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

image

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