个体日光紫外线暴露的评价及其与老年认知功能障碍的关系

Accurate evaluation of long-term ultraviolet (UV) exposure is very important in studies of the relationship between UV exposure and its late biological effect. The frequently used tool for evaluating human UV exposure is the retrospective questionnaire. However, its disadvantages such as time consuming and accuracy of memory limit its application in cognitive impairment subjects. In this study, an algorithm of extracting morphological and texture characteristic of skin will be developed by computer programming. Digital images of sun-exposed skin and unexposed skin from healthy subjects will be collected by skin detector and analyzed by the above method. Intrinsic aging will be removed and photoaging will be kept by the subtraction of the skin indexes between sun-exposed skin and unexposed skin. Then, a long-term cumulative UV exposure prediction model could be built by using skin texture indexes combined with outdoor activity investigation. It could solve the critical issue of objectively measuring the long-term UV exposure in epidemiological studies. Currently, it is not feasible to measure the long-term vitamin D level of human. By using above method, we could explore the association between vitamin D insufficiency caused by lack of UV exposure and cognitive impairment in the old. Thus, we will carry out a screening of cognitive impairment in the old aged above 60 years old in the communities. Digital images of skin will be taken and analyzed for screened out subjects with low cognitive test scores and their control subjects. Then, the long-term UV exposure level will be evaluated by using the above model for each subject. Finally, the association between UV exposure and cognitive impairment in the old will be analyzed after adjusting for the diet history of vitamin D.

紫外线远期生物学效应的研究中，准确评价机体长期紫外线暴露状况是关键，而目前常用的问卷调查方法耗时长、对回忆准确性要求高，不适于认知功能受损类研究对象。本研究拟通过计算机编程，建立皮肤纹理形态和特征分析算法，对收集的正常人群日光紫外线暴露部位和非暴露部位皮肤数字图像进行处理并对比分析，剔除机体自然老化而保留光老化对皮肤的影响，结合户外活动状况调查，构建以皮肤纹理变化定量个体长期累积日光紫外线暴露水平模型，解决客观定量机体长期紫外线暴露的关键问题。在目前很难测量机体长期维生素D水平的情况下，以此为基础还可以从另一角度探索紫外线暴露不足致维生素D长期缺乏与老年认知功能障碍的关系。因此，本研究进一步开展社区老年人认知功能状况调查，对筛查出的认知功能得分较低者及其对照人群采集皮肤图像并分析，通过估计长期累积紫外线暴露水平并控制维生素D摄入膳食史，分析紫外线暴露与老年认知功能障碍的关系。

紫外线的远期生物学效应研究中，准确评价机体的长期紫外线暴露状况是关键。有研究使用环境紫外线水平来衡量个体的长期暴露，但这种方法不能排除如生活习惯和职业等个体因素的影响。目前较常用的回顾性问卷调查方法，耗时长、对回忆的准确性要求高，不适于老年人及认知功能受损的调查对象。本研究通过计算机编程，建立了皮肤纹理形态和特征分析的算法，实现了对皮肤图像的批量转化和处理。通过对皮肤纹理指标的筛选，结合个体日光紫外线暴露状况调查，利用主成分回归，构建了个体长期累积日光紫外线暴露定量模型。本研究进一步开展了对老年人的调查，评估被调查对象的认知功能状况、日光暴露习惯以及富含维生素D食物的食用情况，通过聚类分析的方法，将调查对象分为高暴露组、中等暴露组和低暴露组，结果发现，户外活动时间长、较少使用防晒措施并习惯在上午及午后进行户外活动的个体，其认知功能水平较好。使用构建的皮肤纹理参数提取方法，对老年人暴露部位和非暴露部位皮肤进行分析，得到两部位的皮肤纹理参数，并根据构建的个体长期累积日光紫外线暴露评估模型，评估老年人的累积紫外线暴露水平。经多元有序logistics回归分析发现，对于教育年限大于6年的调查对象，累积紫外线暴露高的个体认知水平更好。本研究建立的以皮肤纹理自动化检测定量个体长期累积日光紫外线暴露的方法可以作为一种基础手段，在探索紫外线生物学效应的各种研究中具有广泛的应用前景。另一方面，鉴于日光紫外线暴露及维生素 D 摄入的可控性，本研究证实长期累积紫外线暴露较低与老年认知功能障碍存在关联，对认知功能障碍的人群预防具有十分重要的意义。