生成对抗性图像修复网络增强牙周临床附着水平预测
本研究采用生成对抗网络(GAN)结合部分卷积的图像修复算法,从咬翼片和根尖片中预测视野外解剖结构,以提升临床附着水平(CAL)预测准确性。回顾性 purposive 抽样纳入牙周健康及患病病例的影像与临床 CAL 记录,训练集 80,326 张、验证集 12,901 张、比较集 10,687 张图像。结果显示,修复方法的 CAL 预测平均绝对误差(MAE)为 1.04mm,优于非修复方法的 1.50mm,Dunn 两两检验显示修复方法显著优于对应非修复方法,最佳方法 Dunn 值为 -63.89。
要点
- 生成对抗修复网络结合部分卷积可预测咬翼片和根尖片视野外解剖,提升 CAL 预测准确性。
- 修复方法 MAE 为 1.04mm,非修复方法为 1.50mm,修复方法优于非修复方法。
- Dunn 两两检验显示修复方法较非修复方法 CAL 预测准确性显著改善,最佳方法 Dunn 值为 -63.89。
核心结论
- 使用生成对抗修复网络结合部分卷积从咬翼片和根尖片预测 CAL 优于非修复方法,且修复方法在 1mm 临床医生测量标准范围内。
研究设计与样本
- 回顾性 purposive 抽样纳入牙周健康及牙周患病病例,使用咬翼片、根尖片及临床医生记录的 CAL,数据时间范围为 2016 年 7 月 1 日至 2020 年 1 月 30 日。
- 训练集 80,326 张图像,验证集 12,901 张图像,比较集 10,687 张图像用于比较非修复方法与修复方法的 CAL 预测。
- 统计分析采用平均偏差误差(MBE)、平均绝对误差(MAE)及 Dunn 两两检验比较 CAL,p=0.05。
主要发现
- 修复方法与非修复方法对应的 p 值显示 CAL 预测准确性有统计学显著改善,MAE 分别为 1.04mm 和 1.50mm。
- Dunn 两两检验表明修复方法较其对应的非修复方法 CAL 预测准确性有统计学显著改善,表现最佳的方法 Dunn 两两值为 -63.89。
关键数值与推荐
- 修复方法 MAE 为 1.04mm,非修复方法 MAE 为 1.50mm。
- Dunn 两两检验 p=0.05,最佳方法 Dunn 两两值为 -63.89。
临床怎么用
- 原文未据此给出临床推荐。
局限性
- 原文未报告局限性。
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: Bone level as measured by clinical attachment levels (CAL) are critical findings that determine the diagnosis of periodontal disease. Deep learning algorithms are being used to determine CAL which aid in the diagnosis of periodontal disease. However, the limited field-of-view of bitewing x-rays poses a challenge for convolutional neural networks (CNN) because out-of-view anatomy cannot be directly considered. This study presents an inpainting algorithm using generative adversarial networks (GANs) coupled with partial convolutions to predict out-of-view anatomy to enhance CAL prediction accuracy. METHODS: Retrospective purposive sampling of cases with healthy periodontium and diseased periodontium with bitewing and periapical radiographs and clinician recorded CAL were utilized. Data utilized was from July 1, 2016 through January 30, 2020. 80,326 images were used for training, 12,901 images were used for validation and 10,687 images were used to compare non-inpainted methods to inpainted methods for CAL predictions. Statistical analyses were mean bias error (MBE), mean absolute error (MAE) and Dunn's pairwise test comparing CAL at p=0.05. RESULTS: Comparator p-values demonstrated statistically significant improvement in CAL prediction accuracy between corresponding inpainted and non-inpainted methods with MAE of 1.04mm and 1.50mm respectively. The Dunn's pairwise test indicated statistically significant improvement in CAL prediction accuracy between inpainted methods compared to their non-inpainted counterparts, with the best performing methods achieving a Dunn's pairwise value of -63.89. CONCLUSIONS: This study demonstrates the superiority of using a generative adversarial inpainting network with partial convolutions to predict CAL from bitewing and periapical images. CLINICAL SIGNIFICANCE: Artificial intelligence was developed and utilized to predict clinical attachment level compared to clinical measurements. A generative adversarial inpainting network with partial convolutions was developed, tested and validated to predict clinical attachment level. The inpainting approach was found to be superior to non-inpainted methods and within the 1mm clinician-determined measurement standard.
引用
- Kearney VP, Yansane AM, Brandon RG, et al. A generative adversarial inpainting network to enhance prediction of periodontal clinical attachment level. J Dent. 2022. PMID: 35760207. doi:10.1016/j.jdent.2022.104211
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。