Vivatech : A challenge won by DataValoris

From May 16 to 18, 2019 was held the Vivatech tradeshow at the Porte de Versailles. We were selected among the 400 Start-ups who had applied to take up the challenges proposed by Sanofi. We were invited to share their booth in Sanofi’s Tech4Health Lab.

The DataValoris team on Sanofi’s showroom in Vivatech

We are the only start-up to be a finalist not just of one but of two challenges! The de-identification of patient data and the reduction of the wandering of the medical diagnosis of rare diseases.

Challenge 2 : De-identification of patient’s date won by DataValoris

Access to many patient data is crucial to advance research. However, it is equally important to preserve the anonymity of the patients concerned by information as sensitive as their state of health. It is not a matter of simply removing the names and surnames of the patients to guarantee their anonymity because we must avoid being able to re-identify them with the data, thanks to their age or other characteristics (to be the only person 1.95m or more than 100 years in a small village, for example.)

It is also necessary that the decisions taken to de-identify do not skew the data (controlled depletion) and therefore the research that will ensue.

These particularly thorny antagonostic objectives require subtle solutions.

Finalist Start-ups for this challenge were : Proxem, Kairntech, Neuroadvise or in DataValoris

We are very proud to announce that we have been selected by Sanofi as the winners of this challenge!

The winning solution :

Here are a few highlights of our proposition :

The ability to search for keywords on social networks patients, and public (like hashtags on Twitter).

Sliders that control the fineness of the disidentification without introducing bias in the data.

An unidentified data-set correctly indexed with the right criteria.

All based on our know-how in data processing and thanks to our technology NNTO of deep-neuroévolution..

Here is an example of deidentified data :

Challenge 5: Reduce the wandering of medical diagnosis in the case of rare diseases won by Médicus AI

Even if the diseases are rare, their diversity is great and the patients affected are plenty ! This is an important public health issue because people with these diseases suffer for many years before they can be properly diagnosed and offered effective treatment if they exist. In order to reduce these delays, Sanofi launched the initiative UNIR published a white book on this matter.

Finalist Start-ups for this challenge were : Anamnèse, Yuni, DataValoris, and Medicus.AI

DataValoris at the heart of health issues during Vivatech

Whether we are talking about e-health, connected objects related to health, or tools of health professionals, Artificial Intelligence will be crucial to make all these data a strength in the advancement of research.

As such, Jean-Patrice Glafkidès, founder of DataValoris was invited to the microphone of Caroline Paré in her show Priority Health on RFI. The subject of this show: Health and Artificial Intelligence.

By his side :

  • David Gruson, professor in Sciences Po Paris. Author of the book La machine, le médecin et moi.
  • Bola Bardet, founder of Susu which allows Africans in France to take care of their family back home.
  • Joseph Quenum, founder ofe-yone

Feel free to listen to the podcast.

The entire DataValoris team would like to thank Sanofi for its invitation to Vivatech. We are delighted and proud to have won this challenge and look forward to starting a collaboration on this subject.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.