Authors

Immunitor

Immunitor is a self-reporting platform, yes, another one! There are so many awesome self-reporting tools being developed, however, being biotech scientists, we saw a need that was still unmet: data validation.

During the CodeVsCOVID19 Hackathon we prototyped a platform where patients that get tested can submit their result and being in full control of how much additional information they want to provide. However, the first question we ask is "Do you have a test serial number?". We want to collect data as anonymously as possible, but we also want to be sure that the entries are valid, that they don't get over counted, and that are compatible with bulk uploads (in case hospitals also want to share data on Immunitor). Using the test serial number allows:

  • to be completely anonymous: we don't need to know the name, the email or anything at all from the user.
  • to tag the data entry as validated.
  • to retrieve information about the type of test (PCR based vs antibody based).
  • to merge data from different cantons, manufacturers, clinics..
  • to collect the data that will come from home testing, that would otherwise be lost.
  • to have a well-organized database (on OpenBis) that can be accessed by researchers and authorities.

Additionally, we allow for subsequent entries, by providing a QR code that the user can store and use again to modify previous entries (for example if you don't know what the test serial number is, and then you find out and want to validate your data) and, most importantly, add subsequent entries that maintain a temporal relationship (for example if you repeat the test and change your status from infected to immune). Last but not least, we provide a neat **live data visualization* page, where anyone can see how their contribution is helping to fight the Covid19 crisis.

Take a look yourself! (but have mercy!):

Here's our website: Immunitor. We are putting all of our hearts and free time in Immunitor, but bear in mind 2 out of 3 people in our team never saw an html file when the project started!  If you see bugs or have any feedback to give us please feel free to comment! We would love your inputs.

Is it really so simple?

On the user side: absolutely yes. On our side: no, it's not. We are aware that implementing a strategy based on serial numbers to track the valid tests will require a huge effort, but we really, really, believe this is a good approach! Reaching out to clinics and manufacturers is not easy, especially now that they are overloaded. However, thanks to the QR code system we're putting in place, it will be possible to retroactively validate the data. When a user is contributing there are no mandatory fields (except for the status: not infected, infected and immune), so even if the test serial number is not available right now it will be possible to add it in a second moment and the entry will be marked as "validated" in our database.

Why oh why?

While great effort is being put in increasing the testing capacity, systematically collecting data regarding the outcome is still very challenging. This will be especially impelling as soon as home testing becomes a viable option. Putting together entries from different clinics, single patients, manufacturers, region, test types etc. is a lot of work. We believe that having a common database that can be compatible with multiple sources and can be validated is crucial to keep the finger on the pulse of this crisis.

Try It out

Hackathons

Technologies

flask, html5, openbis, pandas, python, wtform

Devpost Software Identifier

254321