![]() This study followed the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of the ZOC at Sun Yat-sen University (IRB-ZOC-SYSU). The data were collected from a national project for CC treatment and research: the CC Program of the Chinese Ministry of Health (CCPMOH). Materials and MethodsĪ prospective study was conducted at the Zhongshan Ophthalmic Center (ZOC), Guangdong, China, from June 2011 to February 2019. This system can help ophthalmologists and guardians monitor patients’ visual development ( Long et al., 2017) and adopt necessary visual intervention ( Sniatecki et al., 2015 Wang and Xiao, 2015) in time, thereby contributing to the visual prognosis of young children. In our study, based on clinical data, features of OCT images, and follow-up results of children with childhood cataract (CC), we developed a machine learning system for long-term visual prediction of best corrected visual acuity (BCVA) 3 and 5 years in advance. OCT images have been applied to predict prognostic visual function in age-related macular degeneration (AMD) ( Rohm et al., 2018) and have achieved excellent performance in visual prediction. Existing research ( Rohm et al., 2018) has focused on short-term visual prediction for adults within a year.įundus imaging, especially optical coherence tomography (OCT), is recognized as a key factor for visual prediction ( Guo et al., 2017 Esaka et al., 2019 Park et al., 2020). To date, no children-applicable technology for VA prediction has been reported. However, the nature of ocular growth and myopia drift in young children may disrupt exact visual prediction and affect result accuracy. Exact VA prediction is beneficial for young children, especially children with ophthalmopathy, based on which ophthalmologists and guardians can determine the potential visual prognosis in advance, decide on an intervention plan, and contribute to visual development. Consequently, the results of visual prediction are meaningful by reflecting the tendency and speed of visual development during a future period. Normal visual development and visual acuity (VA) are important for young children and are the basis of infantile brain development ( Stjerna et al., 2015 Danka Mohammed and Khalil, 2020) and ability development ( Wu et al., 2019 Havstam Johansson et al., 2020). The application of our research contributes to the design of visual intervention plans and visual prognosis. This work establishes a reliable method to predict prognosis 5 years in advance. This is the first study to predict post-therapeutic BCVAs in young children. The mean absolute errors (MAEs) between the prediction results and ground truth were 0.1482–0.2117 logMAR for 3-year predictions and 0.1198–0.1845 logMAR for 5-year predictions the root mean square errors (RMSEs) were 0.1916–0.2942 logMAR for 3-year predictions and 0.1692–0.2537 logMAR for 5-year predictions. In the BCVA predictions, small errors within two lines of the visual chart were achieved. Six machine learning algorithms were applied. Two hundred eyes of 132 patients were included. In our study, we developed an intelligent system based on the features of optical coherence tomography images for long-term prediction of best corrected visual acuity (BCVA) 3 and 5 years in advance. The results of visual prediction reflect the tendency and speed of visual development during a future period, based on which ophthalmologists and guardians can know the potential visual prognosis in advance, decide on an intervention plan, and contribute to visual development. 2Center of Precision Medicine, Sun Yat-sen University, Guangzhou, China.1State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, China.Yifan Xiang 1†, Jingjing Chen 1†, Fabao Xu 1†, Zhuoling Lin 1, Jun Xiao 1, Zhenzhe Lin 1* and Haotian Lin 1,2*
0 Comments
Leave a Reply. |