Partner and Head of Patents & Life Sciences Juli Mansnerus delivered a lecture on artificial intelligence in healthcare at HY+ University of Helsinki Centre for Continuing Education, exploring the intersection of artificial intelligence and healthcare regulation with an engaged audience.
Here are some key insights from the session.
Risk-based regulatory approach is essential
AI requirements should be approached on a risk basis and evaluated case-by-case, with AI systems classified into four main risk levels based on their intended use. This framework ensures that regulatory oversight matches the potential impact of each AI application.
Healthcare AI use cases are increasingly diverse
The range of AI use cases in healthcare continues to expand rapidly, from automated patient record drafting and AI-assisted care planning to simultaneous speech interpretation and AI-generated patient guidance. Each application requires careful legal analysis to ensure compliance with applicable regulations.
Multiple regulatory frameworks apply
Healthcare AI implementations must comply with a complex web of regulations, including data protection laws, the AI Act, healthcare-specific rules, medical device regulations, and good governance principles. Understanding how these frameworks interact is crucial for successful implementation.
Five-step implementation framework for success
Effective AI adoption in healthcare requires a structured approach: organisational guidelines, strategic assessment, risk evaluation, contingency planning, and continuous monitoring. This systematic five-step implementation framework helps organisations navigate regulatory requirements while maximising the benefits of AI technology.
Human oversight remains fundamental
Human oversight remains crucial and cannot be eliminated. The same healthcare professional must participate in both the original consultation and record verification, making all relevant decisions based on the AI-generated records to ensure accountability and quality of care.
AI – enabler of good governance
Rather than viewing legal requirements as barriers to AI implementation, organisations should recognise that these frameworks form the foundation for ethical AI use. When implemented correctly, AI enables good and effective governance.
Conclusion
The regulatory landscape is complex, but it remains navigable with the right legal guidance and strategic approach.
If you are interested in discussing more about AI implementation in healthcare or need guidance on navigating the regulatory landscape, please contact Juli Mansnerus, Partner and Head of Patents & Life Sciences