ABSTRACT
@#AIM:To construct an artificial intelligence(AI)system based on convolutional neural network(CNN), which can automatically evaluate the morphological changes of meibomian gland(MG)in meibomian gland dysfunction(MGD)patients. <p>METHODS:The right eyes of 145 subjects who were treated at the Hangzhou Branch of the Eye Hospital of Wenzhou Medical University from January to November 2021 were selected for inclusion in the study. Meibography images of 60 of these subjects were randomly selected for AI training. The meibomian region and each MG in meibography were annotated and formed into datasets. The datasets were used for training and obtaining an AI system based on residual neural network(ResNet)combined with the U-Net model. The AI system was used to automatically analyze the MG morphological parameters in 85 subjects, including 53 patients with obstructive MGD and 32 volunteers with normal meibomian glands. The clinical indices including ocular surface disease index(OSDI), tear meniscus height, tear film break-up time(TBUT), corneal fluorescein staining, lid margin score, meiboscore, and meibomian gland expressibility score were also observed. The correlation between MG morphological parameters and clinical indices were analyzed.<p>RESULTS: After several iterations, we finally obtained an AI system with Intersection over Union of 92.0%. Using this AI system, we found that there was a significant correlation between the MG density in the upper eyelid with OSDI(<i>r</i><sub>s</sub>=-0.320), TBUT(<i>r</i><sub>s</sub>=0.484), lid margin score(<i>r</i><sub>s</sub>=-0.350), meiboscore(<i>r</i><sub>s</sub>=-0.749), and meibum expressibility score(<i>r</i><sub>s</sub>=0.425)(all <i>P</i><0.05). The MG density in the lower eyelid was significantly correlated with OSDI(<i>r</i><sub>s</sub>=-0.420), TBUT(<i>r</i><sub>s</sub>= 0.598), lid margin score(<i>r</i><sub>s</sub>=-0.396), meiboscore(<i>r</i><sub>s</sub>=-0.720), and meibum expressibility score(<i>r</i><sub>s</sub>=0.438)(all <i>P</i><0.05). The MG density in the total eyelid was significantly correlated with OSDI(<i>r</i><sub>s</sub>=-0.404), TBUT(<i>r</i><sub>s</sub>=0.601), lid margin score(<i>r</i><sub>s</sub>=-0.416), meiboscore(<i>r</i><sub>s</sub>=-0.805), and meibum expressibility score(<i>r</i><sub>s</sub>=0.480)(all <i>P</i><0.05).<p>CONCLUSION:The AI system based on CNN in this study is an accurate and efficient MG morphological evaluation system, which can be conveniently used to evaluate the MG morphology of MGD patients quickly and accurately by using the MG density index established by us. MG density is a new quantitative index to evaluate meibomian gland atrophy, which is more accurate than meiboscore.
ABSTRACT
OBJECTIVE: To investigate the interaction of occupational vanadium exposure and heat shock protein 70-hom( HSP70-hom) gene polymorphism on the oxidative stress level of workers. METHODS: By judgment sampling,87 workers from vanadium products plant as vanadium exposure group and 63 workers from chemical industry as control group were enrolled into this investigation. The activities of total superoxide dismutase( T-SOD),manganese-containing superoxide dismutase( Mn-SOD),copper and zinc-containing superoxide dismutase( Cu Zn-SOD) and inducible nitric oxide synthase( iNOS),and the level of malondialdehyde( MDA) in the serum of studied subjects were detected. The HSP70-hom genotype was examined by restricted fragment length polymorphism-polymerase chain reaction. Based on the above data,the comprehensive impact of HSP70-hom gene polymorphism and vanadium exposure on workers' oxidative stress level was analyzed. RESULTS: Compared with the control one,the activities of serum T-SOD and Mn-SOD of vanadium-exposed group were significantly decreased( P < 0. 01),while the level of serum MDA was significantly increased( P < 0. 01).Furthermore,the serum T-SOD activity of T / T genotype was statistically lower than that of T / C + C / C genotype( P <0. 01). Both high serum( T-SOD) activity and high MDA level were significantly influenced by the interaction between vanadium exposure and HSP70-hom polymorphism by univariate Logistic regression analysis( P < 0. 05). Using multiple Logistic regression and crossover analysis,multiplicative and additive interactions on high serum( T-SOD) activity between vanadium exposure and HSP70-hom polymorphism was observed [odds ratio( OR) was 6. 051,95% confidence interval( CI) : 1. 001-36. 572,P = 0. 05]and the attributable proportion due to interaction was 0. 935( 95% CI: 0. 086-1. 783).However,high serum MDA level was significantly associated with vanadium exposure [OR( 95% CI) : 6. 352( 3. 099-13. 021),P < 0. 01] without interaction observed. There was no significant effect of vanadium exposure or HSP70-hom polymorphism on either Cu Zn-SOD or iNOS. CONCLUSION: Vanadium exposure might have an impact on the oxidative stress level of the workers by regulating the activity of serum SOD and the level of serum MDA. Besides,vanadium exposure and HSP70-hom polymorphism could interactively interfere with serum T-SOD activity of workers.