AI & Dementia Initiative#

This project sits at the intersection of dementia care, sleep medicine, caregiver support, and carefully constrained AI.

It is being developed as a clinician-led initiative rather than a generic chatbot product. The goal is not to replace family, staff, or medical judgment. The goal is to extend caregiver capacity, improve the quality of engagement, and preserve dignity in a setting where time, staffing, and emotional bandwidth are often in short supply.

Why This Exists#

My interest in dementia care is personal as well as professional. My mother is in memory care, and I have become increasingly convinced that dementia care needs more than slogans about innovation. It needs thoughtful, humane systems that help families, support staff, and respect the realities of disease progression.

At the same time, I believe sleep is underappreciated in dementia care and dementia risk. Sleep disturbance, sleep apnea, insomnia, REM sleep behavior disorder, hypersomnia, medication effects, and polypharmacy often shape quality of life long before anyone gets coherent help.

Current Focus#

The initiative currently has two connected tracks:

  1. Sleep and dementia care: concierge and facility-based evaluation of sleep problems in older adults and memory-care settings, with attention to obstructive sleep apnea, insomnia, parasomnias, deprescribing, and caregiver education.

  2. Caregiver-centered AI support: a human-reviewed, ethically constrained approach to AI-assisted engagement for people with dementia, designed to support comfort, orientation, routine, and activities of daily living without pretending to be a substitute for human presence.

Design Principles#

  • Human beings remain in charge. Caregivers and clinicians outrank the system.
  • No deceptive framing. The project should not encourage false emotional dependence or hide that an interaction is simulated.
  • Safety before cleverness. Any useful system in this setting needs clear boundaries, escalation rules, and human review.
  • Clinical humility. This is not a diagnostic engine and should not posture as one.
  • Dignity matters. The point is not novelty; the point is better days for patients and less strain for caregivers.

What I Am Building Toward#

  • A practical sleep-medicine and dementia-support offering for families and facilities.
  • A research and pilot pathway for ethically constrained AI-supported engagement.
  • A collaboration model that could include clinicians, memory-care operators, caregivers, and technically capable builders.
  • A more serious public-facing website for the initiative once the product, naming, and partner structure are settled.

Areas of Interest#

  • Sleep evaluation in memory care
  • Polypharmacy and deprescribing
  • Caregiver stress and support
  • Structured engagement for dementia patients
  • Human-in-the-loop AI workflows
  • Cross-disciplinary dementia prediction and trajectory research, including sleep-informed approaches

Collaboration#

I am interested in speaking with:

  • memory-care organizations
  • clinician collaborators
  • caregivers and families
  • technically strong builders interested in responsible medical-adjacent AI
  • funders, grants, and pilot partners aligned with the ethics of the project

For now, this page is the public placeholder for a larger effort taking shape.

Contact: bharris@eusomniamd.com

Contact