Towards an Architecture to Guarantee Both Data Privacy and Utility in the First Phases of Digital Clinical Trials

Abstract

In the era of the Internet of Things (IoT), drug developers can potentially access a wealth of real-world, participant-generated data that enable better insights and streamlined clinical trial processes. Protection of confidential data is of primary interest when it comes to health data, as medical condition influences daily, professional, and social life. Current approaches in digital trials entail that private user data are provisioned to the trial investigator that is considered a trusted party. The aim of this paper is to present the technical requirements and the research challenges to secure the flow and control of personal data and to protect the interests of all the involved parties during the first phases of a clinical trial, namely the characterization of the potential patients and their possible recruitment. The proposed architecture will let the individuals keep their data private during these phases while providing a useful sketch of their data to the investigator. Proof-of-concept implementations are evaluated in terms of performances achieved in real-world environments.

Development of Novel, Value-Based, Digital Endpoints for Clinical Trials: A Structured Approach Toward Fit-for-Purpose Validation

“…The flow of data should be automated, requiring as little manual input as possible to reduce data (entry) errors. The flow should be consistent and allow for the subjects' privacy by design-for example, via encrypted transmission of data (Angeletti et al, 2018). Furthermore, the data flow must be part of the validation and comply with the necessary FDA regulations regarding audit trails and the storage and processing of source data (U.S. Food and Drug Administration, 2017).…”

Audapret: Towards a Cloud Architecture Supporting Multi-Device Research Studies

“…However, Angeletti et al's solution does not specify the data collection process itself. They argue that the devices used in clinical studies must be certified as clinically useful, but do not propose a solution on how to integrate the devices into their architecture (Angeletti et al, 2018).…”

“…Developing customized systems for individual studies involves a considerable expenditure of time and budget. While wearable devices appeal to participants due to their minimal restriction on freedom of movement, they also raise privacy concerns (Angeletti et al, 2018). Such devices constantly collect data and may create data profiles about the respective participant's health (Paul and Irvine, 2014).…”

Fog-Computing-Based Heartbeat Detection and Arrhythmia Classification Using Machine Learning

“…Moreover, depending on the accuracy of the ECG device, even a short recording may lead to a large volume of data, thus affecting the battery performance of the portable devices as well as that of the accompanying smartphone while storing and transmitting the recording wirelessly [4,5], thus limiting their diagnostic use to 24 h monitoring or even less [6]. Finally, the concept of transmitting the sensor data to cloud services for processing and analysis creates a number of security issues that need to be addresses as they are directly related with the privacy of the users [7][8][9]. Existing portable solutions are totally cloud-centric: all personal data collected are stored on the cloud and in most cases users no longer own the data they produce.…”

“…This approach severely limits the ability of the user to maintain control of personal data. Now, more than ever, there is a need for privacy-preserving applications where users are always in control of their sensitive data [8,10].…”