Dear Colleagues:
We are seeing and hearing about more clinicians clients and patients trying to use AI for assessing QEEG maps and NFB sessions.
I have to disagree with blatant statements about AI efficacy with QEEG and NFB on several levels. I have a friend who is a major player in the AI world and AI development. While AI can be a wonderful tool, in the world of
neurofeedback it can be an incredibly damaging tool.
AI is not perfect. It can make mistakes and in the world of mental health it is leading to depression and suicide. Thus, I try to post articles on AI I my science reports to the ISNR listserv.
-
Unless AI knows an individual's complete background, lifestyle, health history, mental health history, medical history (i.e., Hx of TBI), it has no way to accurately interpret a map or neurofeedback session.
-
It has become such a problem with patients and clients using it to interpret brain maps and neurofeedback sessions some clinicians have had clients/patients come in challenging NFB protocols.
-
It has become concerning enough from an ethics standpoint, that Leslie Sherlin and I will be doing our ISNR Ethics webinar on AI this year.
-
AI is not a licensed professional. Any practitioner relying solely on AI could be in violation of their license and face a potential legal battle. Any client or patient relying on AI to interpret a QEEG or determine efficacy of a neurofeedback protocol could
be faced with misinformation and potentially dangerous information.
-
We are advising anyone doing QEEGs and/or Neurofeedback to put a statement in their informed consent that the patient understands AI could provide misleading and potentially dangerous information. We recommend the patient sign both the informed consent and
Initial the AI statement.
-
Any clinician / practitioner should get a formal letter in writing suggesting that AI working well and integrates with their software. Personally, I am recommending against any use of AI as it pertains to QEEG and NFB.
If you have any case examples of clients/patients or clinicians who have had problematic experiences with AI, please channel them to me. Thank you.
Be Well, Stay Safe
Rob
(Retired)
CONFIDENTIALITY NOTICE ~ HIPAA Privacy Notification:
The information contained in this e-mail message may be privileged, confidential, and/or protected from disclosure. This e-mail message may contain protected health information
(PHI); dissemination of PHI should comply with applicable federal and state laws. If you are not the intended recipient, or an authorized representative of the intended recipient, any further review, disclosure, use, dissemination, distribution, or copying
of this message or any attachment (or the information contained therein) is strictly prohibited. If you think that you have received this e-mail message in error, please notify the sender by return e-mail and delete all references to it and its contents from
your systems.