Purpose: To investigate the association between myopia and ganglion cell layer and inner plexiform layer (GCIPL) in diabetic patients without clinically diabetic retinopathy (DR). Methods: Guangzhou Diabetic Eye study was a longitudinal study. A total of 1165 adults aged 30-80 years were recruited in this study which divided into 3 groups according to the presence of myopia (spherical equivalence, SE≤-3 diopters) and diabetics: healthy group (n=508); diabetes mellitus group (n=525); diabetes mellitus (DM)+myopia group(n=132). GCIPL was measured via swept-source optical coherence tomography. Univariable model and multivariable model were used to showing the association of GCIPL change and baseline parameters. Results: SE was 1.07±1.06D in healthy group, 1.02±1.00D in diabetes mellitus group and -5.36±2.30 D in DM + myopia group(P<0.001). The baseline GCIPL thickness were 71.1±0.3μm, 74.4±0.2 μm, 71.7±0.5μm, respectively. The slope of GCIPL thickness was -0.10 (95%CI: 0.05, -2.03) μm/year in healthy group, which was 12 folds faster than those in diabetes group (-1.21 [95%CI: 0.05, -24.04], P<0.001) μm/year and 22 folds higher among those in DM + myopia group (-2.17 [95%CI: 0.10, -21.63], P=0.009) μm/year. Conclusions: Both myopia and diabetes status accelerate macular ganglion cell layer and inner plexiform layer thinning in diabetic patients without diabetic retinopathy.
【 Objective 】 Taking UNSILO platform as a research case, this paper discusses the application of Artificial Intelligence (AI) technology in academic publishing, in order to provide reference for Chinese sci-tech journals. 【 Methods 】 The functions and technical characteristics of UNSILO platform were systematically summarized, and the cooperation cases between UNSILO platform and publishing institutions were deeply analyzed. 【 Result 】 UNSILO platform supports intelligent evaluation of papers and intelligent classification and recommendation of subject content, and its concept extraction technology has obvious advantages in academic publishing. UNSILO has advanced concept extraction technology and friendly and simple interface operation, which provides intelligent solutions for human-computer interaction for many publishing institutions, including Springer Nature. 【 Conclusion 】 AI technology has gradually penetrated into the publishing process of sci-tech journals, and will inject new vitality into the publishing of sci-tech journals.
