Sexy Call Girl Tiruvannamalai Arshi 💚9058824046💚 Tiruvannamalai Escort Service
Artificial Intelligence: Connecting the dots and relieving the pain in Health Care Delivery.
1. Artificial Intelligence: Connecting the dots
and relieving the pain in Health Care Delivery
JAI NAHAR, MD, MBA
Associate Professor of Pediatrics
George Washington University School of Medicine
Attending, Division of Cardiology
Children’s National Hospital, Washington DC
Jan.25th 2020
SAIMS
Indore, India
2. 1. Introductory concepts
2. History & Current State of AI
3. Potential uses in Augmenting Heath care delivery
4. Challenges in Adoption
5. Future Directions
Agenda
3. Why Talk about AI In Health Care?
• Fourth Industrial Revolution
• Digitized world
• Health Care Delivery Challenges
Resources
Access
Education
Care Gaps
Friction
5. AI : Definition
“Artificial Intelligence is the science of making Machines do
things that would require intelligence if done by men.”
Marvin Minsky
6. AI: General Concepts
Artificial
Intelligence AI
Machine
Learning
ML
Deep Learning
DL
AI: Engineering of making Intelligent machines and programs
ML: Ability of machines to learn without being explicitly programmed
DL: Learning based on Deep Neural Networks
10. Supervised Learning
• Goal is to predict a known output orTarget
• Algorithm is taught with right answers (labels) for examples used in
training data set.
• Can be categorized as:
Classification problem: predicting categories, discrete value output
(0,1 etc.)
Regression problem: predict continuous values
Anomaly detection problem: predict unusual pattern
11. Unsupervised Learning
• Goal is to learn the intrinsic structure within data.
• No outputs to predict
• Task is to find hidden pattern/structure in data,
without human feedback
Cluster2
Cluster3
Cluster
1
12. Reinforcement learning
Reinforcement learning led to AlphaGo’s stunning victory over a human Go
champion
https://www.technologyreview.com/s/603501/10-breakthrough-technologies-2017-reinforcement-
learning/
Reinforcement learning is
learning by trial-and-error,
solely from rewards or
punishments.
https://deepmind.com/blog/deep-reinforcement-
learning/
Can be viewed as hybrid
of supervised and
unsupervised learning
13. Artificial Neural Network (ANN)
ANN are modeled after human neurons
• Nodes are like neurons
• Input layer: input data/ predictor variables/ features
• Hidden layer: processing of input
• Output layer:Target (prediction of class or value)
Lee, J.-G., Jun, S., Cho, Y.-W., Lee, H., Kim, G. B., Seo, J. B., & Kim, N. (2017). Deep Learning in Medical Imaging: General
Overview. Korean Journal of Radiology, 18(4), 570–584. http://doi.org/10.3348/kjr.2017.18.4.570
14. Deep Learning
• Part of Machine Learning
• Uses multiple layers of ANN
• Mimics the working of human brain
Lee, J.-G., Jun, S., Cho, Y.-W., Lee, H., Kim, G. B., Seo, J. B., & Kim, N. (2017). Deep Learning in Medical Imaging: General
Overview. Korean Journal of Radiology, 18(4), 570–584. http://doi.org/10.3348/kjr.2017.18.4.570
15. Deep Learning
Three types of Network:
1. Deep Neural Networks: Google’s Deep mind Alpha Go
2. Recurrent Neural Networks (RNN): natural language
processing, handwriting and speech recognition
3. Convolutional Neural Networks (CNN): Computer
vision, image recognition, CV imaging
16. Cognitive Computing
Systems mimicking human cognition, help
replicate human capabilities across the spectrum of sensory
perception, deduction, reasoning, learning and knowledge.
17. Physician’s Brain and Machine-Equivalent capabilities
Chang A. Analytics and Algorithms, Big Data, Cognitive Computing, and Deep Learning in Medicine and Health Care. AI Med Ebook; 2017
18. 1. Introductory concepts
2. History & Current State of AI
3. Potential uses in Augmenting Heath care delivery
4. Challenges in Adoption
5. Future Directions
Agenda
19. The synergistic Convergence
A primer in artificialintelligence in cardiovascularmedicine
J. W. Benjamins · T. Hendriks · J. Knuuti · L. E. Juarez-Orozco · P. van der Harst
Neth Heart J (2019) 27:392–402
https://doi.org/10.1007/s12471-019-1286-6
20. What AITechnologies are available for Health Care ?
• ComputerVision
• Natural Language Processing (NLP)
• Conversational AI
• Virtual Assistants
• HybridTechnologies:AR/VR+ AI, Robotics+AI
• Robotic Process Automation (RPA)
21. Current State of AI: Health Care Delivery
1. Automation, CAD, Diagnostics
2. Risk Stratification
3. Prediction, Early Warning System
4. Precision Medicine
5. Home monitoring/Digital Therapeutics
6. Virtual Assistants/Conversational agents
22. 1. Introductory concepts
2. History & Current State of AI
3. Potential uses in Augmenting Heath care delivery
4. Challenges in Adoption
5. Future Directions
Agenda
24. Radiology: Computer Aided Detection
Critical Care
Suite: UCSF and
GE collaboration.
AI embedded in
Mobile X ray unit
25. AI aided TB screening
http://blog.qure.ai/notes/scaling-up-tb-
screening-with-ai
26. TB: Prompt diagnosis and treatment
http://blog.qure.ai/notes/scaling-up-
tb-screening-with-ai
27. Diabetic retinopathy screening
IDx- DR
First FDA approved AI algorithm
for the detection of DR in the offices of
non-ophthalmic healthcare practitioners
Padhy SK, Takkar B, Chawla R, Kumar A. Artificial intelligence in diabetic
retinopathy: A natural step to the future. Indian J Ophthalmol [serial
online] 2019 [cited 2020 Jan 24];67:1004-9. Available
from: http://www.ijo.in/text.asp?2019/67/7/1004/261034
https://www.eyediagnosis.co/
32. Arrhythmia Detection using AI
Cardiologist-level arrhythmia detection and classification in
ambulatory electrocardiograms using a deep neural network
Hannun,A.Y., Rajpurkar, P., Haghpanahi, M. et al. Cardiologist-level
arrhythmia detection and classification in ambulatory
electrocardiograms using a deep neural network. Nat Med 25, 65–69
(2019) doi:10.1038/s41591-018-0268-3
https://www.ncbi.nlm.nih.gov/pubmed/30617320
37. Phenomapping
Use of unsupervised machine learning in Phenotypic
classification of a heterogeneous clinical syndrome into discrete
phenogroups
PG1
PG2
PG3
PG4
PG5
Phenomapping
38. Deep Phenotyping
Detect Unknown High
risk groups
Develop refined High risk
stratification models
Target timely
prevention/Intervention
Deep Phenotyping
39. Dawn of New Era of
Augmented Intelligence
Physician and AI ( Human/Machine) synergy for facilitating better
• Diagnosis
• Clinical Decisions
• Disease management
Human
Machine
Intelligence
Augmented
Platform for
Precision Medicine
40. AI augmented remote Patient monitoring
Traditional
Hospital
ER
OPD
Home/
Natural
environment
Life Style,
Behavior
SDOH
41. Home monitoring
Smart remote devices
https://hbr.org/2019/05/the-health-care-benefits-of-combining-wearables-and-ai
47. Conversational AI
Conversational AI: Technology which allows Human Machine interaction
through the use of Natural conversation, utilizing voice user interface and
Machine Intelligence.
Human Machine
Conversational AI
Voice
Technology
NLP
ML,
Deep
Learning
Synergistic
Convergence
Conversation
48. Intelligent Virtual Agents
Task Execution using Machine intelligence
1
Perception/Sense
2
Understanding
Comprehension
3
Action/Execution
Speech/Voice /Image
Recognition
Natural Language
processing
Computer Vision
Natural
Language
Generation
4
Learn from past
actions/experiences
Machine
Learning
49. Current State
Conversational AI Ecosystem
Smart
Speakers
Smart
Displays
Smart voice
enabled
devices
Chatbots
Virtual
Assistants
Ambient
Clinical
Intelligence
Conversational
AI
Voice
VUI
AI
Ambient
Sensors
50. Conversational
AI
Outpatient
clinic
Home
On the Go
In Hospital
Conversational AI touchpoints in Health Care
Delivery
1. EHR documentation, navigation
2. Clinical Decision Support
3. Foreign Language Interpretation
1. Appointment Navigation
2. Patient Education
3. Patient Engagement
4. Medication management
5. Chronic care management:
bridging the care gap
1. Mobile health units/EMS
2. Patient Education
3. Care Continuum
4. Access to care providers
5. Access to Information
1. Patient Education, inpatient care
navigation
2. Discharge preparation
3. Clinical Decision Support
4. Operating suites: Hands free clinical
documentation and information retrieval
Decreasing Physician
Burnout: POC Tasks
Home Health
Optimization
Easing the Hospital Journey
Bridging the Care
Gap
51. Livongo for Diabetes program
HIPAA complaint Alexa Health care skill
• Right information
• Right time
• Guides behavior and rational decision making
https://www.youtube.com/watch?v=ShCPsCcy_TM
Task: Information exchange,
behavior modification,
Rational decision making
Technology: Alexa
Environment: Patient’s home
52. 1. Introductory concepts
2. History & Current State of AI
3. Potential uses in Augmenting Heath care delivery
4. Challenges in Adoption
5. Future Directions
Agenda
53. What are the Challenges in AI Adoption?
• Privacy
• Security
• Accuracy
• Reliability
• Explainability
• Trust
• Ethics
• Design and development
54. 1. Introductory concepts
2. History & Current State of AI
3. Potential uses in Augmenting Heath care delivery
4. Challenges in Adoption
5. Future Directions
Agenda
55. Future directions
• AI Education
• Multi stakeholder collaboration
• Sharing of best practices
• Ethical use and Societal Good
56. ClosingThoughts
Man + Machine synergy = Strengthens the power of healing
Team effort: Collaboration of all stakeholders with shared vision and aligned
interests is key to excel.
