AI: Augmenting the ECHO Model

What makes ECHO work is the lasting connection between people: trusted, peer-to-peer learning communities.

Illustration of specialists and health workers teaching and learning from each other on a video call

People bring the context

Frontline workers share what really happens in the field, while experts share best practices to address those local challenges.

AI brings the scale

AI can share knowledge with anyone, anywhere, instantly. But on its own, it has no context about lived experience and what goes on in communities day-to-day.

  1. 1

    A collaborative virtual community

    Specialists and local providers learning from each other's cases, with knowledge curated by experts.

  2. 4

    Shared back with the network

    Insights flow back to every participant, in their own language, and spark the next conversation.

iECHO + AI

Collaborative virtual community

 New knowledge ECHO AI Shared insights
  1. 2

    New knowledge, created together

    Every session, case discussion and conversation on iECHO adds to what the network knows.

  2. 3

    Synthesized securely by AI

    AI securely harnesses insights from across the network and summarizes them, with data privacy and governance built in.

AI augments human connection. It never replaces it.

For learners, the session never really ends

iECHO + AI turns every session into learning they can carry with them, on the phone they already use.

  1. 1

    Session recaps

    The key points of every session, in their language, to read or listen to

  2. 2

    Recommended for them

    Programs and sessions matched to their goals and profession

  3. 3

    Practice cases

    Simulated cases from expert-curated material, between sessions

  4. 4

    Answers with sources

    Grounded in their program's own sessions, with the experts who said it

WelcomeAnjali Patil

Your session recapAI

WED4NOV

Gestational diabetes in primary care

Ended at 03:00 PM IST

Program:Maternal Health ECHO

  • Screen at 24 to 28 weeks
  • Start with diet and activity
  • Refer if fasting glucose stays high

Listen · 2 minमराठी

Recommended for youAI

MON9NOV

Diabetes Care ECHO: Cohort 3

Mondays, 02:00 PM IST

Matches your goal: chronic care

Register

Practice caseAI

Priya, 28, is 24 weeks pregnant. Her fasting glucose is 105 mg/dL. What do you do next?

Repeat testDiet and follow-upRefer

From cases curated by your program's experts

Upcoming Sessions

THU12NOV

Postpartum care in the community

02:00 PM - 03:00 PM IST

Program:Maternal Health ECHO

Join Session

Ask about your sessions

When should we test again after delivery?

Dr. Mehta recommended a glucose test 4 to 12 weeks after delivery, and a yearly check after that.

Session 6 · Nov. 2Session 4 · Oct. 19

Concept design. Names and figures are illustrative.

Insights no general AI can give them

A general AI chatbot

What should I watch for in my patients this month?

Seasonal illnesses such as influenza and gastroenteritis are common. Encourage hand washing and vaccination…

  • Textbook knowledge from the public internet
  • Frozen at its training cut-off
  • No idea what's happening in their community

WelcomeAnjali Patil

Trending in your networkAI

Rising this month

Dengue-like fevers after flooding

Raised in 42 sessions across 9 hubs in western India in the past 30 days

What peers are trying

  • Watching warning signs and platelet counts
  • Clearing standing water with community volunteers

Reviewed by Dr. Rao, hub expert

Upcoming Sessions

THU12NOV

Dengue case management

02:00 PM - 03:00 PM IST

Program:Infectious Disease ECHO

Join Session

  • Current

    From this month's sessions, not last year's training data

  • Local

    From peers facing the same conditions, in the same region

  • Trusted

    Curated and validated by subject matter experts

  • Private

    De-identified, with no patient data, by design

Concept design. Names and figures are illustrative.

The ECHO Listening Network

Every conversation, connected: a living knowledge graph of what the network knows, and where.

  1. 1

    Listen

    Sessions, case discussions and questions, with consent

  2. 2

    Transcribe and translate

    In any language our community speaks

  3. 3

    De-identify

    Names and patient details removed at the source

  4. 4

    Extract

    Agents find expert insights, questions and emerging needs

  5. 5

    Validate

    Subject matter experts validate the findings

Post-floodfever cluster Pune hub Maharashtra Dengue InfectiousDisease ECHO Indonesia Gujarat Dr. Rao Jakarta hub Ahmedabad hub TB India Treatment gaps
  • Geography
  • Hub
  • Program
  • Disease area
  • Expert
  • Insight
  • Session

Fever clusters in western India, past 30 days

1 emerging signal

Dengue-like fevers after flooding

42 sessions9 hubs2 states

Validated by Dr. Rao, hub expert

  • Early signals

    For ministries and public health agencies, before they show up in reports

  • Insights back to learners

    What peers are seeing and trying, in their own language

  • Evidence for decisions

    For program design, funders and research

A team of AI agents for every ECHO practitioner

For participants

Session Assistant

Recaps in their language, in text and voice

Learning Advisor

The right program and session at the right time

Practice Coach

Simulated cases from expert-curated material

Support Agent

Help with iECHO, in 25+ languages

For experts and hubs

Expert Copilot

Prepares sessions and reviews insights for the graph

Analyst Agent

Monitoring, evaluation and reports, written for them

Program Designer

Names, banners and curricula for new programs

Privacy Guardian

Catches patient details before anyone sees them

For ministries, funders and researchers

Listening Agents

Emerging needs and signals from across the network

Research Synthesis Agent

900+ peer-reviewed ECHO publications, synthesized in minutes

Impact Agent

Outcomes across programs and countries, on demand

Policy Briefing Agent

Briefs for decision makers, grounded in the network

Every agent draws on the same knowledge graph, under the same guardrails, with people in the loop.

Responsible AI Governance

Human expertise and validation stay at the center of every AI feature on iECHO.

  1. Open, interoperable public goods

    Built on open standards, so the community is never locked into one vendor

  2. Diverse, representative data

    Grounded in local expertise across sectors, geographies and languages, to reduce bias

  3. Multimodal and multilingual

    Text, voice and video in the languages people speak, so every voice is heard

  4. Human validation and expert curation

    Outputs reviewed by people and traceable to trusted sources; humans stay accountable

  5. Trust and safety architecture

    Privacy and security by design, with de-identification so AI scales without exposing anyone

  • No partner data used to train AI models
  • Partners can opt out of AI features at any time
  • Open about where AI is used and what it does
  • Rolled out gradually, with our partners

The iECHO + AI platform

iECHO platform

iECHO app

Web, mobile and WhatsApp

Open APIs

Every capability, as a service

Data

Programs, sessions, people, outcomes

Trust and safetyGuardrailsDe-identificationEvalsHuman reviewAudit trail

Domain agentsEach with its own skills

Session summarizer Learning advisor Practice coach Analyst Listening agent Support agent

Agent harness

ToolsiECHO APIs and search, over MCP

SkillsReusable playbooks, loaded on demand

Agentloop Plan Act Observe Learn

MemoryWhat each learner and program needs

SchedulerRecaps and reports, on their own

Knowledge graphGrounds every answer

  • Geography, hub, program and disease area
  • Expert-validated insights
  • Vector index for semantic search
  • 900+ ECHO publications

LLM gatewayRouting, caching, failover

GeminiClaudeOpen-weight, in countrySpeechEmbeddings

Guiding health care through the AI transition

AI is changing medicine faster than most health systems can adapt. The ECHO Model can help the global community put it into practice safely and fairly.

  • Tools arrive before guidance

    Clinics adopt AI before there's local evidence for how to use it well.

  • Implementation is the hard part

    Validating tools, fitting them into workflows, training staff and governing their use.

  • The gap is widest where resources are thinnest

    Without support, under-resourced health systems risk being left behind.

AI in Health ECHO Experts at the hub Clinicians Nurses and health workers Hospital leaders Ministries Health IT teams Medical educators

Evaluating AI tools Fitting AI into clinical workflows Governance and regulation Bias and equity Privacy and patient safety AI literacy for the workforce

Illustration of a hub team meeting with a specialist on a video call

Transform health systems at scale, with us

iECHO and responsible AI can bring trusted, local expertise to every health worker, everywhere. Join us and help build it together.

Active users
700,000+
Hubs
1,200+
Countries
70+
Languages
25+

Project ECHO and iECHO are registered trademarks of The University of New Mexico

Use the arrow keys or click to move through the slides

All slides

  1. 1AI: Augmenting the ECHO Model
  2. 2For learners, the session never really ends
  3. 3Insights no general AI can give them
  4. 4The ECHO Listening Network
  5. 5A team of AI agents for every ECHO practitioner
  6. 6Responsible AI Governance
  7. 7The iECHO + AI platform
  8. 8Guiding health care through the AI transition
  9. 9Transform health systems at scale, with us