文献速递口腔修复与种植口腔颌面外科

基于卷积神经网络的CBCT上颌牙槽骨自动分割

Convolutional neural network-based automated maxillary alveolar bone segmentation on cone-beam computed tomography images.
PubMed 文献 · Clin Oral Implants Res · 2023-03-23

本研究开发并评估了一种基于卷积神经网络(CNN)的人工智能工具,用于在锥形束CT(CBCT)图像上自动进行三维上颌牙槽骨分割。共收集141例CBCT扫描用于训练(n=99)、验证(n=12)和测试(n=30)。自动分割的准确性指标表现优异,但手动分割(95% HD: 0.20 ± 0.05 mm; IoU: 95% ± 3.0; DSC: 97% ± 2.0)略优于AI分割(95% HD: 0.27 ± 0.03 mm; IoU: 92% ± 1.0; DSC: 96% ± 1.0)。AI分割(51.5 ± 10.9 s)比手动分割(5973.3 ± 623.6 s)快116倍,差异有统计学意义(p < .001)。结论认为,尽管手动分割表现略好,但新型CNN工具也能提供高精度的上颌牙槽骨及其嵴顶轮廓分割,且耗时仅为手动方法的1/116。

要点

  • CNN模型可自动分割CBCT上颌牙槽骨及嵴顶轮廓,准确性指标优异。
  • 手动分割准确性略优于AI:95% HD 0.20±0.05 mm vs 0.27±0.03 mm。
  • AI分割耗时51.5±10.9秒,比手动分割快116倍(p < .001)。

核心结论

  • 尽管手动分割表现略好,但新型CNN工具也能高精度分割上颌牙槽骨及其嵴顶轮廓,且耗时仅为手动方法的1/116。

研究设计与样本

  • 研究收集141例CBCT扫描,用于CNN模型的训练(n=99)、验证(n=12)和测试(n=30),实现上颌牙槽骨及嵴顶轮廓的自动分割。

主要发现

  • 自动分割的准确性指标均表现优异。
  • 手动分割(95% HD: 0.20 ± 0.05 mm; IoU: 95% ± 3.0; DSC: 97% ± 2.0)略优于AI分割(95% HD: 0.27 ± 0.03 mm; IoU: 92% ± 1.0; DSC: 96% ± 1.0)。
  • 分割方法间耗时差异有统计学意义(p < .001);AI分割(51.5 ± 10.9 s)比手动分割(5973.3 ± 623.6 s)快116倍;R-AI方法耗时居中(1666.7 ± 588.5 s)。

关键数值与推荐

  • 手动分割:95% HD 0.20 ± 0.05 mm,IoU 95% ± 3.0,DSC 97% ± 2.0。
  • AI分割:95% HD 0.27 ± 0.03 mm,IoU 92% ± 1.0,DSC 96% ± 1.0。
  • AI分割耗时51.5 ± 10.9 s,手动分割耗时5973.3 ± 623.6 s,R-AI耗时1666.7 ± 588.5 s;耗时差异p < .001。

临床怎么用

  • CNN工具可用于CBCT上颌牙槽骨及嵴顶轮廓的自动分割,耗时显著少于手动分割。
  • 对于自动分割中分割不足或过度估计的三维模型,可由专家细化生成R-AI分割。

局限性

  • 原文未述局限性

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

OBJECTIVES: To develop and assess the performance of a novel artificial intelligence (AI)-driven convolutional neural network (CNN)-based tool for automated three-dimensional (3D) maxillary alveolar bone segmentation on cone-beam computed tomography (CBCT) images. MATERIALS AND METHODS: A total of 141 CBCT scans were collected for performing training (n = 99), validation (n = 12), and testing (n = 30) of the CNN model for automated segmentation of the maxillary alveolar bone and its crestal contour. Following automated segmentation, the 3D models with under- or overestimated segmentations were refined by an expert for generating a refined-AI (R-AI) segmentation. The overall performance of CNN model was assessed. Also, 30% of the testing sample was randomly selected and manually segmented to compare the accuracy of AI and manual segmentation. Additionally, the time required to generate a 3D model was recorded in seconds (s). RESULTS: The accuracy metrics of automated segmentation showed an excellent range of values for all accuracy metrics. However, the manual method (95% HD: 0.20 ± 0.05 mm; IoU: 95% ± 3.0; DSC: 97% ± 2.0) showed slightly better performance than the AI segmentation (95% HD: 0.27 ± 0.03 mm; IoU: 92% ± 1.0; DSC: 96% ± 1.0). There was a statistically significant difference of the time-consumed among the segmentation methods (p < .001). The AI-driven segmentation (51.5 ± 10.9 s) was 116 times faster than the manual segmentation (5973.3 ± 623.6 s). The R-AI method showed intermediate time-consumed (1666.7 ± 588.5 s). CONCLUSION: Although the manual segmentation showed slightly better performance, the novel CNN-based tool also provided a highly accurate segmentation of the maxillary alveolar bone and its crestal contour consuming 116 times less than the manual approach.

来源:Clin Oral Implants Res,查看原文。版权归原作者所有。

引用

  1. Fontenele RC, Gerhardt MDN, Picoli FF, et al. Convolutional neural network-based automated maxillary alveolar bone segmentation on cone-beam computed tomography images. Clin Oral Implants Res. 2023. PMID: 36906917. doi:10.1111/clr.14063

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。