There’s a growing conversation surrounding the concept of AI regulation within healthcare – but what exactly does this term refer to, and how can it be applied in a clinical setting? In this blog post, we’ll unpack this definition whilst exploring how it can be used ethically and responsibly to shake up the research and development process, benefitting a huge number of individuals across a wide spectrum of indications.
May 30, 2022
What exactly is a distributed approach to clinical AI regulation?
The explosion of interest in AI applications offers a lot of potential in terms of streamlining and improving myriad aspects of the clinical setting – but these changes come with a unique set of challenges that need to be addressed before the healthcare sector can fully embrace them. Issues of ethics within data collection for public health purposes have long been at the forefront of discussions on this topic, with many individuals and organisations alike raising concerns surrounding the use of something as sensitive as medical data. Additionally, the sheer volume of businesses employing AI-based solutions in their daily practice might indicate that a more concrete and comprehensive approach to regulation is on the cards, allowing AI applications to reach their full capability whilst safeguarding data and changing the lives of countless real-world patients.
Centralised regulation – in which a singular entity monitors the expansion of AI practices and solutions - does have its benefits. However, many people in both the data analytics and healthcare field are questioning whether or not this is the most sustainable approach in the long term – and rightly so. The wider implementation of AI poses a variety of hurdles, and it’s no secret that any organisation looking to adopt such policies will need to accept that it all comes with a big learning curve. It’ll take time for the best regulation framework to be developed and put in place, and a distributed approach to this looks like a golden opportunity to establish more coherent best practices than many businesses and individuals are currently working with.
By adopting a decentralised and distributed approach – in which the decision-making power and management surrounding the efficiency, efficacy, effectiveness of AI models and applications - a hybrid model of regulation can become the new normal, involving a wider range of ideas and potential solutions. Concentrating only a small number of regulatory issues within a centralised governing body would free up huge amounts of time, money and effort, giving decentralised administrations the opportunity to decide what works best for them and their organisation without the hassle of an outdated and unnecessary system of approval. For example, this model might be used in high-risk situations in which clinical AI applications are 100% automated and do not entail the supervision of a trained clinician, ensuring that AI-based solutions are monitored when necessary.
Clinical AI opens up a treasure trove of potential in the healthcare sector, but it’s important that it keeps up with the current cultural moment, affording decentralised bodies the capacity to make their own decisions through a streamlined and distributed approach. Laying the appropriate groundwork to make this vision a reality is a solid next step – involving building an accountability framework to continually monitor ongoing progress, developing open data registries, and establishing a comprehensive review process to evaluate areas for improvement.
Medical research is changing, and by taking a step back and reevaluating our attitudes towards these innovative ideas and concepts, we can create a unique opportunity to prepare this field for the future, creating a safer, more responsible, and more equitable healthcare sector for all.