中国实用外科杂志

• 专家论坛 • 上一篇    下一篇

人工智能在疝与腹壁外科中应用的思考

陈    双,黄恩民,周太成,马    涛   

  1. 中山大学附属第六医院胃肠、疝和腹壁外科   广东省胃肠病研究所   广东省结直肠盆底疾病研究重点实验室,广东广州 510655
  • 出版日期:2023-06-01

  • Online:2023-06-01

摘要: 以不同子领域应用为层次,人工智能可分为机器学习、自然语言处理、人工神经网络和计算机视觉4个部分。机器学习可提高疝与腹壁外科疾病的诊断和预后预测效果。在自然语言处理方面,使用电子病历系统的自然语言所构建的模型在术后早期具有优秀的吻合口漏预测能力。在人工神经网络方面,中山大学附属第六医院胃肠、疝和腹壁外科在胃食管反流病中应用人工神经网络的研究结果表明,术前检查的9个参数可以良好预测术后结局,横向对比其他算法后发现人工神经网络在预测预后方面可能更具优势。

关键词: 人工智能, 疝与腹壁外科, 机器学习, 自然语言处理, 人工神经网络, 计算机视觉

Abstract: Artificial Intelligence in hernia and abdominal wall surgery        CHEN Shuang, HUANG En-min, ZHOU Tai-cheng, et al.  Department of Gastroenterological Surgery and Hernia Center, The Sixth Affiliated Hospital, Guangdong Institute of Gastroenterology, Guangdong Provincial Key Laboratory of Colorectal and Pelvic Floor Diseases, Sun Yat-sen University, 510655, Guangzhou, Guangdong, China
Corresponding author: CHEN Shuang, E-mail:chensh223@
mail.sysu.edu.cn
Abstract    AI can be divided into four parts: machine learning, natural language processing, artificial neural networks and computer vision, based on different sub-domain applications. Machine learning can improve the diagnosis and prognosis prediction of hernia and abdominal wall surgery. In terms of natural language processing, the model constructed by natural language using an electronic medical record system has excellent anastomotic leakage prediction ability in the early postoperative period. In terms of artificial neural networks, the research results of Gastrointestinal, Hernia and Abdominal Wall Surgery a the Sixth Affiliated Hospital of Sun Yat-sen University in gastroesophageal reflux disease show that nine parameters of preoperative examination can well predict a postoperative outcome, and horizontal comparison with other algorithms shows that artificial neural networks may have more advantages in predicting prognosis.

Key words: artificial intelligence, hernia and abdominal wall surgery, machine learning, natural language processing, artificial neural networks, computer vision