SlideShare ist ein Scribd-Unternehmen logo
1 von 33
Downloaden Sie, um offline zu lesen
Towards Online Universal Quality
Healthcare through AI
Xavier Amatriain (@xamat)
xavier@curai.com
Being a physician is hard(er)
● Doctors have ~15 minutes to capture
information* about a patient, diagnose,
+ recommend treatment
● *Information
○ Patient’s history
○ Patient’s symptoms
○ Medical knowledge
■ Learned years ago
■ Latest research findings (70+ journal
articles per day)
■ Different demographics
● Data is growing over time, so is complexity
● Manual adaptation is challenging
Being a physician is hard(er) (and tech hasn’t helped!)
Cost of medical errors
● 12M misdiagnoses/y
● Errors cause 400k
deaths & 4M serious
health events
○ Compare to 500k deaths from
cancer or 40k from vehicle
accidents
● Almost half of those
are preventable
Online search and/or Healthcare access?
“72% of internet users say they
looked online for health
information within the past year”
[Pew Research]
1.4M daily
25M daily
Need more than Google can
deliver
Less cost and friction
than PCP visit
We can do better than Google + Webmd
“Roughly 1 percent of searches on
Google are symptom-related”
[Google]
We have an opportunity to
reimagine healthcare
We have an opportunity
obligation to reimagine
healthcare
Towards Online Universal Quality Healthcare
● Online= mobile, always on
● Universal = scalable, low cost, easy to access/use
● Quality = as good as the best doctors, based on
best-known practices and scientific evidence, replicable
● Healthcare != URL. Healthcare = actionable
recommendations + directed access to resources
including human professionals when needed
How?
● Online + physicians in the loop = telehealth
● However, standard telehealth is (1) hard to scale, (2) not better quality
Merrit Hawkins, 2017 survey
shortage of 120,000 physicians by 2030
Barnett et.al. Comparative Accuracy of Diagnosis by Collective Intelligence of
Multiple Physicians vs Individual Physicians JAMA, 2018
How?
Scale + Quality =
Automation (aka AI)
AI-based interaction
AI + Registered Nurse (RN)
AI + Nurse Practitioners
AI + Doctors
Part II.
Medical AI for online Healthcare
Medical AI for Online Healthcare
1. Medical knowledge extraction/representation
2. Conversational Healthcare Systems
3. Automated Triage/Diagnosis/Treatment
4. Multimodal inputs
Part II.
1. Knowledge
Extraction & Representation
Medical Knowledge extraction
● Medical ontologies
● Electronic access to medical
research
● Access to Electronic Health
Records
Knowledge Representation: Medical ontologies
● Snomed Clinical Terms
○ collection of medical terms used in clinical
documentation and reporting.
○ clinical findings, symptoms, diagnoses, procedures,
body structures, organisms substances,
pharmaceuticals, devices…
● UMLS
○ Compendium of many controlled vocabularies
○ Enables translating between terminology systems
● ICD-10
○ International Statistical Classification of Diseases
and Related Health Problems
Health knowledge graphs
Part II.
2. Conversational Healthcare
Understanding patients and doctors
Learning a medically-aware dialog system
Part II.
3. Automated(Assisted) Triage/Dx/Rx
Medical Diagnosis
● Diagnosis (R.A. Miller 1990):
○ Mapping from patient’s data
(history, examination, lab
exams…) to a possible condition.
○ It depends on ability to:
■ Evoke history
■ Surface symptoms and
findings
■ Generate hypotheses that
suggest how to refine or
pursue different hypothesis
○ In a compassionate,
cost-effective manner
An example: Internist-1/QMR/Vddx
● Internist (1971) - Dr. Jack
Myers
● Process for adding a
disease requires 2-4
weeks of full-time effort
and doctors reading 50 to
250 relevant publications
ML/AI Approaches to Diagnosis
● Early DDSS based on Bayesian reasoning (60s-70s)
● Bayesian networks (80s-90s)
● Neural networks (lately)
x
Clinical case simulator
Example of
simulated case
Knowledge
base
Our approach: Expert systems as Prior
Machine learned models
simulated cases
clinical cases simulated
from expert systems
clinical cases other sources
eg. electronic health records
ML classification models for differential diagnosis
Our approach: Incorporating data from EHR
Part II.
4. Multimodality
Multimodal Inputs
Conclusions
Recap
● More people access healthcare through search than PCPs
● Doctors don’t have time and make mistakes
● We should be able to offer a better online experience than Google + Webmd
● Online Universal Quality Healthcare
○ Online, mobile, always on
○ Universal = scalable, low cost, easy to access/use (i.e conversational)
○ Quality = as good as the best doctors, based on best-known practices and scientific evidence,
replicable
○ Healthcare = not only a website you need to parse and figure out on your own, but directed
access to healthcare resources including human professionals when needed
● How?
○ Online + physicians in the loop = telehealth
○ Standard telehealth is (1) hard to scale, (2) not better quality
○ Scale + Quality = automation (aka AI)
References
● “Probabilistic diagnosis using a reformulation of the INTERNIST-1/QMR knowledge base” . Shwe et al. 1991.
● “Computer-assisted diagnostic decision support: history, challenges, and possible paths forward” Miller. 2009.
● “Mining Biomedical Ontologies and Data Using RDF Hypergraphs” Liu et al. 2013.
● “A ‘Green Button’ For Using Aggregate Patient Data At The Point Of Care” Longhurst et al. 2014.
● “Building the graph of medicine from millions of clinical narratives” Finlayson et al. 2014.
● “Comparison of Physician and Computer Diagnostic Accuracy” Semigran et al. 2016.
● “Clinical Tagging with Joint Probabilistic Models” . Halpern et al. 2016.
● “Deep Patient: An Unsupervised Representation to Predict the Future of Patients from EHR”. Miotto et al. 2016.
● “Learning a Health Knowledge Graph from Electronic Medical Records” Rotmensch et al. 2017.
● “Clustering Patients with Tensor Decomposition”. Ruffini et al. 2017.
● “Patient Similarity Using Population Statistics and Multiple Kernel Learning”. Conroy et al. 2017.
● “Diagnostic Inferencing via Clinical Concept Extraction with Deep Reinforcement Learning”. Ling et al. 2017.
● “Generating Multi-label Discrete Patient Records using Generative Adversarial Networks” Choi et al. 2017
● “The Use of Autoencoders for Discovering Patient Phenotypes”. arXiv.org. Suresh, H., et al. 2017.
● “Comparative Accuracy of Diagnosis by Collective Intelligence of Multiple Physicians vs Individual Physicians”
Barnett et.al. 2018
● “Scalable and accurate deep learning with electronic health records”. Rajkomar et al. 2018
● “Learning from the experts: From expert systems to machine-learned diagnosis models”. Ravuri et al. 2018
● “Prototypical Clustering Networks for Dermatological Disease Diagnosis”. Prabhu et al. 2018
References
Towards online universal quality healthcare through AI

Weitere ähnliche Inhalte

Was ist angesagt?

Was ist angesagt? (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
 
Venkatesh_CV
Venkatesh_CVVenkatesh_CV
Venkatesh_CV
 
Computer science for health
Computer science for healthComputer science for health
Computer science for health
 
Healthcare analytics
Healthcare analytics Healthcare analytics
Healthcare analytics
 
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
 
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
 
Stroke Prediction
Stroke PredictionStroke Prediction
Stroke Prediction
 
Healthcare Predicitive Analytics for Risk Profiling in Chronic Care: A Bayesi...
Healthcare Predicitive Analytics for Risk Profiling in Chronic Care: A Bayesi...Healthcare Predicitive Analytics for Risk Profiling in Chronic Care: A Bayesi...
Healthcare Predicitive Analytics for Risk Profiling in Chronic Care: A Bayesi...
 
ICT in Healthcare
ICT in HealthcareICT in Healthcare
ICT in Healthcare
 
Alumni: The Ultimate Outcome Measure of your Training Program
Alumni: The Ultimate Outcome Measure of your Training ProgramAlumni: The Ultimate Outcome Measure of your Training Program
Alumni: The Ultimate Outcome Measure of your Training Program
 
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)
 
Predictive Analytics in Healthcare
Predictive Analytics in HealthcarePredictive Analytics in Healthcare
Predictive Analytics in Healthcare
 
Introduction to Health Informatics and Health Information Technology (Part 2)...
Introduction to Health Informatics and Health Information Technology (Part 2)...Introduction to Health Informatics and Health Information Technology (Part 2)...
Introduction to Health Informatics and Health Information Technology (Part 2)...
 
2015 CVDPG Trinity
2015 CVDPG Trinity2015 CVDPG Trinity
2015 CVDPG Trinity
 
Dr. Obumneke Amadi _Transcript
Dr. Obumneke Amadi _TranscriptDr. Obumneke Amadi _Transcript
Dr. Obumneke Amadi _Transcript
 
Open Educational Resources for Big Data Science
Open Educational Resources for Big Data ScienceOpen Educational Resources for Big Data Science
Open Educational Resources for Big Data Science
 
Towards Patient-Centred Care
Towards Patient-Centred CareTowards Patient-Centred Care
Towards Patient-Centred Care
 

Ähnlich wie Towards online universal quality healthcare through AI

Augmented Personalized Health: using AI techniques on semantically integrated...
Augmented Personalized Health: using AI techniques on semantically integrated...Augmented Personalized Health: using AI techniques on semantically integrated...
Augmented Personalized Health: using AI techniques on semantically integrated...
Amit Sheth
 
ICEGOV2009 - Tutorial 4 - E-Health Standards in Practice: Challenges and Oppo...
ICEGOV2009 - Tutorial 4 - E-Health Standards in Practice: Challenges and Oppo...ICEGOV2009 - Tutorial 4 - E-Health Standards in Practice: Challenges and Oppo...
ICEGOV2009 - Tutorial 4 - E-Health Standards in Practice: Challenges and Oppo...
ICEGOV
 

Ähnlich wie Towards online universal quality healthcare through AI (20)

Andrew Roberts - Mobile Health Apps for Improved Patient Engagement and Educa...
Andrew Roberts - Mobile Health Apps for Improved Patient Engagement and Educa...Andrew Roberts - Mobile Health Apps for Improved Patient Engagement and Educa...
Andrew Roberts - Mobile Health Apps for Improved Patient Engagement and Educa...
 
Day 1: Real-World Data Panel
Day 1: Real-World Data Panel Day 1: Real-World Data Panel
Day 1: Real-World Data Panel
 
Future of Healthcare: 3 Disruptive Trends
Future of Healthcare: 3 Disruptive TrendsFuture of Healthcare: 3 Disruptive Trends
Future of Healthcare: 3 Disruptive Trends
 
Applications of Data Science in Healthcare
Applications of Data Science in HealthcareApplications of Data Science in Healthcare
Applications of Data Science in Healthcare
 
Advancing-OSHMS High-Performance WS in OHM
Advancing-OSHMS High-Performance WS in OHMAdvancing-OSHMS High-Performance WS in OHM
Advancing-OSHMS High-Performance WS in OHM
 
Sun==big data analytics for health care
Sun==big data analytics for health careSun==big data analytics for health care
Sun==big data analytics for health care
 
Health Informatics- Module 1-Chapter 1.pptx
Health Informatics- Module 1-Chapter 1.pptxHealth Informatics- Module 1-Chapter 1.pptx
Health Informatics- Module 1-Chapter 1.pptx
 
Report from 2015 MedInfo Conference Brazil
Report from 2015 MedInfo Conference BrazilReport from 2015 MedInfo Conference Brazil
Report from 2015 MedInfo Conference Brazil
 
Augmented Personalized Health: using AI techniques on semantically integrated...
Augmented Personalized Health: using AI techniques on semantically integrated...Augmented Personalized Health: using AI techniques on semantically integrated...
Augmented Personalized Health: using AI techniques on semantically integrated...
 
Patient relationship management on the cloud
Patient relationship management on the cloudPatient relationship management on the cloud
Patient relationship management on the cloud
 
ICEGOV2009 - Tutorial 4 - E-Health Standards in Practice: Challenges and Oppo...
ICEGOV2009 - Tutorial 4 - E-Health Standards in Practice: Challenges and Oppo...ICEGOV2009 - Tutorial 4 - E-Health Standards in Practice: Challenges and Oppo...
ICEGOV2009 - Tutorial 4 - E-Health Standards in Practice: Challenges and Oppo...
 
1-s2.0-S0167923620300944-main.pdf
1-s2.0-S0167923620300944-main.pdf1-s2.0-S0167923620300944-main.pdf
1-s2.0-S0167923620300944-main.pdf
 
Is Big Data Always Good Data?
Is Big Data Always Good Data? Is Big Data Always Good Data?
Is Big Data Always Good Data?
 
The Learning Health System: Thinking and Acting Across Scales
The Learning Health System: Thinking and Acting Across ScalesThe Learning Health System: Thinking and Acting Across Scales
The Learning Health System: Thinking and Acting Across Scales
 
Person-generated health data: How can it help us to feel better?
Person-generated health data: How can it help us to feel better?Person-generated health data: How can it help us to feel better?
Person-generated health data: How can it help us to feel better?
 
Developing the Informatics Workforce for Scotland's Health and Social Care
Developing the Informatics Workforce for Scotland's Health and Social CareDeveloping the Informatics Workforce for Scotland's Health and Social Care
Developing the Informatics Workforce for Scotland's Health and Social Care
 
Presenting health data to patients
Presenting health data to patientsPresenting health data to patients
Presenting health data to patients
 
NLM Georgia Biomedical Informatics
NLM Georgia Biomedical InformaticsNLM Georgia Biomedical Informatics
NLM Georgia Biomedical Informatics
 
IBM Terkko Pop-up Presentation by Pekka Leppänen
IBM Terkko Pop-up Presentation by Pekka LeppänenIBM Terkko Pop-up Presentation by Pekka Leppänen
IBM Terkko Pop-up Presentation by Pekka Leppänen
 
Introduction to Nursing Informatics
Introduction to Nursing InformaticsIntroduction to Nursing Informatics
Introduction to Nursing Informatics
 

Mehr von Xavier Amatriain

Mehr von 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
 

Kürzlich hochgeladen

EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
Earley Information Science
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
Enterprise Knowledge
 

Kürzlich hochgeladen (20)

Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...
 

Towards online universal quality healthcare through AI

  • 1. Towards Online Universal Quality Healthcare through AI Xavier Amatriain (@xamat) xavier@curai.com
  • 2. Being a physician is hard(er) ● Doctors have ~15 minutes to capture information* about a patient, diagnose, + recommend treatment ● *Information ○ Patient’s history ○ Patient’s symptoms ○ Medical knowledge ■ Learned years ago ■ Latest research findings (70+ journal articles per day) ■ Different demographics ● Data is growing over time, so is complexity ● Manual adaptation is challenging
  • 3. Being a physician is hard(er) (and tech hasn’t helped!)
  • 4. Cost of medical errors ● 12M misdiagnoses/y ● Errors cause 400k deaths & 4M serious health events ○ Compare to 500k deaths from cancer or 40k from vehicle accidents ● Almost half of those are preventable
  • 5. Online search and/or Healthcare access? “72% of internet users say they looked online for health information within the past year” [Pew Research] 1.4M daily 25M daily Need more than Google can deliver Less cost and friction than PCP visit
  • 6. We can do better than Google + Webmd “Roughly 1 percent of searches on Google are symptom-related” [Google]
  • 7. We have an opportunity to reimagine healthcare
  • 8. We have an opportunity obligation to reimagine healthcare
  • 9. Towards Online Universal Quality Healthcare ● Online= mobile, always on ● Universal = scalable, low cost, easy to access/use ● Quality = as good as the best doctors, based on best-known practices and scientific evidence, replicable ● Healthcare != URL. Healthcare = actionable recommendations + directed access to resources including human professionals when needed
  • 10. How? ● Online + physicians in the loop = telehealth ● However, standard telehealth is (1) hard to scale, (2) not better quality Merrit Hawkins, 2017 survey shortage of 120,000 physicians by 2030 Barnett et.al. Comparative Accuracy of Diagnosis by Collective Intelligence of Multiple Physicians vs Individual Physicians JAMA, 2018
  • 11. How? Scale + Quality = Automation (aka AI) AI-based interaction AI + Registered Nurse (RN) AI + Nurse Practitioners AI + Doctors
  • 12. Part II. Medical AI for online Healthcare
  • 13. Medical AI for Online Healthcare 1. Medical knowledge extraction/representation 2. Conversational Healthcare Systems 3. Automated Triage/Diagnosis/Treatment 4. Multimodal inputs
  • 15. Medical Knowledge extraction ● Medical ontologies ● Electronic access to medical research ● Access to Electronic Health Records
  • 16. Knowledge Representation: Medical ontologies ● Snomed Clinical Terms ○ collection of medical terms used in clinical documentation and reporting. ○ clinical findings, symptoms, diagnoses, procedures, body structures, organisms substances, pharmaceuticals, devices… ● UMLS ○ Compendium of many controlled vocabularies ○ Enables translating between terminology systems ● ICD-10 ○ International Statistical Classification of Diseases and Related Health Problems
  • 20. Learning a medically-aware dialog system
  • 22. Medical Diagnosis ● Diagnosis (R.A. Miller 1990): ○ Mapping from patient’s data (history, examination, lab exams…) to a possible condition. ○ It depends on ability to: ■ Evoke history ■ Surface symptoms and findings ■ Generate hypotheses that suggest how to refine or pursue different hypothesis ○ In a compassionate, cost-effective manner
  • 23. An example: Internist-1/QMR/Vddx ● Internist (1971) - Dr. Jack Myers ● Process for adding a disease requires 2-4 weeks of full-time effort and doctors reading 50 to 250 relevant publications
  • 24. ML/AI Approaches to Diagnosis ● Early DDSS based on Bayesian reasoning (60s-70s) ● Bayesian networks (80s-90s) ● Neural networks (lately)
  • 25. x Clinical case simulator Example of simulated case Knowledge base Our approach: Expert systems as Prior Machine learned models simulated cases
  • 26. clinical cases simulated from expert systems clinical cases other sources eg. electronic health records ML classification models for differential diagnosis Our approach: Incorporating data from EHR
  • 30. Recap ● More people access healthcare through search than PCPs ● Doctors don’t have time and make mistakes ● We should be able to offer a better online experience than Google + Webmd ● Online Universal Quality Healthcare ○ Online, mobile, always on ○ Universal = scalable, low cost, easy to access/use (i.e conversational) ○ Quality = as good as the best doctors, based on best-known practices and scientific evidence, replicable ○ Healthcare = not only a website you need to parse and figure out on your own, but directed access to healthcare resources including human professionals when needed ● How? ○ Online + physicians in the loop = telehealth ○ Standard telehealth is (1) hard to scale, (2) not better quality ○ Scale + Quality = automation (aka AI)
  • 32. ● “Probabilistic diagnosis using a reformulation of the INTERNIST-1/QMR knowledge base” . Shwe et al. 1991. ● “Computer-assisted diagnostic decision support: history, challenges, and possible paths forward” Miller. 2009. ● “Mining Biomedical Ontologies and Data Using RDF Hypergraphs” Liu et al. 2013. ● “A ‘Green Button’ For Using Aggregate Patient Data At The Point Of Care” Longhurst et al. 2014. ● “Building the graph of medicine from millions of clinical narratives” Finlayson et al. 2014. ● “Comparison of Physician and Computer Diagnostic Accuracy” Semigran et al. 2016. ● “Clinical Tagging with Joint Probabilistic Models” . Halpern et al. 2016. ● “Deep Patient: An Unsupervised Representation to Predict the Future of Patients from EHR”. Miotto et al. 2016. ● “Learning a Health Knowledge Graph from Electronic Medical Records” Rotmensch et al. 2017. ● “Clustering Patients with Tensor Decomposition”. Ruffini et al. 2017. ● “Patient Similarity Using Population Statistics and Multiple Kernel Learning”. Conroy et al. 2017. ● “Diagnostic Inferencing via Clinical Concept Extraction with Deep Reinforcement Learning”. Ling et al. 2017. ● “Generating Multi-label Discrete Patient Records using Generative Adversarial Networks” Choi et al. 2017 ● “The Use of Autoencoders for Discovering Patient Phenotypes”. arXiv.org. Suresh, H., et al. 2017. ● “Comparative Accuracy of Diagnosis by Collective Intelligence of Multiple Physicians vs Individual Physicians” Barnett et.al. 2018 ● “Scalable and accurate deep learning with electronic health records”. Rajkomar et al. 2018 ● “Learning from the experts: From expert systems to machine-learned diagnosis models”. Ravuri et al. 2018 ● “Prototypical Clustering Networks for Dermatological Disease Diagnosis”. Prabhu et al. 2018 References