ACCV: automatic classification algorithm of cataract video based on deep learning.

PURPOSE: A real-time automatic cataract-grading algorithm based on cataract video is proposed. MATERIALS AND METHODS: In this retrospective study, we set the video of the eye lens section as the research target. A method is proposed to use YOLOv3 to assist in positioning, to automatically identify the position of the lens and classify the cataract after color space conversion. The data set is a cataract video file of 38 people's 76 eyes collected by a slit lamp. Data were collected using five random manner, the method aims to reduce the influence on the collection algorithm accuracy. The video length is within 10 s, and the classified picture data are extracted from the video file. A total of 1520 images are extracted from the image data set, and the data set is divided into training set, validation set and test set according to the ratio of 7:2:1. RESULTS: We verified it on the 76-segment clinical data test set and achieved the accuracy of 0.9400, with the AUC of 0.9880, and the F1 of 0.9388. In addition, because of the color space recognition method, the detection per frame can be completed within 29 microseconds and thus the detection efficiency has been improved significantly. CONCLUSION: With the efficiency and effectiveness of this algorithm, the lens scan video is used as the research object, which improves the accuracy of the screening. It is closer to the actual cataract diagnosis and treatment process, and can effectively improve the cataract inspection ability of non-ophthalmologists. For cataract screening in poor areas, the accessibility of ophthalmology medical care is also increased.

Development and Validation of a Hypoxia-related Prognostic Model for Ovarian Cancer.

BACKGROUND: The high heterogeneity of ovarian cancer (OC) brings great difficulties to its early diagnosis and prognostic forecast. There is an urgent need to establish a prognostic model of OC based on clinicopathological features and genomics. METHODS: We identified hypoxia-related differentially expressed genes (DEGs) between OC tissues from The Cancer Genome Atlas (TCGA) and normal tissues from the Genotype-Tissue Expression (GTEx). LASSO Cox regression analysis was applied for building a prognostic model in the TCGA-GTEx cohorts, and its predictive value was validated in the GEO-OC cohort. Functional enrichment analysis was performed to investigate the underlying mechanisms. By constructing a hypoxia model of the SKOV3 cell line and applying qRT-PCR, we investigated the relationship between hypoxia with two novel genes in the prognostic model (ISG20 and ANGPTL4). RESULTS: Twelve prognostic hypoxia-related DEGs were identified, and nine of them were selected to establish a prognostic model. OC patients were stratified into two risk groups, and the high-risk group showed reduced survival time compared to the low-risk group upon survival analysis. Univariate and multivariate Cox regression analysis demonstrated that the risk score was an independent risk factor for overall survival. The biological function of the identified prognostic hypoxia-related gene signature was involved in immune cell infiltration. Low expression of ISG20 was observed in the CoCl2-mimicked hypoxic SKOV3 cell line and negatively correlated with HIF-1α. CONCLUSION: Our findings showed that this hypoxia-related gene signature could serve as a satisfactory prognostic classifier for OC and will be beneficial to the research and development of targeted therapeutic strategies.

3 CpG Methylation Biomarkers for the Diagnosis of Polycystic Ovary Syndrome (PCOS) in Blood Samples.

BACKGROUND: Polycystic ovary syndrome (PCOS) is a common endocrine disease in women that seriously interferes with patient's metabolic and reproductive functions. The current diagnostic criteria for PCOS are expert-based and still disputed. Previous studies have identified changes in DNA methylation in peripheral blood of women with PCOS, but their diagnostic potential for PCOS remains to be studied. OBJECTIVE: The present study aimed to identify potential methylation biomarkers for the diagnosis of PCOS in blood. METHODS: Methylation profiling of peripheral blood was downloaded from a public database, Gene Expression Omnibus (GEO), including 30 PCOS patients (diagnosed with the revised 2003 Rotterdam consensus criteria) and 30 age-matched healthy women recruited from Centre of Reproductive Medicine, Linyi People's Hospital, Shandong, China. Weighted gene co-expression network analysis (WGCNA) was utilized to identify PCOS-related co-methylation CpG sites (co- MPs). Functional enrichment analysis was performed on the localized genes of PCOS-related co- MPs. The least absolute shrinkage and selection operator (LASSO) regression was used to screen out CpG methylation signatures for PCOS diagnosis, and receiver operating characteristic (ROC) analysis was conducted to evaluate their diagnostic accuracy. To assess the accuracy of the combination of the investigated indicators, multivariate ROC analysis was performed on the predicted probability values obtained using binary logistic regression on the methylation levels of selected CpGs. RESULTS: Seven co-methylation modules were obtained, among which the turquoise module is the most relevant to PCOS, containing 194 co-MPs. The genes that these co-MPs located in were mainly associated with the immune-related pathway. According to LASSO regression, three Co- MPs (cg23464743, cg06834912, cg00103771) were identified as potential diagnostic biomarkers of PCOS. ROC analysis showed an AUC (area under curve) of 0.7556 (sensitivity 60.0%, specificity 83.3%) for cg23464743, 0.7822 (sensitivity 70.0%, specificity 80.0%) for cg06834912, and 0.7611 (sensitivity 63.3%, specificity 83.3%) for cg00103771. The diagnostic accuracy of the combination of these 3 indicators presented to be higher than any single one of them, with the AUC of 0.8378 (sensitivity 73.3%, specificity 93.3%). CONCLUSION: The combination of 3 CpG methylation signatures in blood was identified with a good diagnostic accuracy for PCOS, which may bring new insight into the development of PCOS diagnostic markers in the future.