Writing about web page https://warwick.ac.uk/fac/med/research/hscience/sssh/research/lyncs/
In this blog I consider some of the social and ethical implications of the use of AI for algorithm driven clinical decision-making.
Developments in AI for health seem to challenge the role of and even need for health practitioners.
In the medium term, at least in the UK, society seems to require AI and clinicians to work together. There is a growing consensus that for a decision such as a health decision, where the implications for the individual are considerable, the individual should be informed that the decision is being made using AI and have the right to request that the decision is not made by AI1 2. If a decision is made using AI they should have the right to ask for the decision to be reconsidered1 2. It is not yet clear where responsibility lies for the consequences of advice given directly to the patient based on AI3, and there is a deman for the AI algorithms to be explainable4.
If part of the work of a clinician undertaken by AI, there is a question of how they maintain their expertise5 inorder to take responsibility for decisions and to provide a second opinion when requested. Further, the clinicians need to understand AI in order to be able to explain the decision-making to their patients.
Low and middle income countries (LMIC) do not have the health parctitioner workforce capacity that high income countries enjoy but most of them have good digital infrastructure – far better than water or roads. There is potential for AI to be rapidly taken up in LMIC to keep healthcare cost down whist providing improved services. In this scenario the lack of human capacity in the system might mean the decision-making process is not explained to patients and there is no second opinion available.
1. House of Lords Select Committee on Artificial Intelligence. AI in the UK: ready, willing and able?, 2018.
2. The Alan Turing Institute. The GDPR and Beyond: Elizabeth Denham, UK Information Commissioner, 2018.
3. Floridi L, Taddeo M. What is data ethics?: The Royal Society, 2016.
4. Information Commissioner’s Office. Information Rights Strategic Plan 2017-2021
5. Yang G-Z, Bellingham J, Dupont PE, et al. The grand challenges of Science Robotics. Science Robotics 2018;3(14):eaar7650.