M6A reader YTH structural domain family 2 (YTHDF2) has been recognized to play an oncogenic role in numerous tumors, but its role in cervical cancer has not been extensively discussed yet. This paper was designed to explore the role of YTHDF2 in cervical cancer and identify its underlying mechanism. The expression of YTHDF2 was first determined in cervical cancer cells by quantitative reverse-transcription polymerase chain reaction and western blot. Then, the migration, invasion, and epithelial-mesenchymal transition (EMT) process were observed in YTHDF2knockdown Hela cells using wound healing, transwell and immunofluorescence assays. The cisplatin chemosensitivity of Hela cells was also investigated by assessing cell activity with cell counting kit-8 and TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling). After MeRIP-Seq assay and actinomycin D treatment to confirm the binding relationship between YTHDF2 and AXIN1, the migration, invasion, EMT process, and cisplatin chemosensitivity were assessed again in Hela cells silenced by YTHDF2 and AXIN1 or treated with Wnt agonist. YTHDF2 was increased in cervical cancer cells, and depletion of YTHDF2 led to reduced migration, invasion and EMT process but enhanced chemosensitivity of cisplatin in Hela cells. Furthermore, YTHDF2 could bind to and stabilize the expression of AXIN1. When the YTHDF2-knockdown Hela cells were further transfected with AXIN1 knockdown or treated with Wnt agonist, the effects of YTHDF2 knockdown on the migration, invasion and EMT process were partially abolished, together with reduced cisplatin chemosensitivity. To sum up, we reported that YTHDF2 interference could suppress the EMT of cervical cancer cells and enhance cisplatin chemosensitivity by regulating AXIN1.

Background: To explore the application of neural network models in artificial intelligence (AI)-aided devices fitting for low vision patients.Methods: The data of 836 visually impaired people were collected in southwestern Fujian from May 2014 to May 2017. After a full eye examination, 629 low vision patients were selected from this group. Based on the visual functions, rehabilitation needs, and living quality scores of the selected patients, the professionals chose assistive devices that were the best fit for the patients. The data of these three factors were then subjected to the quantitative analysis, and the results were digitized and labeled. The final datasets were used to train a fully connected deep neural networks to obtain an AI-aided model for assistive device fitting.Results: In this study, the main causes of low vision in southwestern Fujian were congenital diseases, among which congenital cataract was the most common. During the low vision AI-aided devices fitting, we found that the intermediate distance magnifier was suitable for the largest number of patients. Through quantitative analysis of the research results, it was found that AI-aided devices fitting was closely related to visual function, rehabilitation needs and quality of life. If this complex relationship can be mapped into the neural network model, AI-aided device fitting can be realized. We built a fully connected neural network model for AI-aided device fitting. The input of the model was the characteristic data of low vision patients, and the output was the forecast of suitable devices. When the threshold of the model was 0.4, the accuracy was about 80% and the F1 value was about 0.31. This threshold can be used as the classification judgment threshold of the model. Conclusions: Low vision AI-aided device fitting is closely related to visual function, rehabilitation needs, and quality of life scores. The neural network model based on full connection can achieve high accuracy in AI-aided devices fitting. It has a great impact on clinical application.

ObjectiveTo systematically evaluate the safety and adverse event profiles of immune checkpoint inhibitors (ICIs) in patients with esophageal cancer (EPC) or gastroesophageal junction cancer (GEJC).MethodsPubMed, Web of Science, Cochrane Library, and major conference proceedings were systematically searched for all phase II or phase III randomized controlled trials (RCTs) in EPC or GEJC using ICIs. Safety outcomes including treatment-related adverse events (trAEs), immune-related adverse events (irAEs), and serious trAEs were evaluated by network meta-analysis or dichotomous meta-analysis based on the random-effects model.ResultsEleven RCTs involving EPC (five RCTs) and GEJC (six RCTs) were included in the final meta-analysis. NMA showed that placebo was associated with the best safety ranking for grade 3–5 trAEs (SUCRA = 96.0%), followed by avelumab (78.6%), nivolumab (73.9%), ipilimumab (57.0%), and pembrolizumab (56.6%). Conventional pairwise meta-analysis (CPM) showed that ICIs have similar grade 3–5 trAE risk compared with chemotherapy (RR = 0.764, 95% CI: 0.574 to 1.016, I2 = 95.7%, Z = 1.85, P = 0.065). NMA showed that the general safety of grade 3–5 irAEs ranked from high to low is as follows: ChT (85.1%), placebo (76.5%), ipilimumab (56.0%), nivolumab (48.5%), avelumab (48.4%), camrelizumab (41.8%), pembrolizumab (36.4%), and nivolumab + ipilimumab (21.6%). CPM showed that the rates of grade 3–5 irAEs in the ICI group and the chemotherapy group were 7.35% (154/2,095, 95% CI: [6.23%, 8.47%]) versus 2.25% (42/1,869, 95% CI: [1.58%, 2.92%]), with statistical significance (RR = 3.151, 95% CI = 2.175 to 4.563, Z = 6.07, P = 0.000). The most common irAEs in the ICI group were skin reaction (15.76%, 95% CI: [13.67%, 17.84%]), followed by hypothyroidism (9.73%, 95% CI: [8.07%, 11.39%]), infusion-related reactions (5.93%, 95% CI: [4.29%, 7.58%]), hepatitis (5.25%, 95% CI: [4.28%, 6.22%]), and pneumonitis (4.45%, 95% CI: [3.5%, 5.4%]).ConclusionDifferent ICIs had different toxicity manifestations and should not be considered as an entity. Compared with chemotherapy, ICIs were more prone to irAEs, but the overall rates remained low and acceptable. For clinicians, it is important to recognize and monitor the adverse events caused by ICIs for patients with EPC or GEJC.

Background: This study aimed to simulate the visual field (VF) effects of patients with VF defects using deep learning and computer vision technology.Methods: We collected 3,660 Humphrey visual fields (HVFs) as data samples, including 3,263 reliable 24-2 HVFs. The convolutional neural network (CNN) analyzed and converted the grayscale map of reliable samples into structured data. The artificial intelligence (AI) simulations were developed using computer vision technology. In statistical analyses, the pilot study determined 687 reliable samples to conduct clinical trials, and the two independent sample t-tests were used to calculate the difference of the cumulative gray values. Three volunteers evaluated the matching degree of shape and position between the grayscale map and the AI simulation, which was graded from 0 to100 scores. Based on the average ranking, the proportion of good and excellent grades was determined, and thus the reliability of the AI simulations was assessed.Results: The reliable samples in the experimental data consisted of 1,334 normal samples and 1,929 abnormal samples. Based on the existing mature CNN model, the fully connected layer was integrated to analyze the VF damage parameters of the input images, and the prediction accuracy of the damage type of the VF defects was up to 89%. By mapping the area and damage information in the VF damage parameter quintuple data set into the real scene image and adjusting the darkening effect according to the damage parameter, the visual effects in patients were simulated in the real scene image. In the clinical validation, there was no statistically significant difference in the cumulative gray value (P>0.05). The good and excellent proportion of the average scores reached 96.0%, thus confirming the accuracy of the AI model.Conclusions: An AI model with high accuracy was established to simulate the visual effects in patients with VF defects.

Background and purpose: Radioresistance remains a major reason of radiotherapeutic failure in esophageal squamous cell carcinoma (ESCC). Our study is to screen the immune-related long non-coding RNA (ir-lncRNAs) of radiation-resistant ESCC (rr-ESCC) via Gene Expression Omnibus (GEO) database and to construct a prognostic risk model.Methods: Microarray data (GSE45670) related to radioresistance of ESCC was downloaded from GEO. Based on pathologic responses after chemoradiotherapy, patients were divided into a non-responder (17 samples) and responder group (11 samples), and the difference in expression profiles of ir-lncRNAs were compared therein. Ir-lncRNA pairs were constructed for the differentially expressed lncRNAs as prognostic variables, and the microarray dataset (GSE53625) was downloaded from GEO to verify the effect of ir-lncRNA pairs on the long-term survival of ESCC. After modelling, patients are divided into high- and low-risk groups according to prognostic risk scores, and the outcomes were compared within groups based on the COX proportional hazards model. The different expression of ir-lncRNAs were validated using ECA 109 and ECA 109R cell lines via RT-qPCR.Results: 26 ir-lncRNA genes were screened in the GSE45670 dataset with differential expression, and 180 ir-lncRNA pairs were constructed. After matching with ir-lncRNA pairs constructed by GSE53625, six ir-lncRNA pairs had a significant impact on the prognosis of ESCC from univariate analysis model, of which three ir-lncRNA pairs were significantly associated with prognosis in multivariate COX analysis. These three lncRNA pairs were used as prognostic indicators to construct a prognostic risk model, and the predicted risk scores were calculated. With a median value of 2.371, the patients were divided into two groups. The overall survival (OS) in the high-risk group was significantly worse than that in the low-risk group (p < 0.001). The 1-, 2-, and 3-year prediction performance of this risk-model was 0.666, 0.702, and 0.686, respectively. In the validation setting, three ir-lncRNAs were significantly up-regulated, while two ir-lncRNAs were obviouly down-regulated in the responder group.Conclusion: Ir-lncRNAs may be involved in the biological regulation of radioresistance in patients with ESCC; and the prognostic risk-model, established by three ir-lncRNAs pairs has important clinical value in predicting the prognosis of patients with rr-ESCC.