Abstract
Purpose :
Myopia, the dominant form of the ocular disorder, is believed to be heavily influenced by environmental risk factors (ERF) such as nearwork and inadequate time spent outdoors. However, the precise roles of these factors during myopia development are unknown due to the lack of quantification methods. This study proposes an ERF index quantified from continuous monitoring of dynamically viewing distance (VD) and eye-received illuminance (EI) by a novel device ClouclipTM.
Methods :
ClouclipTM was developed to be attached on the spectacle and to continuously measure VD and EI every 5 and 20 seconds respectively. 179 children aged between 10 and 12 were recruited from 3 primary schools in both urban and rural areas. The measurements were taken in 5 days within 1 week for each subject and refractive error (RE) was also recorded. The time series of VD and EI were denoised with a bandpass filter and were mapped to a 2-dimension space, which was divided into a 40-by-40 grid (Figure 1a), so that each pixel in the grid represent an ERF configuration. The percentage of time (PoT) falling into each pixel was calculated for each individual. Weighted linear regression was carried out between PoT and RE and the significant slopes (p<0.05) represent the 'protectiveness' of myopia at the corresponding ERF configuration. Given the VD and EI measurements of a new subject, the ERF index is calculated as the sum of product between its PoT and the significant slopes at corresponding locations. The relation between ERF index and RE is evaluated on the independent subject in cross validation.
Results :
In the VD-EI map, 827 out of 1600 pixels were significantly related to the RE (Figure 1b). The VD-EI map shows that large VD and EI in the upper right corner is positively related to RE while low VD (VD<35cm) area is negatively related to RE (Figure 1b). The ERF index of all subjects were shown in Figure 1c and the median (95% confidence interval) of the ERF index is 2.622 ([0.025, 0.975]). ERF index and RE are positively related with statistical significance (p=0.031; Figure 1d). The slope of the regression is 0.058.
Conclusions :
The significant relationship between ERF index and RE make it a possible quantifier of ERF, which can be used to follow the development of myopia over time.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.