PDF(Pubmed)
Abstract：
Artificial intelligence (AI) has been introduced to interpret the panoramic radiographs (PRs). The aim of this study was to develop an AI framework to diagnose multiple dental diseases on PRs, and to initially evaluate its performance.
The AI framework was developed based on 2 deep convolutional neural networks (CNNs), BDU-Net and nnU-Net. 1996 PRs were used for training. Diagnostic evaluation was performed on a separate evaluation dataset including 282 PRs. Sensitivity, specificity, Youden\'s index, the area under the curve (AUC), and diagnostic time were calculated. Dentists with 3 different levels of seniority (H: high, M: medium, L: low) diagnosed the same evaluation dataset independently. Mann-Whitney U test and Delong test were conducted for statistical analysis (ɑ=0.05).
Sensitivity, specificity, and Youden\'s index of the framework for diagnosing 5 diseases were 0.964, 0.996, 0.960 (impacted teeth), 0.953, 0.998, 0.951 (full crowns), 0.871, 0.999, 0.870 (residual roots), 0.885, 0.994, 0.879 (missing teeth), and 0.554, 0.990, 0.544 (caries), respectively. AUC of the framework for the diseases were 0.980 (95%CI: 0.976-0.983, impacted teeth), 0.975 (95%CI: 0.972-0.978, full crowns), and 0.935 (95%CI: 0.929-0.940, residual roots), 0.939 (95%CI: 0.934-0.944, missing teeth), and 0.772 (95%CI: 0.764-0.781, caries), respectively. AUC of the AI framework was comparable to that of all dentists in diagnosing residual roots (p > 0.05), and its AUC values were similar to (p > 0.05) or better than (p < 0.05) that of M-level dentists for diagnosing 5 diseases. But AUC of the framework was statistically lower than some of H-level dentists for diagnosing impacted teeth, missing teeth, and caries (p < 0.05). The mean diagnostic time of the framework was significantly shorter than that of all dentists (p < 0.001).
The AI framework based on BDU-Net and nnU-Net demonstrated high specificity on diagnosing impacted teeth, full crowns, missing teeth, residual roots, and caries with high efficiency. The clinical feasibility of AI framework was preliminary verified since its performance was similar to or even better than the dentists with 3-10 years of experience. However, the AI framework for caries diagnosis should be improved.
The AI framework was developed based on 2 deep convolutional neural networks (CNNs), BDU-Net and nnU-Net. 1996 PRs were used for training. Diagnostic evaluation was performed on a separate evaluation dataset including 282 PRs. Sensitivity, specificity, Youden\'s index, the area under the curve (AUC), and diagnostic time were calculated. Dentists with 3 different levels of seniority (H: high, M: medium, L: low) diagnosed the same evaluation dataset independently. Mann-Whitney U test and Delong test were conducted for statistical analysis (ɑ=0.05).
Sensitivity, specificity, and Youden\'s index of the framework for diagnosing 5 diseases were 0.964, 0.996, 0.960 (impacted teeth), 0.953, 0.998, 0.951 (full crowns), 0.871, 0.999, 0.870 (residual roots), 0.885, 0.994, 0.879 (missing teeth), and 0.554, 0.990, 0.544 (caries), respectively. AUC of the framework for the diseases were 0.980 (95%CI: 0.976-0.983, impacted teeth), 0.975 (95%CI: 0.972-0.978, full crowns), and 0.935 (95%CI: 0.929-0.940, residual roots), 0.939 (95%CI: 0.934-0.944, missing teeth), and 0.772 (95%CI: 0.764-0.781, caries), respectively. AUC of the AI framework was comparable to that of all dentists in diagnosing residual roots (p > 0.05), and its AUC values were similar to (p > 0.05) or better than (p < 0.05) that of M-level dentists for diagnosing 5 diseases. But AUC of the framework was statistically lower than some of H-level dentists for diagnosing impacted teeth, missing teeth, and caries (p < 0.05). The mean diagnostic time of the framework was significantly shorter than that of all dentists (p < 0.001).
The AI framework based on BDU-Net and nnU-Net demonstrated high specificity on diagnosing impacted teeth, full crowns, missing teeth, residual roots, and caries with high efficiency. The clinical feasibility of AI framework was preliminary verified since its performance was similar to or even better than the dentists with 3-10 years of experience. However, the AI framework for caries diagnosis should be improved.
摘要：
背景：已引入人工智能（AI）来解释全景射线照片（PR）。这项研究的目的是开发一个AI框架来诊断PR上的多种牙科疾病，并初步评估其性能。
方法：AI框架是基于2个深度卷积神经网络（CNN）开发的，BDU-Net和nnU-Net。1996年PR用于培训。在包括282个PR的单独评估数据集上进行诊断评估。灵敏度,特异性，Youden\的索引，曲线下面积（AUC），并计算诊断时间。具有3种不同资历的牙医（H：高，M：中等，L:低)独立诊断相同的评价数据集。采用Mann-WhitneyU检验和Delong检验进行统计学分析(α=0.05)。
结果：灵敏度，特异性，诊断5种疾病的框架和Youden\'s指数分别为0.964、0.996、0.960（阻生齿），0.953,0.998,0.951(全冠),0.871,0.999,0.870(残根),0.885,0.994,0.879(牙齿缺失),和0.554，0.990，0.544（龋齿），分别。疾病框架的AUC为0.980（95CI：0.976-0.983，阻生牙齿），0.975（95CI：0.972-0.978，全冠），和0.935（95CI：0.929-0.940，残余根），0.939（95CI：0.934-0.944，牙齿缺失），和0.772（95CI：0.764-0.781，龋齿），分别。AI框架的AUC与所有牙医诊断残根的AUC相当（p>0.05），其AUC值与M级牙医诊断5种疾病相似（p>0.05）或优于（p<0.05）。但是该框架的AUC在统计学上低于一些H级牙医诊断阻生牙，缺失的牙齿,和龋齿（p<0.05）。框架的平均诊断时间明显短于所有牙医（p<0.001）。
结论：基于BDU-Net和nnU-Net的AI框架在诊断受累牙齿方面表现出高度特异性，全冠,缺失的牙齿,残根,和龋齿效率高。AI框架的临床可行性得到了初步验证，因为其性能与具有3-10年经验的牙医相似甚至更好。然而,应该改进龋齿诊断的AI框架。
方法：AI框架是基于2个深度卷积神经网络（CNN）开发的，BDU-Net和nnU-Net。1996年PR用于培训。在包括282个PR的单独评估数据集上进行诊断评估。灵敏度,特异性，Youden\的索引，曲线下面积（AUC），并计算诊断时间。具有3种不同资历的牙医（H：高，M：中等，L:低)独立诊断相同的评价数据集。采用Mann-WhitneyU检验和Delong检验进行统计学分析(α=0.05)。
结果：灵敏度，特异性，诊断5种疾病的框架和Youden\'s指数分别为0.964、0.996、0.960（阻生齿），0.953,0.998,0.951(全冠),0.871,0.999,0.870(残根),0.885,0.994,0.879(牙齿缺失),和0.554，0.990，0.544（龋齿），分别。疾病框架的AUC为0.980（95CI：0.976-0.983，阻生牙齿），0.975（95CI：0.972-0.978，全冠），和0.935（95CI：0.929-0.940，残余根），0.939（95CI：0.934-0.944，牙齿缺失），和0.772（95CI：0.764-0.781，龋齿），分别。AI框架的AUC与所有牙医诊断残根的AUC相当（p>0.05），其AUC值与M级牙医诊断5种疾病相似（p>0.05）或优于（p<0.05）。但是该框架的AUC在统计学上低于一些H级牙医诊断阻生牙，缺失的牙齿,和龋齿（p<0.05）。框架的平均诊断时间明显短于所有牙医（p<0.001）。
结论：基于BDU-Net和nnU-Net的AI框架在诊断受累牙齿方面表现出高度特异性，全冠,缺失的牙齿,残根,和龋齿效率高。AI框架的临床可行性得到了初步验证，因为其性能与具有3-10年经验的牙医相似甚至更好。然而,应该改进龋齿诊断的AI框架。
