Artificial intelligence may improve the efficiency of NHS surgery

Dear Editor,

To strive for efficiency in healthcare delivery is a moral imperative. The strategic use of data to facilitate this goal, both in the NHS and abroad, should be encouraged.[1] The article by Limb highlights the necessity of sophisticated analysis in operating theatre scheduling.[1] Artificial intelligence (AI) may be able to assist with this task.

AI can make accurate predictions about the future, including with respect to aspects of operative scheduling. For example, previous studies have shown that AI may improve the accuracy of procedure duration prediction for robotic and ophthalmological operations.[2] Additionally, by leveraging diverse datatypes, AI may achieve high levels of performance. For example, medical free text may be used to make predictions relevant to discharge planning. Similarly, pre-operative physiological parameters, imaging and laboratory parameters may be integrated into predictive AI models. Acknowledging the need for generalisable,[3] or locally derived models, AI has great potential for this field.

It is important to note that an accurate prediction does not necessarily convey an improvement in efficiency or care. Appropriate systems need to be in place to act upon such predictions so as to realise tangible benefits. Accordingly, any use of AI for such applications should be interrogated with robust study designs prior to introduction to routine care.[4] Any costs associated with AI algorithm implementation, including upkeep and maintenance costs, should be weighed in the context of demonstrated efficiency savings.

We agree with Limb that the use of additional analytical approaches may convey significant efficiency savings for operating theatre bookings. The investigation of the utility of AI should be encouraged, although not at the cost of the aspects of the system that it is designed to facilitate. Operating theatre scheduling, both in the NHS and other surgical systems, may derive benefit from the intelligent use of artificial intelligence.

References:
1. Limb M. Four ways to fix the inefficiency of NHS surgery. BMJ 2023;380:o2455. doi: 10.1136/bmj.o2455 [published Online First: 2023/01/27]
2. Birkhoff DC, van Dalen ASH, Schijven MP. A review on the current applications of artificial intelligence in the operating room. Surgical Innovation 2021;28(5):611-19.
3. Iqbal U, Celi LA, Hsu Y-HE, et al. Healthcare artificial intelligence: the road to hell is paved with good intentions: BMJ Publishing Group Ltd, 2022:e100650.
4. Kovoor JG, Bacchi S, Gupta AK, et al. Artificial intelligence clinical trials and critical appraisal: a necessity. ANZ J Surg 2023 doi: 10.1111/ans.18263 [published Online First: 2023/01/12]