【作者】 刘江；

【导师】 吴荒；

【作者基本信息】 吉林大学 ， 临床医学硕士（专业学位）， 2024， 硕士

【摘要】 目的:测量立体视觉可以在真实环境下进行,也可以在双眼分视状态下进行,研究人员发现前者测得的结果往往好于后者。但由于各种检查方法之间,测试图形的设计、视差的级距和极值不同,可能会对这种对比造成影响。本研究采用自主设计的立体视觉检测方法,将测试视标形状、测试级距与极值、测试距离等因素统一,明确在真实状态和双眼分视状态下,立体视锐度是否存在差异。方法:建立基于综合验光仪和4K智能手机的立体视觉测试系统,和小尺寸Frisby立体视觉测试系统(FNS),对30名受试者进行测量。在确定其远距离屈光状态后,于0.65m处确定测试所需的视力状态。遮盖左眼,在右眼前加正镜,确定视力达到0.5log MAR时最高的正镜(最低的负镜)度数并记录(设定为组6);然后去雾视,分别记录视力达到0.4log MAR(组5)、0.3log MAR(组4)、0.2log MAR(组3)、0.1log MAR(组2)和0log MAR(组1)时的球镜度。遮盖右眼,在左眼前加正镜,确定并记录视力达到0log MAR时最高的正镜(最低的负镜)度数。分别测量真实状态下(小尺寸FNS)和双眼分视状态下(智能手机)的立体视锐度。采用棱镜调整受试者的视轴,使之在0.65m处呈现平行状态,而后测量组1在小尺寸FNS的立体视锐度。结果:正态性检验结果显示,本研究中绝大部分数据不符合正态分布,故均采用非参数检验进行分析。在保证左眼视力0log MAR的情况下,右眼的视力变化从0.5log MAR到0log MAR。双眼视力差在0.2log MAR及以下时,小尺寸FNS的测量结果明显好于4K智能手机(Wilcoxon signed-rank检验:组1,Z=-3.879,P<0.001;组2,Z=-3.478,P=0.001;组3,Z=-3.977,P<0.001)。而当双眼视力差距超过0.2log MAR时,两者的测量结果则无统计学差异(组4,Z=-1.880,P=0.060;组5,Z=-1.784,P=0.074;组6,Z=-1.812,P=0.070)。对于组1,采用棱镜将双眼视轴调整平行后,小尺寸FNS组与智能手机组(Z=-3.879,P<0.001)、小尺寸FNS组与FNS加棱镜组(Z=-4.396,P<0.001)测量结果有显著性差异;而智能手机组与FNS加棱镜组(Z=-0.021,P=0.983)测量结果无显著性差异。结论:真实环境下立体视觉的测量结果明显优于双眼分视状态下的测量结果。双眼的集合功能,可能是造成这种差异的主要原因。更多还原

【Abstract】 Objective:This study aims to categorize stereopsis measurement into two distinct approaches: measurements conducted in a real-world environment and those performed under dichoptic presentation conditions.There is a prevailing view among researchers that real environment measurements yield more accurate results than those obtained via dichoptic presentation.However,significant variations exist among the different commercial tools available for stereopsis measurement.These variations result in inconsistent stereoacuity values,complicating accurate comparative analysis.To address this issue,we have developed a novel testing system,which incorporates the use of smartphones and phoropters.This system ensures uniformity in pattern size and shape,test step ranges,extreme values,and testing distance.The primary focus of this study is to rigorously investigate whether there are notable differences in stereopsis values obtained through real-world measurements as opposed to dichoptic presentation.Methods:This study employs two stereopsis testing systems: one based on a phoropter with two 4K smartphones,and the other utilizing a modified Frisby Near Stereotest(FNS).A total of 30 subjects were enrolled in the study.Initially,each subject’s distance refraction was determined,followed by near vision testing at a distance of 0.65 m.The testing process involved sequentially covering the left eye and placing positive lenses before the right eye,with the aim of determining the highest positive diopter that allowed the subject to achieve a visual acuity(VA)of 0.5log MAR.This diopter level was recorded and categorized as Group6.Subsequently,the lens power was gradually reduced(’unfogging’),and the spherical diopter corresponding to VAs of 0.4log MAR,0.3log MAR,0.2log MAR,0.1log MAR,and 0log MAR were recorded,forming Groups5 to 1 respectively.The same procedure was then repeated for the right eye,with the objective of identifying and recording the highest positive diopter when the VA reached 0log MAR.Stereoacuity was assessed in two conditions: a real-world scenario using the modified FNS,and a separated eyes condition utilizing the smartphone-based system.To ensure accurate measurement,prisms were utilized to align the visual axes of the subjects parallelly at a distance of0.65 m.Specifically,the stereoacuity of Group 1 in the modified FNS was measured with the assistance of prisms.Results:The normality test conducted on the data from this study indicated that most of the data did not conform to a normal distribution.Consequently,non-parametric statistical methods were employed for analysis.Throughout the study,the visual acuity(VA)of the left eye was consistently maintained at 0log MAR,while the VA of the right eye varied from 0.5log MAR to 0log MAR.When the discrepancy in VA between the two eyes was 0.2log MAR or less,the stereoacuities measured by the modified Frisby Near Stereotest(FNS)were significantly superior to those obtained using the4 K smartphones.This was evidenced by the Wilcoxon signed-rank test results for Group 1(Z=-3.879,P<0.001),Group 2(Z=-3.478,P=0.001),and Group 3(Z=-3.977,P<0.001).Conversely,when the VA difference between the two eyes exceeded 0.2log MAR,no statistically significant difference was observed between the two measurement methods,as indicated in Group 4(Z=-1.880,P=0.060),Group 5(Z=-1.784,P=0.074),and Group 6(Z=-1.812,P=0.070).For Group 1,significant differences were found between the modified FNS measurements with and without the aid of prisms,and between these measurements and those obtained using the smartphone-based system.Specifically,the Wilcoxon signed-rank test showed significant differences between the modified FNS group and the smartphone group(Z=-3.879,P<0.001),and between the modified FNS group and the FNS plus prism group(Z=-4.396,P<0.001).However,there was no significant difference between the smartphone group and the FNS plus prism group(Z=-0.021,P=0.983).Conclusions:The study concludes that stereoacuity measured in a real-world environment is significantly better than that measured in dichoptic presentation conditions.The primary factor contributing to this difference is likely the vergence mechanism of binocular vision.更多还原