Graduate School of Science and Engineering, Ritsumeikan University, 1-1-1 Noji-higashi, Kusatsu, Shiga, 525-8577, Japan.
TAKARA TELESYSTEMS Corporation, 1-17-17 Nihonbashi, Chuo-ku, Osaka, 542-0073, Japan.
Radiol Phys Technol. 2021 Mar;14(1):64-69. doi: 10.1007/s12194-020-00603-1. Epub 2021 Jan 5.
Panoramic dental X-ray imaging is an established method for the diagnosis of dental problems. However, the resolution of panoramic dental X-ray images is relatively low. Thus, early lesions are often overlooked. As the first step in the development of a computer-aided diagnosis scheme for panoramic dental X-ray images, we propose a computerized method for the segmentation of teeth using U-Net with a loss function weighted on the tooth edge. Our database consisted of 162 panoramic dental X-ray images. The training dataset consisted of 102 images, while the remaining 60 images were used as the test dataset. The loss function obtained by the cross entropy (CE) in the entire image is usually used in training U-Net. To improve the segmentation accuracy of the tooth edge, a loss function weighted on the tooth edge is proposed by adding the CE in the tooth edge region to the CE for the entire image. The mean Jaccard index and Dice index for U-Net with the loss function combining the CEs for the entire image and tooth edge were 0.864 and 0.927, respectively, which were significantly larger than those for U-Net with the CE for the entire image (0.802 and 0.890, p < 0.001) and U-Net with the CE for the tooth edge (0.826 and 0.905, p < 0.001). U-Net with the new loss function exhibited a higher segmentation accuracy of the tooth in panoramic dental X-ray images than that obtained by U-Net with the conventional loss function.
全景牙科 X 光成像已被广泛应用于牙科疾病的诊断。然而，全景牙科 X 光图像的分辨率相对较低，因此早期病变常常被忽略。作为开发全景牙科 X 光图像计算机辅助诊断方案的第一步，我们提出了一种基于 U-Net 的计算机牙齿分割方法，该方法使用加权于牙边缘的损失函数。我们的数据库包含 162 张全景牙科 X 光图像。训练数据集由 102 张图像组成，而其余 60 张图像则用于测试数据集。在训练 U-Net 时，通常使用整个图像的交叉熵 (CE) 获得损失函数。为了提高牙边缘的分割精度，通过在牙边缘区域添加 CE 到整个图像的 CE，提出了加权于牙边缘的损失函数。U-Net 结合整个图像和牙边缘的 CE 的平均 Jaccard 指数和 Dice 指数分别为 0.864 和 0.927，显著大于仅使用整个图像的 CE 的 U-Net（0.802 和 0.890，p < 0.001）和仅使用牙边缘的 CE 的 U-Net（0.826 和 0.905，p < 0.001）。与传统的损失函数相比，新的损失函数使得 U-Net 在全景牙科 X 光图像中具有更高的牙齿分割精度。
