人工智能赋能的全髋关节置换术进展  

Advances in AI-Enabled Total Hip Arthroplasty

作　　者：黄寅殊 单浩杰 于晓巍[2] HUANG Yinshu;SHAN Haojie;YU Xiaowei(College of Engineering Science and Technology,Shanghai Ocean University,Shanghai 201306,China;Department of Orthopedics,Shanghai Sixth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine,Shanghai 200233,China)

机构地区：[1]上海海洋大学工程学院,上海201306 [2]上海交通大学医学院附属第六人民医院骨科,上海200233

出　　处：《医用生物力学》2024年第6期1228-1234,共7页Journal of Medical Biomechanics

基　　金：上海市2022年度“科技创新行动计划”技术标准项目(22DZ2203600);上海市创新医疗器械应用示范项目(23SHS02800)。

摘　　要：全髋关节置换术(total hip arthroplasty,THA)的术前规划、术中导航和术后康复等环节,都因人工智能(artificial intelligence,AI)技术的介入而得到显著优化。本文综述AI技术在医疗影像分割和配准方面的最新进展,特别关注其在THA中的应用。医疗影像与自然影像的显著差别对AI算法的设计构成挑战。深度学习技术,特别是CNN、U-Net和Transformer模型,在各项医疗影像分割和配准任务上表现突出。AI技术通过深度学习分析CT影像,显著提高了髋部病变的识别准确性。在术中引导方面,AI系统利用智能分割和运动状态模拟,为手术提供实时导航和精确定位,有效提升手术效率。AI技术还涉及手术成本预测和术后康复,通过马尔可夫模型等方法,为医疗决策提供了有力的数据支持。随着深度学习技术的不断进步,医疗影像分析正逐步实现自动化和智能化,对改善患者的整体手术体验和临床结果具有重大的临床意义,预示着未来在医疗影像领域可能实现的新突破。Preoperative planning,intraoperative navigation,and postoperative rehabilitation of total hip arthroplasty(THA)have been significantly enhanced by the integration of artificial intelligence(AI)technology.This review summarizes the latest advancements in AI technology for medical image segmentation and registration,with a particular focus on its application in THA.The notable differences between medical and natural images present challenges for the design of AI algorithms.Deep learning techniques,especially CNN,U-Net,and Transformer models,have demonstrated an outstanding performance in various medical image segmentation and registration tasks.The AI technology,through deep learning analysis of CT images,has significantly improved the accuracy of identifying hip pathologies.In terms of intraoperative guidance,AI systems provide real-time navigation and precise positioning for surgeries by utilizing intelligent segmentation and motion state simulation,effectively enhancing surgical efficiency.AI technology also encompasses surgical cost prediction and postoperative recovery,offering robust data support for medical decision-making through method such as Markov models.As deep learning technology continues to advance,the analysis of medical images is progressively achieving automation and intelligence,which has significant clinical implications for improving patients’overall surgical experiences and outcomes,suggesting potential new breakthroughs in the field of medical imaging in the future.

关 键 词：人工智能 髋关节置换术 术前规划 医疗影像分割 医疗影像配准 

分 类 号：R318.01[医药卫生—生物医学工程]