Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
AI for Health Care
Yu-Chuan (Jack) Li, M.D., Ph.D., FACMI Professor in Biomedical Informatics,
Dean, College of Medical Science and Technology Taipei Medical University
A bit about myself
• Professor in Biomedical Informatics
• Board-certified Dermatologist
• Elected Fellow, ACMI (American College of Medical Informatics) and IAHIS (International Academy of Health Information Science)
• Fellow, ACHI (Australian College of Health Informatics)
• Editor-in-Chief, Computer Methods and Programs in Biomedicine (IF 2.7)
• Editor-in-Chief, International Journal for Quality in Healthcare (IF 2.6)
http://Jackli.cc
Computer Methods and Programs in Biomedicine
International Journal for Quality in Health Care
Editor-in-Chief
ISQua / OUP Elsevier 2000 submissions, 360 paper published / year
3
Defining AI
Artificial intelligence (AI) is intelligence exhibited by machines.
…
Colloquially, the term "artificial intelligence" is applied
when a machine mimics "cognitive" functions that
humans associate with other human minds, such as
"learning" and "problem solving".
… "We don't need Artificial Intelligence if we don't have Natural Stupidity!" - Professor Allan T. Pryor
6
Evolution of AI • 1960 Age of Reasoning
• Logic-based
• heuristic search
• 1990 Age of Representation
• Rule-based
• Knowledge engineering
• Expert system
• 2015~ Age of Machine Learning
• Big Data-driven
• Autonomous learning
• 2045 Age of Superintelligence?
7
Why AIHC in Taiwan • Taiwan has a strong ICT industry/academia
• Taiwan has one of the most“high performance”healthcare system in the world
• Very high outpatient visit – 15 visits /pers/yr
• Diagnoses/Drugs coded by physicians, NOT coders
• Accurate e-prescription – $$$ by NHI x 200
• 100% e-claim since 1995 95% EHR
• Highly accessible exams/tests 2M CT|MR /yr
• Very standard coding and data schema AI in Medicine market value
9.2 Billion USD in 2019
Key Issues in Current Health Care
•Medical Errors 醫療錯誤
•Poor/Inconsistent Quality 品質不佳
•One-size-fits-all Approach 以偏概全
•Prediction → Prevention 輕忽預防
Top 5 Causes of Death
Ref: Medical error—the third leading cause of death in the US, British Medical Journal(BMJ), 2016;353:i2139 doi: 10.1136/bmj.i2139
611 K
585 K
251 K
149 K
41 K
1st Heart Disease
2nd Cancer
3rd Medical Error
4th COPD
5th Suicide
Causes of Death Per Year, USA
98 K 2000
2013
Poor Quality
• 45% did NOT receive recommended care
(US adult)
• Pneumonia 61% X
• Asthma 47% X
• Hypertension 35% X
• 41% did NOT receive standard care (AU
Children)
McGlynn et al., New England Journal of Medicine, 2003
Braithwaite et al., JAMA, 2018
12
One-size-fits-all Medicine
• A clinical practice hinges on “average”
• Lab data range: No difference between
10 y/o girl and 80 y/o man
• Diagnostic accuracy: 20% wrong
• Allergy Info: 40% missing
• Family Hx: 90% incomplete
• Genomic, Behavioral and Environmental
data NOT AVAILABLE to doctors
13
Preventive Medicine is Hard
• No visible target
• Repetitive & Slow
• No pain
• People don’t understand probability
• Science can’t produce reliable/useful
predictions
Low market value
To Prevent Medication Errors
733.4 Millions Prescriptions
80 Million Dx-Med and 2.25 Million Med-Med Associations Explored
Medication codes Mapped to 1,500 unique WHO
codes
Diagnoses ICD codes
20,000 unique ICD codes
Machine Learning Learn from Doctors‘ Behavior
1.34B 2.53B
Results after 2 months of running
A Medical Center in Taiwan
• Patients:72,378
Reminders:2140 (3%)
Agreed:1038
• High risk medications
• Patients :17,793人
Reminders :114
Agreed :62
A Healthcare System in the US
• Patients : 31,728人次
Reminders : 2,723 (8.6%)
*Estimated
Patient
Profile
Diagnosis
/Problem
Procedures Medication
Lab/Exam
7
11
Age, sex, allergy, weight,
height, blood type, body
temperature, …etc.
YC (Jack) Li et. al., 2004
Current and/or chronic
dz, DM, H/T,
Pregnancy…etc.
Surgery, transfusion,
endoscopy,
angiogram, PTCA,
rehabilitation…etc.
Propanolol vs
theophylline,
Cipro vs aminophylline,
Acetaminophen vs
Phenytoin…etc.
CBC, D/C, Chem-
20, hCG, PT,
APTT, INR…etc.
e.g. Coumadin vs
INR
e.g. Wafarin vs
angiogram
e.g. Penicillin vs
PCN allergy
e.g. Retinoids vs
pregnancy
Data Interaction Model for Adverse Event detection
2x
2x
2x
2x
1x
Input Variables for AIHC
Patient Profile
Diagnosis /Problem
Procedures Medication
Lab/Exam
12
9
7
8
5
11
10
4 3
2
Birth
YC (Jack) Li et. al., 2016
Phenotype
(Environmental)
AIHC的八大類資料來源與時間軸
Output Variables of AIHC
Death
YC (Jack) Li et. al., 2016
Clinical Events
Treatment Rehabilitation
Prognosis
Management
Diagnosis
Prediction Early
Detection Suggestion/Recommendation
Augmented
Intelligence
AI and Medicine – for NOW
Adapted from: Charles P. Friedman. J Am Med Inform Assoc. 2009;16:169 –170.
Rule the World!
Men lose jobs
Conclusion
• AI and Healthcare can go hand-in-hand
• AI can help on QPS, one-size-fits-all and prediction/prevention
• AI has to change the future of medicine (or we may not have one)
• because we deserve it!