文献速递关节外科

THA-Net:用于下一代模板设计和患者特异性手术执行的深度学习解决方案

THA-Net: A Deep Learning Solution for Next-Generation Templating and Patient-specific Surgical Execution.
PubMed 文献 · J Arthroplasty · 2023-08-22

本研究提出THA-Net深度学习图像修复算法,可从单张术前骨盆X光片生成模拟全髋关节置换术(THA)的术后X光片,算法可无条件或由外科医生条件选择植入物。基于单中心14,357例患者356,305对X光片训练后,合成术后X光片的手术执行有效性评分显著高于真实片(10分制Likert平均差0.8至1.1分,P<.001),盲法专家审阅无法区分真实与合成片的真实感。

要点

  • THA-Net可从单张术前骨盆X光片生成逼真的模拟THA术后X光片。
  • 合成片手术执行有效性评分显著高于真实片(10分制Likert平均差0.8至1.1分,P<.001)。
  • 盲法专家审阅无法区分合成片与真实片的真实感。

核心结论

  • 作者团队开发了THA下一代模板设计工具,其生成的合成X光片在最终手术执行评分上高于训练数据中的真实X光片。
  • 作者认为进一步优化该工具有望促进患者特异性手术规划,并赋能机器人、导航和增强现实等技术。

研究设计与样本

  • THA-Net是一种深度学习图像修复算法,接收术前X光片输入后,将目标髋关节替换为THA植入物以生成合成但逼真的术后X光片。
  • 模型训练使用来自单一机构全关节登记处的14,357例患者的356,305对X光片。
  • 评估内容为输出的有效性(手术执行质量)和真实感(区分真实与合成X光片的能力),对照基于人工和基于软件的标准。

主要发现

  • 合成术后X光片的手术有效性显著高于真实片(10分制Likert平均差0.8至1.1分,P<.001)。
  • 在盲法专家审阅中,合成片与真实片在真实感方面无法被区分。
  • 使用已验证的深度学习模型分析时,合成图像显示出极好的有效性和真实感。

关键数值与推荐

  • 手术有效性:合成片较真实片10分制Likert评分平均差0.8至1.1分(P<.001)。

临床怎么用

  • 作者开发的THA-Net可从单张术前骨盆X光片生成模拟术后X光片,并可选择由算法或外科医生决定植入物。
  • 作者认为该工具需进一步优化后可能用于患者特异性手术规划及机器人、导航、增强现实等技术。

局限性

  • 作者指出该工具需进一步优化(Further refinement of this tool may potentiate patient-specific surgical planning and enable technologies such as robotics, navigation, and augmented reality)。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND: This study introduces THA-Net, a deep learning inpainting algorithm for simulating postoperative total hip arthroplasty (THA) radiographs from a single preoperative pelvis radiograph input, while being able to generate predictions either unconditionally (algorithm chooses implants) or conditionally (surgeon chooses implants). METHODS: The THA-Net is a deep learning algorithm which receives an input preoperative radiograph and subsequently replaces the target hip joint with THA implants to generate a synthetic yet realistic postoperative radiograph. We trained THA-Net on 356,305 pairs of radiographs from 14,357 patients from a single institution's total joint registry and evaluated the validity (quality of surgical execution) and realism (ability to differentiate real and synthetic radiographs) of its outputs against both human-based and software-based criteria. RESULTS: The surgical validity of synthetic postoperative radiographs was significantly higher than their real counterparts (mean difference: 0.8 to 1.1 points on 10-point Likert scale, P < .001), but they were not able to be differentiated in terms of realism in blinded expert review. Synthetic images showed excellent validity and realism when analyzed with already validated deep learning models. CONCLUSION: We developed a THA next-generation templating tool that can generate synthetic radiographs graded higher on ultimate surgical execution than real radiographs from training data. Further refinement of this tool may potentiate patient-specific surgical planning and enable technologies such as robotics, navigation, and augmented reality (an online demo of THA-Net is available at: https://demo.osail.ai/tha_net).

来源:J Arthroplasty,查看原文。版权归原作者所有。

引用

  1. Rouzrokh P, Khosravi B, Mickley JP, et al. THA-Net: A Deep Learning Solution for Next-Generation Templating and Patient-specific Surgical Execution. J Arthroplasty. 2023. PMID: 37619804. doi:10.1016/j.arth.2023.08.063

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。