Monday, March 23, 2020

From Tracing to Preemption



There are a number of application (app) assisted contact-tracing projects currently being developed in response to Covid-19. Their objective is to notify persons who may have been infected as quickly as possible so they can be tested or take other precautionary measures. However, with a proper information security architecture, it is possible to notify persons that they may be coming in contact with an infectious individual. This allows the person to preempt the contact or take precautions, such as wearing a mask.

This publication introduced the information security architecture needed for Virus Radar:
Stodolsky, D. S. (1997). Automation of Contagion Vigilance. Methods of Information in Medicine, 36(3), 220-232.

https://sites.google.com/a/secureid.net/dss/automation-of-contagion-vigilance

This study showed acceptability of the approach and how it could be extended to location associated risks (contaminated surfaces, etc.):

Stodolsky, D. S. & Zaharia, C. N. (2009). Acceptance of Virus Radar. The European Journal of ePractice, 8, 77-93. 

https://drive.google.com/open?id=0B_zxYlTkSnKQZXFsXzNwSDd3ZGs

There are two things that are needed for Virus Radar to function. First, the infectious agent must be tracked. In the simplest case, when two phones come within a couple meters of each other they exchange information in order to record a risky contact. However, a risky contact can be preempted, if a person’s phone can’t present a digital health certificate. Failure to present a certificate can occur for many reasons, so the risk of stigmatization is avoided.

Second, if a person can’t present a health certificate, they can avoid potential ostracism by getting tested or resolving malfunction of the Virus Radar application or of their phone. New health certificates are delivered to all users on a daily basis, so information about the extent of infection is always current. This information can be updated rapidly, because those suspected of being newly infected report for testing upon not receiving a fresh certificate.

An advantage of this approach is that it uses peer pressure to accelerate adoption of the app. If comprehensive coverage can be achieved, then it becomes possible to rapidly unroll the infection chains and prevent all likely infected persons from contact with the at-risk population. Both contact tracing and motivating of persons to be tested results from peer influence. Professionals can focus upon testing of persons who come for assistance. 

It appears that TraceTogether can be extended to include a “MeetSecure” capability.

https://www.gov.sg/article/help-speed-up-contact-tracing-with-tracetogether

6 comments:

  1. So... what are European governments waiting for? This app is what we need right now!

    ReplyDelete
    Replies
    1. An app could be out within a month, with government cooperation. Vaccine development and testing will take at least a year. So, once lockdowns are lifted, we will likely see new outbreaks, unless something like this is available. This article suggests that containment requires immediate notification of at-risk persons:

      http://www.ox.ac.uk/news/2020-03-17-infectious-disease-experts-provide-evidence-coronavirus-mobile-app-instant-contact

      Therefore, the upgrade I have suggested may be crucial.

      If anyone can get a response from a government or company that is willing to support this development, I am ready to give it my full attention. Research proposals to the European Union Framework Program and to the Danish government yielded no response. After half a dozen attempts to get funding, I gave up on the idea of a contagion management test. After the #MeToo media explosion, I decided that I might pursue a test by focussing on the frontend of the design. A successful workshop led nowhere.

      Delete
  2. This comment has been removed by the author.

    ReplyDelete
  3. Probably the simplest explanation of my distributed contagion vigilance proposal is based upon an analogy between computer and biological viruses:

    Assume that everyone is carrying a mobile phone and these phones can propagate a computer virus.
    If a person comes within ten meters of you, their computer virus will be transmitted to your phone via Bluetooth or WiFi. This generates an alert sound from your phone, allowing you to avoid risky contact with that person.

    It gets extremely complicated to implement this in a secure way, so as to avoid stigma and to ensure cooperation. Normally, I wouldn’t even try to explain how it works, since people would get the idea from using the application. Find a set of slides illustrating this process attached: Dynamic Risk-based surveillance - PDF and Keynote formats.


    We can estimate the cost from related implementations:

    The TraceTogether app required about a thousand hours of work to implement.
    We estimated that it would cost a few million Euro to develop and test a system like this for protection of hospitals.


    The application would allow us to trace any agent entering a population and isolate all infected persons. This would apply to all epidemiological agents spread by direct contact and via surfaces - with the more advanced model developed for hospitals.
    I attach a Virus Radar Simulation in mov format - SWF format available.

    ReplyDelete
  4. The above mentioned files have been uploaded to the files area of the Medical Ethics list - subscription required for access.

    https://groups.io/g/MedicalEthics/files/

    ReplyDelete
  5. All followers of this blog, David Stodolsky has died, likely sometime in the evening of September 26. Thanks in advance for your condolences. Please only contact me if there are special issues, as I (brother, Marvin.Stodolsky@gmail.com) am managing considerable traffic.

    ReplyDelete