Myia Health

Simon MacGibbon, Myia Health

Enterprise tech, IBM , McKinsey, BCGDV cofounder

Launched Myia at 2017


Individualied population health

Belief: country is struggling to move to a value based healthcare system.

Impossible, unless you can take care of pateints outside of the clinical setting

Hence we built a learning system that helps healthcare workers beyond the four walls of the hospital

med is traditionally episodic, facility based

but most of the medical expense is lived in the real world

So we curate data from the edge (working with 3rd party OEMs)

Kit things

Pair with home hubs, and send to patients

100 data feature, eg functional markers, biomarkers, everyday life, environmental context

Zip code can reveal a lot of information from a social and financial perspective as well

ALl of this can be fed into

Work in Myia, but store in Epic. 2 way exchange with EHR


How is AI relevant

1 Predict events before they happen

2 Automate clinical grunt work

3 Optimisation of resource utilisation

Use of asynchronous info. Break out of serial flooding of patients

4 Virtualise venue of care

Opens up new world of care

5 Personalise care plans and pathways

Next steps for this specific patients?

6 Learn across systems

Healthcare is not a learnign system. Clinicians do not learn from each other or from their past selves

7 Restore humanity in clinician / patient relationship


Mercy Virtual

Future is already here, but it's just unevenly distributed

A hospital built in Chesterton, Missouri

Also has a engagement program (population health and care program for chronic patients)

Operate under a value based program

They take on 100% risk with payer on Medicare Advantaged population???

ALso happens to have the largest ACO in the country???

Nurses, coordinators, APs take care of patients virtually

A hospital without beds

Telemetry coming from the edge is crucial

Nutrition works, social workers, chaplains, in addition to PCP

AI & ML is essential to extract meaning from edge data

50% reduction in avoidable hospital readmission

30% reduction in hospital expenses

High 90s patient scores in this program


Context and complications

  • They keep giving them all these new electronic systems... they spend your whole appointment cussing at the computer

Launching incremental technology further draws away the clinician from the patient interaction


Rural patients or broadband-less patients

Offer IoT router that goes along with the product

Silicon Valley culture overlooks the fact that not everyone has wifi


USFDA is formulating guidance and regulation around AI/ML

Generally good direction with natural constraints

eg models updating their priors instantly aren't compatible with algorithm checkpoints

Myia is a non device clinical decision support platform

Must let physicians independently evaluate the basis of a recommendation

Not looking at images

Fit into


Project Nightingale

Have to be on the front foot on the issue of trust

Otherwise the entire industry will be slowed down


Hippocratic oath?

visionary call to action, search for STAT article


digital silo

interoperability regulation released by

let patients get control over their EHR

getting data across several providers to piece together a single view of the patient



Who owns the data? What rights do the patients have over their data?

Contract is between patient and provider

Myia signs a HIPAA BAA with the provider

Do not have personally identifiable individual (PII) information

// relevant to cybersecurity as well

Do you see more adoption of AI in public or private sector in AI? What is driving the adoption?

A large community hospitals in 2015 had a new vision

Engaged and invested in us

Incredibly hard to scale without ML & AI

You'll only receive email when they publish something new.

More from 9631
All posts