【作者】 杜持新；

【导师】 顾扬顺；

【作者基本信息】 浙江大学 ， 临床医学， 2012， 博士

【摘要】 目的构建大量程光学相干层析成像系统(ultra-long scan depth optical coherence tomography,UL-OCT),对整个眼前节实时成像,检验其在不同调节状态测量的重复性和可靠性,并用于调节时眼前节各生物参数的测量和比较。方法构建基于迈克耳逊干涉仪的UL-OCT系统,该系统的光源采用中心波长为840nm,带宽为50 nm的超发光二级管SLD,利用特殊设计的参考臂上两个反射镜的转换,改变了零光程位置,再将两次干涉取得的信号图像叠加,实现了一次扫描得到眼前节的完整图像。该方法不仅保留了较高的分辨率而且成功补偿了单幅图性性噪比的下降。本系统光输出功率为1.0 mW,受CCD相机像素大小的制约扫描速度为4KHz；成像深度为12.055mm；轴向和横向分辨率分别约为11.0μm和20μm。用构建的UL-OCT对41例(41眼)平均年龄33.5±6.9岁,平均等效球镜度-2.5±2.6(D)的正常人在最小和最大调节状态下的眼前节进行重复成像,并用自行设计的软件对图像进行矫正和测量。测量参数包括：中央角膜厚度、前房宽度和深度、瞳孔直径、晶状体前后的曲率半径、晶体厚度、晶状体中心位置和眼前节长度。运用SPSS 16.0对数据进行统计分析。所有参数以平均数±标准差显示,以配对t检验比较同一调节状态下两次重复检查间和不同调节状态下的区别,以重复性系数和组内相关系数来表示测量的重复性和可靠性。结果在最大和最小生理调节状态下两次检查间的所有参数无显著性差别(p,0.05)。因此,将两次检查的平均值作为其各自调节状态的测量值以用于不同调节状态间的比较。在最小调节和最大调节状态下中央角膜厚度、前房宽度和深度、晶体厚度、晶状体中心位置和眼前节长度测量的可重复系数极好,在1.23%～3.59%之间；瞳孔直径测量的可重复系数较高,在最小和最大调节状态分别为18.90%和21.63%；水平方向晶状体前后的曲率半径测量的可重复性中等至相当,为34.86%～42.70%。在最小和最大调节状态中央角膜厚度、前房深度、晶体厚度、晶状体中心位置和眼前节长度测量的组内相关系数极佳,为0.969～0.998；瞳孔直径和前房宽度测量的相当高,为0.874～0.925；而晶状体前后的曲率半径的则中等至相当,为0.251～0.669。最大调节时瞳孔直径、前房深度、晶状体前后表面的曲率半径和晶状体中心位置明显减小,晶状体厚度和眼前节长度明显增大(均P<0.01),而中央角膜厚度和前房宽度无明显变化(P=0.0602,0.8345)。结论本研究构建的UL-OCT可对调节时整个眼前节包括角膜至晶状体后表面实时成像,该系统在人眼调节眼前节生物参数的测量中具有很好的重复性和可靠性。调节时晶状体厚度增加,前后表面变凸,整个晶状体中心位置前移,同时伴随有前房变浅和瞳孔缩小。更多还原

【Abstract】 PurposeTo measure by ultra-long scan depth optical coherence tomography (UL-OCT) dimensional changes in the anterior segment of human eyes during accommodation.MethodsUtilizing a fiber-based Michelson interferometer, the UL-OCT system was constructed with a superluminescent diode laser-based light source with center wavelength of 840 nm and full-width at half maximum bandwidth of 50 nm. An alternative placement of the zero-delay line on the top and bottom in the reference arm was specially designed to acquire multiple images in one scan. Image enhancement was realized by overlapping the two acquired images. This method was used to compensate the drop of signal-to-noise ratio through the entire image depth. The light was focused on the anterior segment, and the total exposure power at the surface of the cornea was 1.10 mW. The scan speed was set to 4,000 A-scans/sec, a limitation imposed by the charged-couple device (CCD) line scan camera. The scan depth of the system was 12.055 mm. The resolution of the UL-OCT system was measured as 11.Oμm axially and approximately 20μm transversely for 2,048 pixels in the eye.UL-OCT was used to image the cornea to the back surface of the crystalline lens. 41 right eyes of healthy subjects with a mean age of 33.5±6.9 years (range,22-41 years) and a mean refraction of-2.5±2.6 diopters (D) (range,+0.50D～-7.00D)were imaged in two repeated measurements at minimal and maximal accommodation. Custom software corrected the optical distortion of the images and yielded the measurements. The dimensional results included central corneal thickness (CCT), anterior chamber depth and width (ACD, ACW), pupil diameter (PD), horizontal radii of the lens anterior and posterior surface curvatures (LAC, LPC), lens thickness (LT), lens central position (LCP) and anterior segment length (ASL).Statistical Procedures for the Social Sciences 16.0 was used for descriptive statistics and data analysis. All data were presented as means±standard deviations. Paired t-tests were for comparisons between repeated measures of the same accommodative state and between the two different accommodative states. Coefficient of repeatability (COR) and the intraclass correlation coefficient (ICC) were calculated to evaluate the repeatability and reliability.ResultsFor all anterior segment dimensions, there were no significant differences between the two repeated measurements of the same accommodative state (paired t-tests, all p-values> 0.05). The average of the two measurements for each accommodative state was used to determine if there were differences between two states. The CORs for CCT, ACW, ACD, LT, LCP, and ASL were excellent, ranging from 1.23% to 3.59% in the minimal and maximal accommodative states. The CORs for PD were relatively high, 18.90% and 21.63% in minimal and maximal accommodative states respectively. The CORs for LAC and LPC ranged from 34.86% to 42.70% in both accommodative states. The ICCs for CCT, ACD, LT, LCP, and ASL were very high, varying between 0.969 and 0.998. They were slightly lower for PD and ACW,0.874 to 0.925, and showed moderate and fair reliability for LAC and LPC (0.251 to 0.669). Compared to minimal accommodation, PD, ACD, LAC, LPC and LCP decreased and LT and ASL increased significantly at maximal accommodation (P< 0.01), while CCT and ACW did not change (P= 0.0602,0.8345) ConclusionsUL-OCT system successfully imaged accommodative change of the whole anterior segment from the cornea to the back surface of the crystalline lens in vivo. UL-OCT measured changes in anterior segment dimensions during accommodation with good repeatability and reliability. During accommodation, the anterior and posterior surface of the crystalline lens became steeper as the lens moved forward, with decrease of the anterior chamber depth and constriction of the pupil.更多还原