中国实用口腔科杂志 ›› 2024, Vol. 17 ›› Issue (6): 722-726.DOI: 10.19538/j.kq.2024.06.014

• 综述 • 上一篇    下一篇

数字化技术辅助牙隐裂检测研究进展

张艾頔1,曾涵柔2,李长芳1,王    玮1   

  1. 1. 口颌系统重建与再生全国重点实验室,国家口腔疾病临床医学研究中心,陕西省口腔医学重点实验室,空军军医大学口腔医院牙体牙髓病科,陕西 西安 710032;2. 西北工业大学生命学院,陕西 西安 710072
  • 出版日期:2024-11-30 发布日期:2024-11-30
  • 通讯作者: 王玮
  • 基金资助:
    国家自然科学基金面上项目(82370942)

  • Online:2024-11-30 Published:2024-11-30

摘要: 牙隐裂是继龋病和牙周病之后成年人后牙缺失的第三大病因。由于目前传统的检测技术和检测仪器精准度不高,易造成漏诊、误诊而耽误最佳治疗时间,早发现、早诊断对于隐裂牙的存留尤为关键。随着数字化技术在口腔医学领域的广泛应用,牙隐裂的诊断有了新的技术保障。文章就吲哚菁绿辅助近红外光成像、造影剂加强锥形术CT成像、X射线暗场断层扫描成像、数字图像相关法、人工智能技术辅助成像、定量光诱导荧光技术等数字化技术在牙隐裂检测中的应用研究进展做一综述,简要阐明以上技术的优缺点、临床应用现状及前景展望,为未来相关研究方向提供参考依据。

关键词: 牙隐裂, 数字化技术, 吲哚菁绿, 造影剂, X射线暗场断层扫描, 数字图像相关法, 卷积神经网络, 定量光诱导荧光

Abstract: Cracked teeth are the third leading cause after dental caries and periodontal disease of posterior tooth loss in adults. Because the precision of traditional testing technology and testing instruments is not good,it is easy to have missed diagnosis and misdiagnosis,which results in the miss of the best time for treatment. Early detection and early diagnosis are particularly critical for the preservation of cracked teeth. With the wide application of digital technology in dentistry,the diagnosis of cracked teeth has been guaranteed by new technologies. The article provides a review of the research progress on the application of digital techniques in the detection of cracked teeth, including indocyanine green-assisted near-infrared light imaging,contrast-enhanced subconical CT imaging,X-ray dark-field tomography imaging,digital image correlation,artificial intelligence technology-assisted imaging,and quantitative light-induced fluorescence technique. The advantages and disadvantages of the above techniques,the current status of their clinical application,and their future prospects are briefly described to provide a reference for the future direction of related research.

Key words: cracked tooth, digital technology, indocyanine green, contrast medium, X-ray dark-field tomography, digital image correlation, convolutional neural networks, quantitative light-induced fluorescence