Knowledge vs privacy: The healthcare dilemmaEscrito por Edgar Batista el 17/11/2020 a las 11:09:09
(Phd Candidate. Universidad Rivira i Virgili)
Information and communication technologies have irreversibly changed our lives. As a result of the generalised use of computers, smartphones, tablets, smartwatches and gadgets alike, many services are already provided through them. This seamless connection between the physical and the digital worlds has paved the way towards the Industry 4.0, aiming to interconnect multiple cyber-physical systems, with the capability to derive non-apparent knowledge and to support autonomous decisions based on vast amounts of data. The healthcare industry, one of the world’s largest and fastest-growing industries, puts many efforts in adopting the latest technologies into medical daily practice. Technology is key to telemedicine, and it enables the delivery of digital health and care services to patients. Moreover, it helps to increase diagnosis accuracy, shorten waiting times, and decrease medical expenditures. All in all, the improvement and optimisation of medical treatments and care protocols is aligned with a better patients’ quality of life.
In parallel, the miniaturisation of electronic circuits has facilitated the integration of a large variety of sensors in many electronic devices with high sensitivity at lower costs and energy consumption. The ubiquity and communication capabilities of these devices, such as wearables and the Internet of Things, have enabled the gathering of large volumes of user-centric data, especially related to health status, such as heart rate, respiratory rate and even electrocardiography. From the patients’ perspective, this medical data can be augmented with mobility data, e.g., GPS data, and contextual data, e.g., using the sensing infrastructure of context-aware environments such as smart cities, smart homes or smart hospitals. The meaningful exploitation of these data has opened the door to the so-called smart health, a trendy healthcare paradigm, introduced by Solanas et al. , focused on the provision of added-value personalised healthcare services to patients.
Aligned with the smart health objectives, today’s healthcare institutions struggle to become more effective, cost-efficient, flexible and sustainable. Among others, the proper management of business processes is crucial to understand and specify how organisations operate. Within the healthcare sector, a business process could refer to the procedure followed to take care of patients in an emergency department, or the procedure followed to treat a certain disease. It has been proven that efficient business processes contribute to improve the overall performance of organisations. Unfortunately, the ideal/theoretical execution of processes might not always conform with reality in practice. This is particularly common in the healthcare domain due to the high complexity, dynamism, and multi-disciplinary nature of healthcare processes. To support the analysis of business processes, the process mining discipline emerged . This research discipline aims to discover, monitor and improve real processes by extracting knowledge from event data recorded by healthcare information systems. The intelligent analysis of event data (representing the actual execution of processes) can provide valuable insights into process executions, record processes deviations, recommend measures on how to improve processes, identify bottlenecks, monitor resources utilisation, and predict times and costs, among others. The integration of process mining tools into Business Intelligence solutions will be paramount in the years to come, enabling healthcare managers to make better decisions based on the knowledge gathered from process mining related analyses.
The management of healthcare data, from its storage to its analysis, arises a number of serious concerns due to its highly confidential nature. In particular, privacy is a fundamental human right in the spotlight as a result of the numerous data leakages and worldwide scandals. To avert the disclosure of personal data to unauthorised parties, applicating privacy-preserving techniques is mandatory. Surprisingly enough, these techniques were barely studied within the process mining context before the enforcement of the General Data Protection Regulation (GDPR) in 2018, the biggest legislation step towards harmonising the data protection laws across the European Union. Consequently, research on privacy-preserving process mining, still in an embryonic stage, aims to apply the classical privacy protection techniques into process mining and develop privacy-enhancing solutions that guarantee people’s privacy during process mining analyses. However, applying privacy-preserving solutions lowers the quality of the process mining results, and therefore the knowledge that can be inferred from them. This trade-off between knowledge and privacy is unavoidable.
In a nutshell, organisations consider users’ data one of their most valuable assets, because of the priceless benefits that they may obtain from their proper analysis (healthcare organisations are not an exception). Hence, they invest lots of time, money and efforts into getting advanced knowledge and becoming more efficient and competitive. In this context, it would not be out of place to guarantee users’ privacy by introducing privacy-preserving techniques and pursue the best knowledge-privacy equilibrium, which represents the main focus of our study.
Acknowledgments: This research was funded by the Government of Catalonia under the Industrial Doctorate Plan (grant number 2017-DI-002). Special thanks to Dr. Agusti Solanas, project supervisor at Universitat Rovira i Virgili, and to Mr. Diego Donoso, project supervisor at SIMPPLE, S.L.
 A. Solanas, C. Patsakis, M. Conti, I. S. Vlachos, V. Ramos, F. Falcone, O. Postolache, P. A. Pérez-Martínez, R. Di Pietro, D. N. Perrea and A. Martínez-Ballesté, Smart health: A context-aware health paradigm within smart cities. IEEE Communications Magazine 52(8), pp. 74-81 (2014)
 W. M. P. van der Aalst, Process Mining: Discovery, Conformance and Enhancement of Business Processes. Springer (2011)