Keynote
In person
LeadersGoldDiscovery

Closing the dementia detection gap with Agentic AI

  • Date
    7 July 2026
    Timeframe
    15:00 - 15:15
    Duration
    15 minutes
    • Days
      Hours
      Min
      Sec

    Dementia already affects more than 50 million people worldwide and is projected to triple by 2050, with two thirds of cases occurring in low and middle income countries (LMICs). Yet diagnosis remains profoundly unequal: underdiagnosis rates in LMICs reach 75 to 90%, compared to approximately 50% in high income countries. This gap is exacerbated by workforce shortages: there are only approximately 0.1 neurologists per 100,000 people in LMICs versus approximately 7.1 per 100,000 in HICs. Despite this disproportionate burden, most AI initiatives for Alzheimer’s disease and related dementias have been developed and validated in high income settings, typically using highly educated, clinic based cohorts. This mismatch perpetuates inequities, as models fail to generalize to the cultural, linguistic, and healthcare realities of LMIC populations.

    This project directly addresses these gaps by developing an Agentic AI conversational system for early dementia detection, specifically designed for low education, low income populations in Argentina as a model for LMICs. By leveraging ubiquitous platforms, such as WhatsApp, and portable neuropsychological protocols validated in local pilots, the project delivers scalable, culturally adapted, and accessible screening tools that democratize early detection. Its core hypothesis is that multimodal conversational and cognitive signals, captured through text, speech, video, and behavior, can detect early Mild Cognitive Impairment with accuracy comparable to gold standards, while remaining feasible in resource constrained environments.

    Share this session with your network
    • Session starts in
      Days
      Hours
      Min
      Sec

    Are you sure you want to remove this speaker?