Doctolib plans to use your health data for its AI research

LaRedaction Posted messages 879 Registration date   Status Administrator Last intervention   -  
durarex Posted messages 15 Registration date   Status Member Last intervention   -
https://fr.123rf.com/photo_106158386_ai-intelligence-artificielle-dans-la-technologie-m%C3%A9dicale-moderne-iot-et-automatisation.html?downloaded=1
Doctolib plans to use its users' medical data for an AI research project, unless they object. This includes demographic and health data of users and their relatives linked to their Doctolib account. The data will be pseudonymized, not anonymized. While this does not allow direct identification, it remains possible to re-identify a person with additional information. Doctolib assures that objection to this use will have no impact on access to Doctolib services. What do you think?
Source
0

3 answers

  1. durarex Posted messages 15 Registration date   Status Member Last intervention   4
     

    Thank you for the information.

    The Doctolib opposition form is 'easy' to use.

    With the reminder that I was not born on 07/09/2026, it is impossible to enter my birth date directly.

    You have to go back in time month by month, year after year with the slider.

    A somewhat clumsy way for Doctolib to discourage unenrollment...

    Which had the opposite effect for me...

    1
    1. brucine Posted messages 25218 Registration date   Status Member Last intervention   4 182
       

      Hello,

      Maybe we should tackle the problem from the other side of the telescope.

      A certain number of doctors no longer have medical secretaries, or have them outsourced to a call center and are difficult to reach.

      If, in a number of cases, it is not about consulting one's GP when you have one, but about finding a practitioner in your area, any practitioner for the moment’s procedure as long as they have availability (to caricature, a dentist if you don’t have one habitually), it would be a struggle to navigate given the previous reasons, and Doctolib would then be almost indispensable.

      0
      1. babacool45 Posted messages 88 Registration date   Status Member Last intervention   > brucine Posted messages 25218 Registration date   Status Member Last intervention  
         

        Your comment has nothing to do with the problem... we’re not blaming Doctolib, but rather its way of once again draining our data under a fallacious pretext. In fact, they want with our data just to train an AI

        -1
      2. brucine Posted messages 25218 Registration date   Status Member Last intervention   4 182 > babacool45 Posted messages 88 Registration date   Status Member Last intervention  
         

        The one in <1> says that since that’s how it is, he unsubscribes from it.

        He therefore has to do with the issue because if Doctolib does not back down on this use or if it is impossible unlike what is announced to oppose it (otherwise the problem does not exist) we have the choice (as I have often suggested, somewhat relative) to use it or not.

        Note that for example Google or Microsoft, not to mention the various social networks, do not refrain from doing the same and we hear fewer shrieks of outrage.

        1
    2. olle
       

      Just click the pencil icon in the popup to enter the date directly...

      0
      1. durarex Posted messages 15 Registration date   Status Member Last intervention   4 > olle
         

        Except that: When I did the operation on my smartphone, I had to go field by field – dd – mm – yyyy – and it was impossible to enter the date directly.

        -1
  2. HelpiOS Posted messages 8473 Registration date   Status Moderator Last intervention   11 958
     

    Hello,

    Here is the direct link to the objection form.


    1
    1. brucine Posted messages 25218 Registration date   Status Member Last intervention   4 182
       

      Hello,

      Users receive this link preceded by a long ramble explaining that this data has been submitted to the CNIL, is directed to 3 research organizations (INRIA, Inserm, Université Paris Cité) and that this data, which is only submitted to these authorized research teams, is secure and does not allow direct identification of the subscriber.
       

      Something of the same spirit exists intermittently in the Virenque fashion (without your knowledge or consent) where, for example, a nominative collection of patient data from a few voluntary hospitals has recently been authorized and sent to clinical research teams also duly authorized.

      More generally, health insurance schemes collect and analyze everything that is submitted for reimbursement for personal purposes (which practitioner billed "too many" procedures of a given category...) or global purposes (mandatory declarations of certain diseases where, by definition, the index patient is identified...).

      There are theoretically barriers, including medical confidentiality, but the administrative level of these funds is not subject to them; it concerns only the consulting practitioners; I have occasionally told their dear studies employees who contested such element not administrative but medical of an act, rightly or wrongly, but who do not have the competence to know about it or to judge it.

      -2
  3. Bayesien
     
    Personally, it doesn’t bother me at all. Science saved my life four times, for different reasons, including a serious accident. In return, it has never asked me for anything. It simply allowed me to continue living, to see my family grow, and to enjoy life. So, if I can contribute in return, within my means, it seems completely natural to me. I do it with joy, but also with deep gratitude. Even if it means losing a bit of anonymity or a small part of my freedom, it is not a sacrifice in my eyes.

    Without the advances of science (of which AI is already a part and will become an even bigger part in the future) I would not be here today! In a sense, the freedom and the pleasure of living that I still enjoy, I owe entirely to it. I therefore consider it both fair and normal to give back, as much as possible, a portion of what I have received.
    0