Vascular Surgery Department, Royal Blackburn Teaching Hospital, Haslingden Road, Blackburn, BB23HH, UK.
Faculty of Biology, Medicine and Health, The University of Manchester, Manchester, UK.
Eur J Trauma Emerg Surg. 2021 Jun;47(3):757-762. doi: 10.1007/s00068-020-01444-8. Epub 2020 Jul 26.
Artificial intelligence (AI) is a field involving computational simulation of human intelligence processes; these applications of deep learning could have implications in the specialty of emergency surgery (ES). ES is a rapidly advancing area, and this review will outline the most recent advances.
A literature search encompassing the uses of AI in surgery was conducted across large databases (Pubmed, OVID, SCOPUS). Two doctors (LR, CH) both collated relevant papers and appraised them. Papers included were published within the last 5 years, and a "snowball effect" used to collate further relevant literature.
AI has been shown to provide value in predicting surgical outcomes and giving personalised patient risks based on inputted data. Further to this, image recognition technology within AI has showed success in fracture identification and breast cancer diagnosis. Regarding theatre presence, supervised robots have carried out suturing and anastomosis of bowel in controlled environments to a high standard.
AI has potential for integration across surgical services, from diagnosis to treatment, and aiding the surgeon in key decision-making for risks per patient. Fully automated surgery may be the future, but at present, AI needs human supervision.
人工智能（AI）是一个涉及人类智能过程的计算模拟领域；这些深度学习的应用可能会对急诊外科（ES）这一专业产生影响。ES 是一个快速发展的领域，本综述将概述最近的进展。
在大型数据库（Pubmed、OVID、SCOPUS）中进行了涵盖 AI 在外科手术中应用的文献检索。两位医生（LR、CH）都整理了相关论文并进行了评估。所纳入的论文均在过去 5 年内发表，并采用“滚雪球效应”来进一步收集相关文献。
AI 已被证明在预测手术结果和根据输入数据为患者提供个性化风险方面具有价值。除此之外，AI 中的图像识别技术在骨折识别和乳腺癌诊断方面取得了成功。在手术室方面，经过监督的机器人已经在受控环境中以高标准完成了缝合和肠吻合术。
AI 有可能在从诊断到治疗的各个外科服务中得到整合，并帮助外科医生针对每位患者的风险做出关键决策。完全自动化的手术可能是未来的趋势，但目前，AI 需要人类的监督。
