目的: 基于虚拟现实牙科模拟系统UniDental开发龋坏识别软件,应用于口腔医学专业本科生临床前期龋坏识别教学,提高学生识别龋坏组织的能力。方法: 基于UniDental系统进行龋坏识别软件开发,在软件中建立抽象龋洞模型,将龋洞分为浅、中、深3层,各层内赋予相同的硬度值和粗糙度值,其中硬度值随深度增加而增大,粗糙度值随深度增加而减小。被试选取北京大学口腔医学院2014级口腔医学专业临床前期教学阶段学生64人,使用Unidental系统的龋坏识别软件,操作力反馈手柄,分别在龋洞模型的浅层、中层、深层进行垂直向硬度探查和水平向粗糙度探查操作培训,完成培训后进行问卷调查,以1~5分评价各项力反馈效果的真实性和对龋坏识别能力提高的帮助度,选择个人倾向的教学模式,并进行相应的统计学分析。结果: 各层硬度、粗糙度及其各自变化梯度评分的中位数均为4,表示比较真实;浅层、中层、深层硬度评分结果对硬度变化梯度评分的影响均有统计学意义(P<0.05);浅层、中层粗糙度评分结果对粗糙度变化梯度评分的影响有统计学意义(P<0.05);本实验对学生能力的提高比较有帮助(中位数为4), 各层硬度和硬度变化梯度评分结果对学生能力提高帮助度的影响有统计学意义(P<0.05);90.4%的学生支持传统离体牙教学模式。结论: 虚拟现实技术对学生龋坏识别能力提高比较有帮助,但无法取代传统教学, 可作为传统离体牙教学模式的补充。

Objective: To develop a software based on “UniDental” system which is a virtual reality dental simulation system and applied to undergraduate majoring in stomatology to improve the ability of identifying caries.Methods: A software was developed applying to identify virtual dental caries based on UniDental system. In the software, a virtual dental caries model was designed and carious tissue was separated to 3 layers by the depth. The stiffness was the same within each layer which was increasing gradually layer by layer. The roughness was also the same within each layer which was decreasing gradually layer by layer. Sixty-four participants in pre-clinical stage of the class of 2014 majoring in stomatology from Peking University School of Stomatology were trained with the software. During the training, the students should probe on the virtual dental carious tissue layer by layer and feel the difference of vertical stiffness and horizontal roughness of each layer by using a handpiece with realistic force feedback. After training, a questionnaire survey was conducted to evaluate the software including a score of 1-5 for haptic fidelity of stiffness and roughness and their relevant gradient and benefit of improving the ability of identifying caries, choosing the preferred training method. The data were statistically analyzed using Kruskal-Wallis test.Results: The median of subjective evaluation scores of the proposed metrics were all “4”, demonstrating that the software operated above medium fidelity. The stiffness scores of all 3 layers were statistically significant (P<0.05) on the stiffness gradient score. The roughness scores of the 1st and 2nd layers were statistically significant (P<0.05) on the roughness gradient score. The training was helpful to improve the ability of identifying caries (median was 4). The scores of all 3 layers stiffness and relevant gradient were statistically significant (P<0.05) on the score of benefit of improving the ability of identifying caries. 90.4% of the participants preferred the traditional extracted teeth training method.Conclusion: The virtual reality dental simulation system was helpful to improve students’ ability of identifying caries. It couldn’t replace the traditional extracted teeth training method by now, it should be used as a supplement to the traditional training method.