data preparation activities.
In this paper, we address this problem by proposing a novel and general approach to data curation in data lakes based on: (i) the specification of integrity constraints over a conceptual representation of the data lake and (ii) the automatic translation and enforcement of such constraints over the actual data. We discuss the advantages of this idea and the challenges behind its implementation
Heterogeneous Data Translation through XML Conversion
No full textIn this paper, we illustrate an approach to the translation of Web data between heterogeneous formats. This work fits into a larger project whose aim is the development of a tool for the management of data described according to a large variety of models and formats used on the Web and the automatic translation of schemes and instances from one model to another. Data translations operate over XML representations of schemes and instances and rely on a uniform description of models that we call metamodel. The metamodel shows structural diversities and dictates the needed transformations. Complex translations are derived automatically by combining a number of predefined basic procedures. These procedures perform XML transformations and are implemented by means of XML query languages. Practical examples are provided to show the effectiveness of the approach
- …
