SlideShare a Scribd company logo
AI for healthcare: Scaling access and
quality of care for everyone
Anitha Kannan
Xavier Amatriain
MLConf 10/08/2019
● >50% world’s population with
no access to essential health
services
● US…
○ 10% of adult population
has no health insurance
○ 28% of working adults are
under insured
Healthcare access is a major issue
Kaiser Family Foundation analysis of the 2017 National Health Interview Survey
Merrit Hawkins, 2017 survey
shortage of 120,000 physicians by 2030
Patient-Doctor interaction
● Doctors have ~15 minutes to capture
pertinent information about a patient,
diagnose + recommend treatment
● 30% of the medical errors causing
~400k deaths a year are due to
misdiagnosis
2069 doctors solve 1572 HumanDx cases
Online search and/or Healthcare access?
“72% of internet users say
they looked online for
health information within
the past year”
“More than ⅓ use Internet
to self-diagnose”
[Pew Research]
1.4M daily
25M daily
Need more than Google can
deliver
Less cost and friction
than PCP visit
We have an opportunity to
reimagine healthcare
We have an obligation
opportunity to reimagine
healthcare
Looking Forward: Towards AI powered Learning Health Systems
● AI + human practitioners
for Quality Care
● Less than 20% of the cost
for best healthcare access
● Mobile First Care, 24/7
always on
What are we doing?
● Mission: Provide the world's
best healthcare for everyone
● Product: User-facing mobile
primary care app
● Team: Building an awesome
and diverse team
● Approach: State-of-the-art
AI/ML + product/UX/clinical
AI-based interaction
AI + Health coaches
AI + Doctors
Breakthroughs in AI & healthcare
Peer-reviewed research at Curai
AI “in the wild”: Learning health systems
Automation/AIAutomation/AI
Automation/AI in healthcare
Pertinent
information
gathering
Assessment
(Diagnosis,
Triaging etc)
Plan
(Next steps,
treatments)
Chief
complaints
AI in the wild: Desired properties
● Easily extensible
○ Incrementally/iteratively learn from
“physician-in-the-loop” or from
additional data
● Knows what it does not know
○ Models uncertainty in prediction
○ Enables fall-back to
“physician-in-the-loop”
Automation/AI in healthcare
Automation/AIAutomation/AI
Pertinent
information
gathering
Assessment
(Diagnosis,
Triaging etc)
Plan
(Next steps,
treatments)
Chief
complaints
AI for assisted diagnosis (since 1980s)
● Expert systems
○ Mycin, Internist-1, DxPlain, VDDx,
QMR
● Covers over 1000 diseases
and 3500+ findings
○ Most comprehensive diagnosis
model, so far
○ 30+ years of expert curation
based on research and
evidence-based literature
Expert systems in the wild?
● Not easy to extend
○ Costly, time consuming and
time-delayed
○ Poor generalization to new places
● Does not know what “it
doesn’t know”
○ Constrained to diseases in the
system
Assisted diagnosis in the wild
1. Extensibility
a. Diagnosis as a ML task
i. Expert systems as a prior
b. Modeling less prevalent diseases
i. Low-shot learning
2. Knowing what you don’t know
a. Measures of uncertainty in prediction
b. Allows fall-back to
“physician-in-the-loop”
Assisted diagnosis in the wild
1. Extensibility
a. Diagnosis as a ML task
i. Expert systems as a prior
b. Modeling less prevalent diseases
i. Low-shot learning
2. Knowing what you don’t know
a. Measures of uncertainty in prediction
b. Allows fall-back to
“physician-in-the-loop”
x
Clinical case simulator
Example of simulated case
Knowledge base
central to expert
systems
Expert systems as prior
ML models for diagnosis
clinical cases simulated
from expert system
From expert systems to ML model for diagnosis
ML models for diagnosis
clinical cases simulated
from expert system
From expert systems to ML model for diagnosis
clinical cases from other sources eg.
electronic health records
Assisted diagnosis in the wild
1. Extensibility
a. Diagnosis as a ML task
i. Expert systems as a prior
b. Modeling less prevalent diseases
i. Low-shot learning
2. Knowing what you don’t know
a. Measures of uncertainty in prediction
b. Allows fall-back to
“physician-in-the-loop”
Assisted diagnosis in the wild
1. Extensibility
a. Diagnosis as a ML task
i. Expert systems as a prior
b. Modeling less prevalent diseases
i. Low-shot learning
2. Knowing what you don’t know
a. Measures of uncertainty in prediction
b. Allows fall-back to
“physician-in-the-loop”
Open-set diagnosis
Amblyopia
Gastroenteritis
Diseases within
diagnostic scope
Open-Set diagnosis
Universe of diseases
Amblyopia
Diabetic
Ophthalmoplegia
Gastroenteritis
is aware and avoid misclassifying unknown diseases as known
Diseases within
diagnostic scope
Open-Set diagnosis
Universe of diseases
Amblyopia
Diabetic
Ophthalmoplegia
Extra diseases
Gastroenteritis
avoids
misclassifying
unknown
diseases as
known.
Diseases within
diagnostic scope
Entropic open-set loss: Maximize predictive
entropy of unseen examples
AI in-the-wild: Desired properties
● Easily extensible
○ Incrementally/iteratively learn from
“physician-in-the-loop” or from
additional data
● Knows what it does not know
○ Models uncertainty in prediction
○ Enables fall-back to
“physician-in-the-loop”
Medical Information gathering “in-the-wild”
Real users with health
issues that an AI medical
agent may not understand
Looking Forward...
● AI + human practitioners
for Quality Care
● Less than 20% of the cost
for best healthcare access
● Mobile First Care, 24/7
always on
AI-based interaction
AI + Health coaches
AI + Doctors
https://firstopinionapp.com/
39.6 M Californians
with access to high
quality affordable
primary care

More Related Content

What's hot

How Real-time Analysis turns Big Medical Data into Precision Medicine
How Real-time Analysis turns Big Medical Data into Precision MedicineHow Real-time Analysis turns Big Medical Data into Precision Medicine
How Real-time Analysis turns Big Medical Data into Precision Medicine
Matthieu Schapranow
 
Introduction to Healthcare Analytics
Introduction to Healthcare Analytics Introduction to Healthcare Analytics
Introduction to Healthcare Analytics
Experfy
 
Overview of Health IT
Overview of Health ITOverview of Health IT
Overview of Health IT
Nawanan Theera-Ampornpunt
 
Statistics For Health Science and Its Impacts
Statistics For Health Science and Its ImpactsStatistics For Health Science and Its Impacts
Statistics For Health Science and Its Impacts
Cashews
 
Analytics in healthcare
Analytics in healthcareAnalytics in healthcare
Analytics in healthcare
AnushkaAlok
 
GuoJian CV2014.6.26提交
GuoJian CV2014.6.26提交GuoJian CV2014.6.26提交
GuoJian CV2014.6.26提交
JIan Guo
 
ICT in Healthcare
ICT in HealthcareICT in Healthcare
ICT in Healthcare
Nawanan Theera-Ampornpunt
 
Healthcare analytics
Healthcare analytics Healthcare analytics
Healthcare analytics
Arun K
 
Health Informatics- UMKC, Presentation
Health Informatics- UMKC, Presentation Health Informatics- UMKC, Presentation
Health Informatics- UMKC, Presentation
Nikhil Kassetty
 
Venkatesh_CV
Venkatesh_CVVenkatesh_CV
Data to help patients 101
Data to help patients 101Data to help patients 101
Data to help patients 101
Wen Dombrowski MD MBA
 
Overview of Health IT (October 2, 2016)
Overview of Health IT (October 2, 2016)Overview of Health IT (October 2, 2016)
Overview of Health IT (October 2, 2016)
Nawanan Theera-Ampornpunt
 
IRJET - An Effective Stroke Prediction System using Predictive Models
IRJET -  	  An Effective Stroke Prediction System using Predictive ModelsIRJET -  	  An Effective Stroke Prediction System using Predictive Models
IRJET - An Effective Stroke Prediction System using Predictive Models
IRJET Journal
 
Predictive Analytics in Healthcare
Predictive Analytics in HealthcarePredictive Analytics in Healthcare
Predictive Analytics in Healthcare
ICFAIEDGE
 
Stroke Prediction
Stroke PredictionStroke Prediction
Stroke Prediction
MamathaGuntu1
 
Computer science for health
Computer science for healthComputer science for health
Computer science for health
Md. Shafiuzzaman Hira
 
Data explosion in medicine: challenges and opportunities
Data explosion in medicine: challenges and opportunitiesData explosion in medicine: challenges and opportunities
Data explosion in medicine: challenges and opportunities
Ourlad Alzeus Tantengco
 
Health Informatics and Patient Safety
Health Informatics and Patient SafetyHealth Informatics and Patient Safety
Health Informatics and Patient Safety
Health Informatics New Zealand
 
Medical Informatics
Medical InformaticsMedical Informatics
Medical Informatics
Reena Titoria
 
resume
resumeresume
resume
Xinran Huang
 

What's hot (20)

How Real-time Analysis turns Big Medical Data into Precision Medicine
How Real-time Analysis turns Big Medical Data into Precision MedicineHow Real-time Analysis turns Big Medical Data into Precision Medicine
How Real-time Analysis turns Big Medical Data into Precision Medicine
 
Introduction to Healthcare Analytics
Introduction to Healthcare Analytics Introduction to Healthcare Analytics
Introduction to Healthcare Analytics
 
Overview of Health IT
Overview of Health ITOverview of Health IT
Overview of Health IT
 
Statistics For Health Science and Its Impacts
Statistics For Health Science and Its ImpactsStatistics For Health Science and Its Impacts
Statistics For Health Science and Its Impacts
 
Analytics in healthcare
Analytics in healthcareAnalytics in healthcare
Analytics in healthcare
 
GuoJian CV2014.6.26提交
GuoJian CV2014.6.26提交GuoJian CV2014.6.26提交
GuoJian CV2014.6.26提交
 
ICT in Healthcare
ICT in HealthcareICT in Healthcare
ICT in Healthcare
 
Healthcare analytics
Healthcare analytics Healthcare analytics
Healthcare analytics
 
Health Informatics- UMKC, Presentation
Health Informatics- UMKC, Presentation Health Informatics- UMKC, Presentation
Health Informatics- UMKC, Presentation
 
Venkatesh_CV
Venkatesh_CVVenkatesh_CV
Venkatesh_CV
 
Data to help patients 101
Data to help patients 101Data to help patients 101
Data to help patients 101
 
Overview of Health IT (October 2, 2016)
Overview of Health IT (October 2, 2016)Overview of Health IT (October 2, 2016)
Overview of Health IT (October 2, 2016)
 
IRJET - An Effective Stroke Prediction System using Predictive Models
IRJET -  	  An Effective Stroke Prediction System using Predictive ModelsIRJET -  	  An Effective Stroke Prediction System using Predictive Models
IRJET - An Effective Stroke Prediction System using Predictive Models
 
Predictive Analytics in Healthcare
Predictive Analytics in HealthcarePredictive Analytics in Healthcare
Predictive Analytics in Healthcare
 
Stroke Prediction
Stroke PredictionStroke Prediction
Stroke Prediction
 
Computer science for health
Computer science for healthComputer science for health
Computer science for health
 
Data explosion in medicine: challenges and opportunities
Data explosion in medicine: challenges and opportunitiesData explosion in medicine: challenges and opportunities
Data explosion in medicine: challenges and opportunities
 
Health Informatics and Patient Safety
Health Informatics and Patient SafetyHealth Informatics and Patient Safety
Health Informatics and Patient Safety
 
Medical Informatics
Medical InformaticsMedical Informatics
Medical Informatics
 
resume
resumeresume
resume
 

Similar to AI for healthcare: Scaling Access and Quality of Care for Everyone

Big Data Provides Opportunities, Challenges and a Better Future in Health and...
Big Data Provides Opportunities, Challenges and a Better Future in Health and...Big Data Provides Opportunities, Challenges and a Better Future in Health and...
Big Data Provides Opportunities, Challenges and a Better Future in Health and...
Cirdan
 
2016 Healthcare Trends
2016 Healthcare Trends2016 Healthcare Trends
2016 Healthcare Trends
GSW
 
What Do Patients Really Want Out Of Adherence Technology?
What Do Patients Really Want Out Of Adherence Technology?What Do Patients Really Want Out Of Adherence Technology?
What Do Patients Really Want Out Of Adherence Technology?
Inspire
 
Power to the Patient
Power to the PatientPower to the Patient
Power to the Patient
Bastian Greshake
 
How to Use Data to Improve Patient Safety: A Two-Part Discussion
How to Use Data to Improve Patient Safety: A Two-Part DiscussionHow to Use Data to Improve Patient Safety: A Two-Part Discussion
How to Use Data to Improve Patient Safety: A Two-Part Discussion
Health Catalyst
 
There Is A 90% Probability That Your Son Is Pregnant: Predicting The Future ...
There Is A 90% Probability That Your Son Is Pregnant:  Predicting The Future ...There Is A 90% Probability That Your Son Is Pregnant:  Predicting The Future ...
There Is A 90% Probability That Your Son Is Pregnant: Predicting The Future ...
Health Catalyst
 
AI in Healthcare.pptx
AI in Healthcare.pptxAI in Healthcare.pptx
AI in Healthcare.pptx
Dr. Anik Chakraborty
 
Wearable Health, Fitness Trackers, and the Quantified Self
Wearable Health, Fitness Trackers, and the Quantified SelfWearable Health, Fitness Trackers, and the Quantified Self
Wearable Health, Fitness Trackers, and the Quantified Self
Steven Tucker
 
Meeting healthcare challenges: what are the challenges and what is the role o...
Meeting healthcare challenges: what are the challenges and what is the role o...Meeting healthcare challenges: what are the challenges and what is the role o...
Meeting healthcare challenges: what are the challenges and what is the role o...
Mohammad Al-Ubaydli
 
Patient experience: where can we improve?
Patient experience: where can we improve?Patient experience: where can we improve?
Patient experience: where can we improve?
Consumers Health Forum of Australia
 
20130226 impact van zorg 2 0 op onze samenleving
20130226 impact van zorg 2 0 op onze samenleving20130226 impact van zorg 2 0 op onze samenleving
20130226 impact van zorg 2 0 op onze samenleving
D3 Consutling
 
20130226 impact van zorg 2 0 op onze samenleving
20130226 impact van zorg 2 0 op onze samenleving20130226 impact van zorg 2 0 op onze samenleving
20130226 impact van zorg 2 0 op onze samenleving
Ann Huygelier
 
Big Data and the Promise and Pitfalls when Applied to Disease Prevention and ...
Big Data and the Promise and Pitfalls when Applied to Disease Prevention and ...Big Data and the Promise and Pitfalls when Applied to Disease Prevention and ...
Big Data and the Promise and Pitfalls when Applied to Disease Prevention and ...
Philip Bourne
 
"Think, think, think": e-Patient Dave address to ECPC annual meeting
"Think, think, think": e-Patient Dave address to ECPC annual meeting"Think, think, think": e-Patient Dave address to ECPC annual meeting
"Think, think, think": e-Patient Dave address to ECPC annual meeting
e-Patient Dave deBronkart
 
MedicalResearch.com: Medical Research Exclusive Interviews December 31 2014
MedicalResearch.com:  Medical Research Exclusive Interviews December 31 2014MedicalResearch.com:  Medical Research Exclusive Interviews December 31 2014
MedicalResearch.com: Medical Research Exclusive Interviews December 31 2014
Marie Benz MD FAAD
 
How AI is transforming Mobile Health
How AI is transforming Mobile HealthHow AI is transforming Mobile Health
How AI is transforming Mobile Health
VentureDive
 
Superpatients webinar for Natl Network of Libraries of Medicine
Superpatients webinar for Natl Network of Libraries of MedicineSuperpatients webinar for Natl Network of Libraries of Medicine
Superpatients webinar for Natl Network of Libraries of Medicine
e-Patient Dave deBronkart
 
How predictive analytics can help find the rare disease patient
How predictive analytics can help find the rare disease patientHow predictive analytics can help find the rare disease patient
How predictive analytics can help find the rare disease patient
IMSHealthRWES
 
R ppt.pdf
R ppt.pdfR ppt.pdf
Augmented Personalized Healthcare: Panel Version
Augmented Personalized Healthcare: Panel VersionAugmented Personalized Healthcare: Panel Version
Augmented Personalized Healthcare: Panel Version
Artificial Intelligence Institute at UofSC
 

Similar to AI for healthcare: Scaling Access and Quality of Care for Everyone (20)

Big Data Provides Opportunities, Challenges and a Better Future in Health and...
Big Data Provides Opportunities, Challenges and a Better Future in Health and...Big Data Provides Opportunities, Challenges and a Better Future in Health and...
Big Data Provides Opportunities, Challenges and a Better Future in Health and...
 
2016 Healthcare Trends
2016 Healthcare Trends2016 Healthcare Trends
2016 Healthcare Trends
 
What Do Patients Really Want Out Of Adherence Technology?
What Do Patients Really Want Out Of Adherence Technology?What Do Patients Really Want Out Of Adherence Technology?
What Do Patients Really Want Out Of Adherence Technology?
 
Power to the Patient
Power to the PatientPower to the Patient
Power to the Patient
 
How to Use Data to Improve Patient Safety: A Two-Part Discussion
How to Use Data to Improve Patient Safety: A Two-Part DiscussionHow to Use Data to Improve Patient Safety: A Two-Part Discussion
How to Use Data to Improve Patient Safety: A Two-Part Discussion
 
There Is A 90% Probability That Your Son Is Pregnant: Predicting The Future ...
There Is A 90% Probability That Your Son Is Pregnant:  Predicting The Future ...There Is A 90% Probability That Your Son Is Pregnant:  Predicting The Future ...
There Is A 90% Probability That Your Son Is Pregnant: Predicting The Future ...
 
AI in Healthcare.pptx
AI in Healthcare.pptxAI in Healthcare.pptx
AI in Healthcare.pptx
 
Wearable Health, Fitness Trackers, and the Quantified Self
Wearable Health, Fitness Trackers, and the Quantified SelfWearable Health, Fitness Trackers, and the Quantified Self
Wearable Health, Fitness Trackers, and the Quantified Self
 
Meeting healthcare challenges: what are the challenges and what is the role o...
Meeting healthcare challenges: what are the challenges and what is the role o...Meeting healthcare challenges: what are the challenges and what is the role o...
Meeting healthcare challenges: what are the challenges and what is the role o...
 
Patient experience: where can we improve?
Patient experience: where can we improve?Patient experience: where can we improve?
Patient experience: where can we improve?
 
20130226 impact van zorg 2 0 op onze samenleving
20130226 impact van zorg 2 0 op onze samenleving20130226 impact van zorg 2 0 op onze samenleving
20130226 impact van zorg 2 0 op onze samenleving
 
20130226 impact van zorg 2 0 op onze samenleving
20130226 impact van zorg 2 0 op onze samenleving20130226 impact van zorg 2 0 op onze samenleving
20130226 impact van zorg 2 0 op onze samenleving
 
Big Data and the Promise and Pitfalls when Applied to Disease Prevention and ...
Big Data and the Promise and Pitfalls when Applied to Disease Prevention and ...Big Data and the Promise and Pitfalls when Applied to Disease Prevention and ...
Big Data and the Promise and Pitfalls when Applied to Disease Prevention and ...
 
"Think, think, think": e-Patient Dave address to ECPC annual meeting
"Think, think, think": e-Patient Dave address to ECPC annual meeting"Think, think, think": e-Patient Dave address to ECPC annual meeting
"Think, think, think": e-Patient Dave address to ECPC annual meeting
 
MedicalResearch.com: Medical Research Exclusive Interviews December 31 2014
MedicalResearch.com:  Medical Research Exclusive Interviews December 31 2014MedicalResearch.com:  Medical Research Exclusive Interviews December 31 2014
MedicalResearch.com: Medical Research Exclusive Interviews December 31 2014
 
How AI is transforming Mobile Health
How AI is transforming Mobile HealthHow AI is transforming Mobile Health
How AI is transforming Mobile Health
 
Superpatients webinar for Natl Network of Libraries of Medicine
Superpatients webinar for Natl Network of Libraries of MedicineSuperpatients webinar for Natl Network of Libraries of Medicine
Superpatients webinar for Natl Network of Libraries of Medicine
 
How predictive analytics can help find the rare disease patient
How predictive analytics can help find the rare disease patientHow predictive analytics can help find the rare disease patient
How predictive analytics can help find the rare disease patient
 
R ppt.pdf
R ppt.pdfR ppt.pdf
R ppt.pdf
 
Augmented Personalized Healthcare: Panel Version
Augmented Personalized Healthcare: Panel VersionAugmented Personalized Healthcare: Panel Version
Augmented Personalized Healthcare: Panel Version
 

More from Xavier Amatriain

Data/AI driven product development: from video streaming to telehealth
Data/AI driven product development: from video streaming to telehealthData/AI driven product development: from video streaming to telehealth
Data/AI driven product development: from video streaming to telehealth
Xavier Amatriain
 
AI-driven product innovation: from Recommender Systems to COVID-19
AI-driven product innovation: from Recommender Systems to COVID-19AI-driven product innovation: from Recommender Systems to COVID-19
AI-driven product innovation: from Recommender Systems to COVID-19
Xavier Amatriain
 
Lessons learned from building practical deep learning systems
Lessons learned from building practical deep learning systemsLessons learned from building practical deep learning systems
Lessons learned from building practical deep learning systems
Xavier Amatriain
 
From one to zero: Going smaller as a growth strategy
From one to zero: Going smaller as a growth strategyFrom one to zero: Going smaller as a growth strategy
From one to zero: Going smaller as a growth strategy
Xavier Amatriain
 
Recommender Systems In Industry
Recommender Systems In IndustryRecommender Systems In Industry
Recommender Systems In Industry
Xavier Amatriain
 
Recsys 2016 tutorial: Lessons learned from building real-life recommender sys...
Recsys 2016 tutorial: Lessons learned from building real-life recommender sys...Recsys 2016 tutorial: Lessons learned from building real-life recommender sys...
Recsys 2016 tutorial: Lessons learned from building real-life recommender sys...
Xavier Amatriain
 
Past present and future of Recommender Systems: an Industry Perspective
Past present and future of Recommender Systems: an Industry PerspectivePast present and future of Recommender Systems: an Industry Perspective
Past present and future of Recommender Systems: an Industry Perspective
Xavier Amatriain
 
Staying Shallow & Lean in a Deep Learning World
Staying Shallow & Lean in a Deep Learning WorldStaying Shallow & Lean in a Deep Learning World
Staying Shallow & Lean in a Deep Learning World
Xavier Amatriain
 
Machine Learning for Q&A Sites: The Quora Example
Machine Learning for Q&A Sites: The Quora ExampleMachine Learning for Q&A Sites: The Quora Example
Machine Learning for Q&A Sites: The Quora Example
Xavier Amatriain
 
BIG2016- Lessons Learned from building real-life user-focused Big Data systems
BIG2016- Lessons Learned from building real-life user-focused Big Data systemsBIG2016- Lessons Learned from building real-life user-focused Big Data systems
BIG2016- Lessons Learned from building real-life user-focused Big Data systems
Xavier Amatriain
 
Strata 2016 - Lessons Learned from building real-life Machine Learning Systems
Strata 2016 -  Lessons Learned from building real-life Machine Learning SystemsStrata 2016 -  Lessons Learned from building real-life Machine Learning Systems
Strata 2016 - Lessons Learned from building real-life Machine Learning Systems
Xavier Amatriain
 
Past, present, and future of Recommender Systems: an industry perspective
Past, present, and future of Recommender Systems: an industry perspectivePast, present, and future of Recommender Systems: an industry perspective
Past, present, and future of Recommender Systems: an industry perspective
Xavier Amatriain
 
Barcelona ML Meetup - Lessons Learned
Barcelona ML Meetup - Lessons LearnedBarcelona ML Meetup - Lessons Learned
Barcelona ML Meetup - Lessons Learned
Xavier Amatriain
 
10 more lessons learned from building Machine Learning systems - MLConf
10 more lessons learned from building Machine Learning systems - MLConf10 more lessons learned from building Machine Learning systems - MLConf
10 more lessons learned from building Machine Learning systems - MLConf
Xavier Amatriain
 
10 more lessons learned from building Machine Learning systems
10 more lessons learned from building Machine Learning systems10 more lessons learned from building Machine Learning systems
10 more lessons learned from building Machine Learning systems
Xavier Amatriain
 
Machine Learning to Grow the World's Knowledge
Machine Learning to Grow  the World's KnowledgeMachine Learning to Grow  the World's Knowledge
Machine Learning to Grow the World's Knowledge
Xavier Amatriain
 
MLConf Seattle 2015 - ML@Quora
MLConf Seattle 2015 - ML@QuoraMLConf Seattle 2015 - ML@Quora
MLConf Seattle 2015 - ML@Quora
Xavier Amatriain
 
Lean DevOps - Lessons Learned from Innovation-driven Companies
Lean DevOps - Lessons Learned from Innovation-driven CompaniesLean DevOps - Lessons Learned from Innovation-driven Companies
Lean DevOps - Lessons Learned from Innovation-driven Companies
Xavier Amatriain
 
10 Lessons Learned from Building Machine Learning Systems
10 Lessons Learned from Building Machine Learning Systems10 Lessons Learned from Building Machine Learning Systems
10 Lessons Learned from Building Machine Learning Systems
Xavier Amatriain
 
Recsys 2014 Tutorial - The Recommender Problem Revisited
Recsys 2014 Tutorial - The Recommender Problem RevisitedRecsys 2014 Tutorial - The Recommender Problem Revisited
Recsys 2014 Tutorial - The Recommender Problem Revisited
Xavier Amatriain
 

More from Xavier Amatriain (20)

Data/AI driven product development: from video streaming to telehealth
Data/AI driven product development: from video streaming to telehealthData/AI driven product development: from video streaming to telehealth
Data/AI driven product development: from video streaming to telehealth
 
AI-driven product innovation: from Recommender Systems to COVID-19
AI-driven product innovation: from Recommender Systems to COVID-19AI-driven product innovation: from Recommender Systems to COVID-19
AI-driven product innovation: from Recommender Systems to COVID-19
 
Lessons learned from building practical deep learning systems
Lessons learned from building practical deep learning systemsLessons learned from building practical deep learning systems
Lessons learned from building practical deep learning systems
 
From one to zero: Going smaller as a growth strategy
From one to zero: Going smaller as a growth strategyFrom one to zero: Going smaller as a growth strategy
From one to zero: Going smaller as a growth strategy
 
Recommender Systems In Industry
Recommender Systems In IndustryRecommender Systems In Industry
Recommender Systems In Industry
 
Recsys 2016 tutorial: Lessons learned from building real-life recommender sys...
Recsys 2016 tutorial: Lessons learned from building real-life recommender sys...Recsys 2016 tutorial: Lessons learned from building real-life recommender sys...
Recsys 2016 tutorial: Lessons learned from building real-life recommender sys...
 
Past present and future of Recommender Systems: an Industry Perspective
Past present and future of Recommender Systems: an Industry PerspectivePast present and future of Recommender Systems: an Industry Perspective
Past present and future of Recommender Systems: an Industry Perspective
 
Staying Shallow & Lean in a Deep Learning World
Staying Shallow & Lean in a Deep Learning WorldStaying Shallow & Lean in a Deep Learning World
Staying Shallow & Lean in a Deep Learning World
 
Machine Learning for Q&A Sites: The Quora Example
Machine Learning for Q&A Sites: The Quora ExampleMachine Learning for Q&A Sites: The Quora Example
Machine Learning for Q&A Sites: The Quora Example
 
BIG2016- Lessons Learned from building real-life user-focused Big Data systems
BIG2016- Lessons Learned from building real-life user-focused Big Data systemsBIG2016- Lessons Learned from building real-life user-focused Big Data systems
BIG2016- Lessons Learned from building real-life user-focused Big Data systems
 
Strata 2016 - Lessons Learned from building real-life Machine Learning Systems
Strata 2016 -  Lessons Learned from building real-life Machine Learning SystemsStrata 2016 -  Lessons Learned from building real-life Machine Learning Systems
Strata 2016 - Lessons Learned from building real-life Machine Learning Systems
 
Past, present, and future of Recommender Systems: an industry perspective
Past, present, and future of Recommender Systems: an industry perspectivePast, present, and future of Recommender Systems: an industry perspective
Past, present, and future of Recommender Systems: an industry perspective
 
Barcelona ML Meetup - Lessons Learned
Barcelona ML Meetup - Lessons LearnedBarcelona ML Meetup - Lessons Learned
Barcelona ML Meetup - Lessons Learned
 
10 more lessons learned from building Machine Learning systems - MLConf
10 more lessons learned from building Machine Learning systems - MLConf10 more lessons learned from building Machine Learning systems - MLConf
10 more lessons learned from building Machine Learning systems - MLConf
 
10 more lessons learned from building Machine Learning systems
10 more lessons learned from building Machine Learning systems10 more lessons learned from building Machine Learning systems
10 more lessons learned from building Machine Learning systems
 
Machine Learning to Grow the World's Knowledge
Machine Learning to Grow  the World's KnowledgeMachine Learning to Grow  the World's Knowledge
Machine Learning to Grow the World's Knowledge
 
MLConf Seattle 2015 - ML@Quora
MLConf Seattle 2015 - ML@QuoraMLConf Seattle 2015 - ML@Quora
MLConf Seattle 2015 - ML@Quora
 
Lean DevOps - Lessons Learned from Innovation-driven Companies
Lean DevOps - Lessons Learned from Innovation-driven CompaniesLean DevOps - Lessons Learned from Innovation-driven Companies
Lean DevOps - Lessons Learned from Innovation-driven Companies
 
10 Lessons Learned from Building Machine Learning Systems
10 Lessons Learned from Building Machine Learning Systems10 Lessons Learned from Building Machine Learning Systems
10 Lessons Learned from Building Machine Learning Systems
 
Recsys 2014 Tutorial - The Recommender Problem Revisited
Recsys 2014 Tutorial - The Recommender Problem RevisitedRecsys 2014 Tutorial - The Recommender Problem Revisited
Recsys 2014 Tutorial - The Recommender Problem Revisited
 

Recently uploaded

QA or the Highway - Component Testing: Bridging the gap between frontend appl...
QA or the Highway - Component Testing: Bridging the gap between frontend appl...QA or the Highway - Component Testing: Bridging the gap between frontend appl...
QA or the Highway - Component Testing: Bridging the gap between frontend appl...
zjhamm304
 
"What does it really mean for your system to be available, or how to define w...
"What does it really mean for your system to be available, or how to define w..."What does it really mean for your system to be available, or how to define w...
"What does it really mean for your system to be available, or how to define w...
Fwdays
 
Getting the Most Out of ScyllaDB Monitoring: ShareChat's Tips
Getting the Most Out of ScyllaDB Monitoring: ShareChat's TipsGetting the Most Out of ScyllaDB Monitoring: ShareChat's Tips
Getting the Most Out of ScyllaDB Monitoring: ShareChat's Tips
ScyllaDB
 
Christine's Supplier Sourcing Presentaion.pptx
Christine's Supplier Sourcing Presentaion.pptxChristine's Supplier Sourcing Presentaion.pptx
Christine's Supplier Sourcing Presentaion.pptx
christinelarrosa
 
Essentials of Automations: Exploring Attributes & Automation Parameters
Essentials of Automations: Exploring Attributes & Automation ParametersEssentials of Automations: Exploring Attributes & Automation Parameters
Essentials of Automations: Exploring Attributes & Automation Parameters
Safe Software
 
Session 1 - Intro to Robotic Process Automation.pdf
Session 1 - Intro to Robotic Process Automation.pdfSession 1 - Intro to Robotic Process Automation.pdf
Session 1 - Intro to Robotic Process Automation.pdf
UiPathCommunity
 
AWS Certified Solutions Architect Associate (SAA-C03)
AWS Certified Solutions Architect Associate (SAA-C03)AWS Certified Solutions Architect Associate (SAA-C03)
AWS Certified Solutions Architect Associate (SAA-C03)
HarpalGohil4
 
[OReilly Superstream] Occupy the Space: A grassroots guide to engineering (an...
[OReilly Superstream] Occupy the Space: A grassroots guide to engineering (an...[OReilly Superstream] Occupy the Space: A grassroots guide to engineering (an...
[OReilly Superstream] Occupy the Space: A grassroots guide to engineering (an...
Jason Yip
 
Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |
AstuteBusiness
 
Poznań ACE event - 19.06.2024 Team 24 Wrapup slidedeck
Poznań ACE event - 19.06.2024 Team 24 Wrapup slidedeckPoznań ACE event - 19.06.2024 Team 24 Wrapup slidedeck
Poznań ACE event - 19.06.2024 Team 24 Wrapup slidedeck
FilipTomaszewski5
 
AppSec PNW: Android and iOS Application Security with MobSF
AppSec PNW: Android and iOS Application Security with MobSFAppSec PNW: Android and iOS Application Security with MobSF
AppSec PNW: Android and iOS Application Security with MobSF
Ajin Abraham
 
GlobalLogic Java Community Webinar #18 “How to Improve Web Application Perfor...
GlobalLogic Java Community Webinar #18 “How to Improve Web Application Perfor...GlobalLogic Java Community Webinar #18 “How to Improve Web Application Perfor...
GlobalLogic Java Community Webinar #18 “How to Improve Web Application Perfor...
GlobalLogic Ukraine
 
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and BioinformaticiansBiomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Neo4j
 
GNSS spoofing via SDR (Criptored Talks 2024)
GNSS spoofing via SDR (Criptored Talks 2024)GNSS spoofing via SDR (Criptored Talks 2024)
GNSS spoofing via SDR (Criptored Talks 2024)
Javier Junquera
 
Apps Break Data
Apps Break DataApps Break Data
Apps Break Data
Ivo Velitchkov
 
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectorsConnector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
DianaGray10
 
Containers & AI - Beauty and the Beast!?!
Containers & AI - Beauty and the Beast!?!Containers & AI - Beauty and the Beast!?!
Containers & AI - Beauty and the Beast!?!
Tobias Schneck
 
"Frontline Battles with DDoS: Best practices and Lessons Learned", Igor Ivaniuk
"Frontline Battles with DDoS: Best practices and Lessons Learned",  Igor Ivaniuk"Frontline Battles with DDoS: Best practices and Lessons Learned",  Igor Ivaniuk
"Frontline Battles with DDoS: Best practices and Lessons Learned", Igor Ivaniuk
Fwdays
 
Northern Engraving | Nameplate Manufacturing Process - 2024
Northern Engraving | Nameplate Manufacturing Process - 2024Northern Engraving | Nameplate Manufacturing Process - 2024
Northern Engraving | Nameplate Manufacturing Process - 2024
Northern Engraving
 
"NATO Hackathon Winner: AI-Powered Drug Search", Taras Kloba
"NATO Hackathon Winner: AI-Powered Drug Search",  Taras Kloba"NATO Hackathon Winner: AI-Powered Drug Search",  Taras Kloba
"NATO Hackathon Winner: AI-Powered Drug Search", Taras Kloba
Fwdays
 

Recently uploaded (20)

QA or the Highway - Component Testing: Bridging the gap between frontend appl...
QA or the Highway - Component Testing: Bridging the gap between frontend appl...QA or the Highway - Component Testing: Bridging the gap between frontend appl...
QA or the Highway - Component Testing: Bridging the gap between frontend appl...
 
"What does it really mean for your system to be available, or how to define w...
"What does it really mean for your system to be available, or how to define w..."What does it really mean for your system to be available, or how to define w...
"What does it really mean for your system to be available, or how to define w...
 
Getting the Most Out of ScyllaDB Monitoring: ShareChat's Tips
Getting the Most Out of ScyllaDB Monitoring: ShareChat's TipsGetting the Most Out of ScyllaDB Monitoring: ShareChat's Tips
Getting the Most Out of ScyllaDB Monitoring: ShareChat's Tips
 
Christine's Supplier Sourcing Presentaion.pptx
Christine's Supplier Sourcing Presentaion.pptxChristine's Supplier Sourcing Presentaion.pptx
Christine's Supplier Sourcing Presentaion.pptx
 
Essentials of Automations: Exploring Attributes & Automation Parameters
Essentials of Automations: Exploring Attributes & Automation ParametersEssentials of Automations: Exploring Attributes & Automation Parameters
Essentials of Automations: Exploring Attributes & Automation Parameters
 
Session 1 - Intro to Robotic Process Automation.pdf
Session 1 - Intro to Robotic Process Automation.pdfSession 1 - Intro to Robotic Process Automation.pdf
Session 1 - Intro to Robotic Process Automation.pdf
 
AWS Certified Solutions Architect Associate (SAA-C03)
AWS Certified Solutions Architect Associate (SAA-C03)AWS Certified Solutions Architect Associate (SAA-C03)
AWS Certified Solutions Architect Associate (SAA-C03)
 
[OReilly Superstream] Occupy the Space: A grassroots guide to engineering (an...
[OReilly Superstream] Occupy the Space: A grassroots guide to engineering (an...[OReilly Superstream] Occupy the Space: A grassroots guide to engineering (an...
[OReilly Superstream] Occupy the Space: A grassroots guide to engineering (an...
 
Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |Astute Business Solutions | Oracle Cloud Partner |
Astute Business Solutions | Oracle Cloud Partner |
 
Poznań ACE event - 19.06.2024 Team 24 Wrapup slidedeck
Poznań ACE event - 19.06.2024 Team 24 Wrapup slidedeckPoznań ACE event - 19.06.2024 Team 24 Wrapup slidedeck
Poznań ACE event - 19.06.2024 Team 24 Wrapup slidedeck
 
AppSec PNW: Android and iOS Application Security with MobSF
AppSec PNW: Android and iOS Application Security with MobSFAppSec PNW: Android and iOS Application Security with MobSF
AppSec PNW: Android and iOS Application Security with MobSF
 
GlobalLogic Java Community Webinar #18 “How to Improve Web Application Perfor...
GlobalLogic Java Community Webinar #18 “How to Improve Web Application Perfor...GlobalLogic Java Community Webinar #18 “How to Improve Web Application Perfor...
GlobalLogic Java Community Webinar #18 “How to Improve Web Application Perfor...
 
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and BioinformaticiansBiomedical Knowledge Graphs for Data Scientists and Bioinformaticians
Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
 
GNSS spoofing via SDR (Criptored Talks 2024)
GNSS spoofing via SDR (Criptored Talks 2024)GNSS spoofing via SDR (Criptored Talks 2024)
GNSS spoofing via SDR (Criptored Talks 2024)
 
Apps Break Data
Apps Break DataApps Break Data
Apps Break Data
 
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectorsConnector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
Connector Corner: Seamlessly power UiPath Apps, GenAI with prebuilt connectors
 
Containers & AI - Beauty and the Beast!?!
Containers & AI - Beauty and the Beast!?!Containers & AI - Beauty and the Beast!?!
Containers & AI - Beauty and the Beast!?!
 
"Frontline Battles with DDoS: Best practices and Lessons Learned", Igor Ivaniuk
"Frontline Battles with DDoS: Best practices and Lessons Learned",  Igor Ivaniuk"Frontline Battles with DDoS: Best practices and Lessons Learned",  Igor Ivaniuk
"Frontline Battles with DDoS: Best practices and Lessons Learned", Igor Ivaniuk
 
Northern Engraving | Nameplate Manufacturing Process - 2024
Northern Engraving | Nameplate Manufacturing Process - 2024Northern Engraving | Nameplate Manufacturing Process - 2024
Northern Engraving | Nameplate Manufacturing Process - 2024
 
"NATO Hackathon Winner: AI-Powered Drug Search", Taras Kloba
"NATO Hackathon Winner: AI-Powered Drug Search",  Taras Kloba"NATO Hackathon Winner: AI-Powered Drug Search",  Taras Kloba
"NATO Hackathon Winner: AI-Powered Drug Search", Taras Kloba
 

AI for healthcare: Scaling Access and Quality of Care for Everyone

  • 1. AI for healthcare: Scaling access and quality of care for everyone Anitha Kannan Xavier Amatriain MLConf 10/08/2019
  • 2. ● >50% world’s population with no access to essential health services ● US… ○ 10% of adult population has no health insurance ○ 28% of working adults are under insured Healthcare access is a major issue Kaiser Family Foundation analysis of the 2017 National Health Interview Survey Merrit Hawkins, 2017 survey shortage of 120,000 physicians by 2030
  • 3. Patient-Doctor interaction ● Doctors have ~15 minutes to capture pertinent information about a patient, diagnose + recommend treatment ● 30% of the medical errors causing ~400k deaths a year are due to misdiagnosis 2069 doctors solve 1572 HumanDx cases
  • 4. Online search and/or Healthcare access? “72% of internet users say they looked online for health information within the past year” “More than ⅓ use Internet to self-diagnose” [Pew Research] 1.4M daily 25M daily Need more than Google can deliver Less cost and friction than PCP visit
  • 5. We have an opportunity to reimagine healthcare
  • 6. We have an obligation opportunity to reimagine healthcare
  • 7. Looking Forward: Towards AI powered Learning Health Systems ● AI + human practitioners for Quality Care ● Less than 20% of the cost for best healthcare access ● Mobile First Care, 24/7 always on
  • 8. What are we doing? ● Mission: Provide the world's best healthcare for everyone ● Product: User-facing mobile primary care app ● Team: Building an awesome and diverse team ● Approach: State-of-the-art AI/ML + product/UX/clinical AI-based interaction AI + Health coaches AI + Doctors
  • 9. Breakthroughs in AI & healthcare
  • 11. AI “in the wild”: Learning health systems
  • 13. AI in the wild: Desired properties ● Easily extensible ○ Incrementally/iteratively learn from “physician-in-the-loop” or from additional data ● Knows what it does not know ○ Models uncertainty in prediction ○ Enables fall-back to “physician-in-the-loop”
  • 15. AI for assisted diagnosis (since 1980s) ● Expert systems ○ Mycin, Internist-1, DxPlain, VDDx, QMR ● Covers over 1000 diseases and 3500+ findings ○ Most comprehensive diagnosis model, so far ○ 30+ years of expert curation based on research and evidence-based literature
  • 16. Expert systems in the wild? ● Not easy to extend ○ Costly, time consuming and time-delayed ○ Poor generalization to new places ● Does not know what “it doesn’t know” ○ Constrained to diseases in the system
  • 17. Assisted diagnosis in the wild 1. Extensibility a. Diagnosis as a ML task i. Expert systems as a prior b. Modeling less prevalent diseases i. Low-shot learning 2. Knowing what you don’t know a. Measures of uncertainty in prediction b. Allows fall-back to “physician-in-the-loop”
  • 18. Assisted diagnosis in the wild 1. Extensibility a. Diagnosis as a ML task i. Expert systems as a prior b. Modeling less prevalent diseases i. Low-shot learning 2. Knowing what you don’t know a. Measures of uncertainty in prediction b. Allows fall-back to “physician-in-the-loop”
  • 19. x Clinical case simulator Example of simulated case Knowledge base central to expert systems Expert systems as prior
  • 20. ML models for diagnosis clinical cases simulated from expert system From expert systems to ML model for diagnosis
  • 21. ML models for diagnosis clinical cases simulated from expert system From expert systems to ML model for diagnosis clinical cases from other sources eg. electronic health records
  • 22. Assisted diagnosis in the wild 1. Extensibility a. Diagnosis as a ML task i. Expert systems as a prior b. Modeling less prevalent diseases i. Low-shot learning 2. Knowing what you don’t know a. Measures of uncertainty in prediction b. Allows fall-back to “physician-in-the-loop”
  • 23. Assisted diagnosis in the wild 1. Extensibility a. Diagnosis as a ML task i. Expert systems as a prior b. Modeling less prevalent diseases i. Low-shot learning 2. Knowing what you don’t know a. Measures of uncertainty in prediction b. Allows fall-back to “physician-in-the-loop”
  • 25. Open-Set diagnosis Universe of diseases Amblyopia Diabetic Ophthalmoplegia Gastroenteritis is aware and avoid misclassifying unknown diseases as known Diseases within diagnostic scope
  • 26. Open-Set diagnosis Universe of diseases Amblyopia Diabetic Ophthalmoplegia Extra diseases Gastroenteritis avoids misclassifying unknown diseases as known. Diseases within diagnostic scope Entropic open-set loss: Maximize predictive entropy of unseen examples
  • 27. AI in-the-wild: Desired properties ● Easily extensible ○ Incrementally/iteratively learn from “physician-in-the-loop” or from additional data ● Knows what it does not know ○ Models uncertainty in prediction ○ Enables fall-back to “physician-in-the-loop”
  • 28. Medical Information gathering “in-the-wild” Real users with health issues that an AI medical agent may not understand
  • 29. Looking Forward... ● AI + human practitioners for Quality Care ● Less than 20% of the cost for best healthcare access ● Mobile First Care, 24/7 always on AI-based interaction AI + Health coaches AI + Doctors https://firstopinionapp.com/ 39.6 M Californians with access to high quality affordable primary care