深度学习在数字化根尖片中的研究进展

Research Progress of Deep Learning in Digital Periapical Radiograph

摘要: 数字化根尖片是口腔医生进行疾病筛查、疾病诊断、治疗评估等的重要辅助工具。传统的影像图片分析是依靠临床医生的经验和影像拍摄的技术。人工判读存在主观性强、早期病变易漏诊、效率低下等局限。深度学习(Deep Learning, DL)凭借卷积神经网络(CNN)等模型强大的图像特征提取与模式识别能力，目前广泛应用于口腔医学领域。深度学习模型被开发应用于牙齿识别分割、疾病检测、疾病预后评估等方面。许多研究表明，深度学习技术在口腔数字化根尖片中的应用已经取得了重大进展，准确性已经达到临床医生的水平。然而，在某一些领域，其精度仍有待提高。本文综述了深度学习在口腔医学常用的数字化根尖片影像中的应用现状，强调其潜力，并提出了未来的研究方向。

Abstract: Digital periapical radiograph is an important auxiliary tool for dentists to carry out disease screening, disease diagnosis, treatment evaluation and so on. Traditional image analysis relies on the experience of clinicians and the technology of image shooting. Manual interpretation has limitations such as strong subjectivity, easy missed diagnosis of early lesions, and low efficiency. With the powerful image feature extraction and pattern recognition capabilities of models such as convolutional neural network (CNN), deep learning (DL) is widely used in the field of stomatology. Deep learning models have been developed and applied to tooth recognition and segmentation, disease detection, and disease prognosis evaluation. Many studies have shown that the application of deep learning technology in oral digital periapical radiographs has made significant progress, and the accuracy has reached the level of clinicians. However, in some fields, its accuracy still needs to be improved. This article reviews the application status of deep learning in digital periapical radiograph commonly used in stomatology, emphasizes its potential, and proposes future research directions.