AI and Medicine
Preliminary qs
How do the four fields of ethics, law, ai and medicine interact
What are the key ethical principles or legal frameworks at play
What are the potential benefits and risks of ai in healthcare
Who should be held accountable when medical ai systems fail
How can bias in ai algorithms be identified and mitigated
What are the future challenges and opprtunities for the ai health sector
History of AI
1950
1955s Dartmouth conference
1960s AI explosion
1970s AI winter
1980s-2000s AI renaissance
NOW
Stages of AI
Stage 1
Artificial narrow intelligence ANI
Machine learning
Specialises in one area and solves one probelm
Stage 2
Artificial general inteligence AGI
Machine intelligence
Referes to a computer that is at smart as the human across the board
Stage 3
Artificial super intelligence ASI
Machine conciousness
An intellect that is much smarter than the best human brains in practically every field
AI Family
Machine Learning (ML)
Natural Langauge procesing (NLP)
Expert systems
Vision
Speech
Planning
Robotics
EU Artificial Intelligent Act
Regulation and directives
Purpose
Ensure that AI systems are Safe, respect fundemental human rights and union values
Ensure llegal certainty to facilitate investment ad innovation in AI
Facilitate the development of single market for lawful, safe and trustworthy AI applications
Risk Based Approach to AI Regulation
Lower risk - less regulation
Medical AI regulation / Law
Requirements for surveillance
EU GDPR
EU Medical device frameworks
Product liability and safety legislation
Proposed AI liability Directive
Clinical Applications of AI
Diagnostics and imaging
Predictive analytics
Personalised medicine
Clinical decision support
Robotics and surgery
Drug discorvery and development
VHA telemedicine and remote care
Admin efficiency
Population health managment
Mental health
Core Prinicples of Medical Ethics
Autonomy
Beneficence
Non-maleficence
Justice
Bias Mitigation
Diverse and Representative data, bias audits
Integration of fairness contraints
Transparency and explainability XAI
Human oversight continous monitoring , feedback loops
Ethical AI
Regulatory complicance
Post market surveillance
Education and training
Liability
EU act
Providers
Users
Ensure system works according to intention
Bear primary compliance responsibility
AI Liability Directive
Fault-based liability
Strict liability for operators of high risk AI