Home / Blog / Decentralised Epidemic Alert Systems leveraging on Digital Twins

Decentralised Epidemic Alert Systems leveraging on Digital Twins

Schematic showing the basic principles of decentralized epidemic alert system (DEAS). Decentralized systems run on multiple computers and leverage distributed ledger technologies (DLT), such as blockchains. Image credit: Juuso Autiosalo

I have received this post from Juuso who is working on Digital Twins as Doctoral Candidate at Aalto University and I am very pleased to share it on the FDC blog. It presents several interesting ideas worth of discussion and hopefully of implementation.

By Juuso Autiosalo

Roberto Saracco has documented the use of personal digital twins for epidemics in several previous blog posts. Some of the contents of this post rely on his ideas and they are recommended background reading material. That said, I did have a similar idea on Sunday March 15 2020 after reading the enlightening COVID-19 article by Tomas Pueyo. My thoughts wondered to privacy which the Western societies so enthusiastically protect and the previous experience with DLTs enabled me to think of a solution that leverages this infant technology group.

To convey the idea of decentralized epidemic alert system (DEAS), basic understanding of DLTs is described. If you know DLTs well already, you can jump over the next four paragraphs.

Distributed ledger technologies (DLT), such as blockchains, offer a new kind of decentralized solution for preserving transparency and privacy in internet-based systems. Transparency is enabled by the public nature of transactions in a distributed ledger (an implementation of DLT), whereas privacy is ensured by using pseudonyms (usually public cryptographic keys) for the transactions. An apparent issue with privacy is that all transactions signed with single public-private key pair can be associated to each other.

The security of DLT based solutions can be considered superior to traditional centralized systems thanks to their innate use of cryptographic hashing with private-public key pairs and the distributed means of ensuring information. At least when the DLT-specific security issues are taken care of.

Blockchain-based cryptocurrencies are the common example of how DLTs are leveraged currently, but distributed ledgers (implementations of DLT) can in fact be assembled in a wide variety of ways. Ledgers can trigger actions based on predefined rules (i.e. smart contracts), the mining capacity of participants can be used for meaningful computing instead of calculating trivial math problems, and a ledger can even be controlled by a trusted administrator. (It is negotiable if an administered ledger should be called a distributed ledger, but the technology can be used this way nevertheless.)

Furthermore, a distributed ledger doesn’t actually need a cryptocurrency, it is just a convenient way of ensuring computing and storage capacity in a truly distributed ledger. If computing and storage is provided some other way, a currency is not needed. For example, one or more governmental bodies can ensure the computing and storage capacity of a ledger, which might be the viable solution for the system presented in this post.

Decentralized epidemic alert system (DEAS) resides in a DLT-based cloud, catering the needs of the real world through via digital twins (DT). The DEAS cloud communicates with the personal digital twins (PDT) of people, and with the DTs of other stakeholder organizations, namely a hospital, a health organization and a governmental body. The PDT of a medical doctor has a special role in DEAS.

The main function of DEAS is to analyze the health information coming from the PDTs of the people. DEAS looks for possible signs of health problems and upcoming epidemics. The PDTs can analyze health data locally or send raw health data to the DEAS. (The health data may include e.g. heartbeat readings from a smartwatch and/or body temperature from a tracker ring.) The analysis algorithms can detect an anomaly in health data, implying a possible virus infection. Accompanied with location data, DEAS can detect clusters of multiple infections and detect an upcoming epidemic before it even starts. Detection is performed by machine learning algorithms that actively monitor people that get sick around the same time, backtracking the location and searching for possible common locations. This way, the origin of the virus infection is tracked automatically, and the potentially contaminated people are notified via their PDTs. DEAS also generates reports on the data, providing crucial information to decision makers.

The main figure of this post describes a simple example case in which one person is potentially infected during a known pandemic outbreak, accompanied with consequential notifications and other information flow. The members of Group 1, Anna and Jon have been practicing social distancing and their PDTs are sending their personal data to the DEAS system. In return of sharing the data, Anna and Jon can rely on the knowledge that they will be notified by DEAS in case of nearby infection. If they wish, they can also review the public situation reports generated by DEAS.

Contrastingly, Group 2 is in the middle of an unfortunate situation. The DEAS algorithm has detected a potential infection for Alex, who is sent an alert about the detection. The other Group 2 members Mary and Bob receive a warning. The personal doctor of Alex also receives an alert, along with the personal heath data from the PDT of Alex that the doctor can use to pre-analyze the situation.

The alert is also sent to the nearby hospital that can access the personal data of Alex if needed and make the required preparations. The hospital is up to date of the underlying epidemic situation thanks to the reports provided by DEAS, which are also sent to a health organization and a governmental body. The health organization receives warnings of suspected epidemic outbreaks and the government may receive personal data if enforcement procedures are required to keep the pandemic in control. Each of the three organizations may also be granted various types of control privileges to DEAS.

The reason why also the organizations have DTs is to simplify the information interfaces of the system; this way DEAS only needs to communicate only with entities that follow the same DT data exchange standards (which are still to be defined).

The contents of DEAS consist of three main software components and other supporting code, which are all implemented as open source. Interfaces convey data in and out of the system, data analysis algorithms crunch the numbers to draw conclusions, and the notification system alerts the stakeholders. Open source enables transparency, creating trust for the system.

There are still open questions to be answered. Should people be able to limit their data sharing? How can the system be supported economically? What kind of DLT implementation should DEAS be based on? Who manages the updates of the system? Will DEAS be a national or global system? Some of these questions need decisions by leaders of the world, whereas others may only be solved on the go.

DEAS presents a novel way for decentralized pandemic prevention. There are known problems with DLTs, but they are being solved by numerous researchers and practitioners. The whole field is developing fast, enabling new kinds of implementations. Implementing the DEAS system described in this post is becoming increasingly trivial, we just need to make the initial decisions and start building it.

Finally, I would like to thank Roberto for the encouragement to publish these thoughts and for providing an excellent venue for this post.

About Roberto Saracco

Roberto Saracco fell in love with technology and its implications long time ago. His background is in math and computer science. Until April 2017 he led the EIT Digital Italian Node and then was head of the Industrial Doctoral School of EIT Digital up to September 2018. Previously, up to December 2011 he was the Director of the Telecom Italia Future Centre in Venice, looking at the interplay of technology evolution, economics and society. At the turn of the century he led a World Bank-Infodev project to stimulate entrepreneurship in Latin America. He is a senior member of IEEE where he leads the Industry Advisory Board within the Future Directions Committee and co-chairs the Digital Reality Initiative. He teaches a Master course on Technology Forecasting and Market impact at the University of Trento. He has published over 100 papers in journals and magazines and 14 books.