- with Finance and Tax Executives
- with readers working within the Utilities industries
The first reported use of live AI assistance during brain surgery highlights both the opportunities and challenges AI presents for NHS and private medical organisations.
The reported success of the world's first patient to undergo live AI-assisted brain surgery is an important milestone in the development of artificial intelligence (AI) in healthcare.
The procedure involved AI analysing live surgical footage and assisting the neurosurgical team in identifying important anatomical structures during the removal of a brain tumour. The surgeon remained responsible for the procedure, with the technology providing additional information in real time.
The outcome provides an encouraging illustration of how AI may augment clinical expertise and potentially improve patient safety. However, due caution is required to ensure that innovation is accompanied by appropriate clinical governance and risk management.
An opportunity for patient safety
AI has the potential to make a significant contribution to patient care.
It is already being explored and used in areas including diagnostics, imaging, clinical documentation, monitoring and decision support. As the technology develops, its role is likely to extend further into treatment planning and, as the recent neurosurgical procedure demonstrates, real-time assistance during procedures.
Used appropriately, this could support better decision-making and reduce some of the risks associated with human error. The potential patient-safety benefits are substantial. However, these benefits should not lead to an assumption that AI is inherently safer than human decision-making. AI is an additional clinical tool; it does not remove the need for appropriate professional judgment and oversight.
The importance of human oversight
One of the key considerations for NHS organisations will be establishing an appropriate relationship between the clinician and the technology.
AI systems have limitations. Their performance can vary depending upon the data upon which they have been developed, the clinical environment in which they are deployed and the circumstances of the individual patient.
There is also a risk of automation bias, where too much reliance is placed on an automated recommendation because it appears authoritative or sophisticated.
Training should therefore cover how an AI system operates and its limitations, and outputs must always remain grounded in clinical judgment.
The role of insurers and indemnifiers
AI is also relevant to NHS insurers and indemnifiers.
As adoption increases, insurers will need to understand how AI is being used within the organisations they support, particularly where it has a direct role in clinical decision-making.
AI may ultimately reduce risk and improve patient safety, and insurers are well placed to support organisations in understanding both the benefits and emerging risks.
Claims and incident data may, over time, provide valuable insight into whether particular applications of AI are achieving their intended safety benefits.
There is also value in ensuring that contractual and insurance arrangements remain appropriate as technology develops.
Clear allocation of responsibilities between healthcare providers, technology suppliers and other organisations involved in implementation can help reduce uncertainty and support effective risk management.
Looking ahead
The first reported live AI-assisted brain surgery is a significant step and clearly demonstrates the potential for technology to augment, rather than replace, clinical expertise. Similar developments are likely to become increasingly common in clinical practice.
For NHS Trusts, the focus will be on ensuring an environment in which innovation and patient safety are mutually supportive.
For insurers and indemnifiers, understanding how AI is being deployed will be increasingly important in the identification of emerging risks, in recognising where technology may reduce existing sources of clinical harm and in respect of liability and indemnity issues.
AI is clearly a clinical tool with considerable potential, and its successful adoption will depend on the quality of the governance around it, it will therefore be important to ensure that governance keeps pace with innovation.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
[View Source]
