RESUMEN
Objective:To construct an artificial intelligence (AI)-assisted system based on deep learning for corneal in vivo confocal microscopy (IVCM) image recognition and to evaluate its value in clinical applications. Methods:A diagnostic study was conducted.A total of 18 860 corneal images were collected from 331 subjects who underwent IVCM examination at Renmin Hospital of Wuhan University and Zhongnan Hospital of Wuhan University from May 2021 to September 2022.The collected images were used for model training and testing after being reviewed and classified by corneal experts.The model design included a low-quality image filtering model, a corneal image diagnosis model, and a 4-layer identification model for corneal epithelium, Bowman membrane, stroma, and endothelium, to initially determine normal and abnormal corneal images and corresponding corneal layers.A human-machine competition was conducted with another 360 database-independent IVCM images to compare the accuracy and time spent on image recognition by three senior ophthalmologists and the AI system.In addition, 8 trainees without IVCM training and with less than three years of clinical experience were selected to recognize the same 360 images without and with model assistance to analyze the effectiveness of model assistance.This study adhered to the Declaration of Helsinki.The study protocol was approved by the Ethics Committee of Renmin Hospital of Wuhan University (No.WDRY2021-K148).Results:The accuracy of this diagnostic model in screening high-quality images was 0.954.Its overall accuracy in identifying normal/abnormal corneal images was 0.916 and 0.896 in the internal and external test sets, respectively.Its accuracy reached 0.983, 0.925 in the internal test sets and 0.988, 0.929 in the external test sets in identifying corneal layers of normal and abnormal images, respectively.In the human-machine competition, the overall recognition accuracy of the model was 0.878, which was similar to the average accuracy of the three senior physicians and was approximately 300 times faster than the experts in recognition speed.Trainees assisted by the system achieved an accuracy of 0.816±0.043 in identifying corneal layers of normal and abnormal images, which was significantly higher than 0.669±0.061 without model assistance ( t=6.304, P<0.001). Conclusions:A deep learning-based assistant system for corneal IVCM image recognition is successfully constructed.This system can discriminate normal/abnormal corneal images and diagnose the corresponding corneal layer of the images, which can improve the efficiency of clinical diagnosis and assist doctors in training and learning.
RESUMEN
Objective:To investigate the inhibitory effect of miR-497 on the corneal epithelial healing in diabetic mice and its possible mechanism.Methods:Forty healthy clean-grade wild-type C57BL/J6 mice were randomly divided into a blank control group and a model control group, with 20 mice in each group.Another 20 CRISPR/Cas9-mediated miR-497 knockout mice and miR-497 overexpression mice were taken as miR-497 knockout and miR-497 overexpression groups, respectively.The diabetes model was constructed by continuous intraperitoneal injection of streptozotocin (STZ) to the mice in model control, miR-497 knockout and miR-497 overexpression groups, and the mice in blank control group were injected with an equal amount of citrate buffer, followed by 8-week normal feeding.After the establishment of diabetes model, the corneal epithelial injury model was further constructed by scraping off part of the corneal epithelium with a central diameter of 2 mm.The corneal epithelial defect area of mice in 0, 12, 24 and 36 hours after corneal epithelial injury was observed by corneal fluorescein sodium staining.The expression of Wnt3a and β-catenin proteins in mice corneal tissues was detected by Western blot.The expression of miR-497 as well as the mRNA expression levels of cell proliferation-associated factor genes CyclinD1, c-Myc, and Ki-67 mRNA was detected by real-time quantitative fluorescence PCR.The targeting relationship between miR-497 and wnt3a was detected by a dual luciferase reporter gene assay.Human corneal epithelial cells (HCEC) were cultured in vitro and transfected with miR-497 mimics, miR-497 mimics negative control, miR-497 inhibitor, and miR-497 inhibitor negative control by Lipo8000 as miR-497 mimics group, mimics negative control group, miR-497 inhibitor group, andmiR-497 inhibitor negative control group, respectively, all of which were cultured in high glucose medium containing 25% glucose.Another two groups of HCEC were taken and cultured in medium containing 5% and 25% glucose as control and high glucose groups, respectively.The cell proliferation viability was determined by CCK8 method.The use and care of animals complied ith the ARVO statement.The study protocol was approved by the Ethics Committee of Renmin Hospital of Wuhan University (2019K-K010). Results:Eight weeks after STZ injection, the blood glucose of mice was significantly higher and the weight was significantly lower in each diabetic model group than those of blank control group (all at P<0.05). At 12, 24 and 36 hours after the corneal epithelial injury, the percentages of corneal epithelial defect area observed by slit-lamp microscopy in model control group were significantly higher than those in blank control group and miR-497 knockout group and lower than those in miR-497 overexpression group, and the differences were statistically significant (all at P<0.05). The relative expressions of wnt3a and β-catenin proteins in the corneal tissues of model control group were significantly lower than those of blank control group and miR-497 knockout group, but higher than those of miR-497 overexpression group, and the differences were statistically significant (all at P<0.05). The relative expressions of CyclinD1, c-Myc and Ki-67 mRNA in model control group were lower than those in miR-497 knockout group, but higher than those in miR-497 overexpression group, and the differences were statistically significant (all at P<0.05). The relative expression of miR-497 in model control group, miR-497 knockout group and miR-497 overexpression group was 1.00±0.02, 0.63±0.06 and 1.48±0.03, respectively, with a statistically significant difference ( F=19.62, P<0.01). The luciferase activity of miR-497-5p mimics group in wild-type wnt3a transfected cells was lower than that of miR-497-5p negative control group and empty vector group, and the differences were statistically significant (all at P<0.05). In the mutant wnt3a transfected cells, there was no significant difference in the luciferase activity among various groups ( F=0.73, P=0.59). The cell proliferation A value of high glucose group was 0.59±0.03, which was significantly lower than 0.59±0.03 of normal control group and 0.88±0.08 of miR-497 inhibitor group, but significantly higher than 0.48±0.11 of miR-497 mimics group (all at P<0.05). Conclusions:The silencing of miR-497 may promote the repair of diabetic corneal epithelial defects by targeting wnt/β-catenin pathway.
RESUMEN
The " Internet plus" development approach for early phase clinical study management is of great significance to improve the implementation quality and management effectiveness of clinical trials. In December 2020, a tertiary hospital used the internet platform SaaS mode to build its early clinical study management system, with the design concepts of simplicity, convenience and adaptability. Based on cloud computing and multi-level data flow mode, the system formed a simple and feasible system architecture, real-time follow-up system process and dynamic visual project information through the adaptive design of mobile terminal application link and system user interface, with the advantages of low cost, high flexibility, strong specificity and multi-party interoperability.Since its launched in January 2021, as of May 2022, the system had included 56 early clinical trial projects of the hospital, effectively improving the implementation progress and quality of early clinical trials, strengthening the risk control in the trial, so as to provide reference for the digital development of hospital clinical research.
RESUMEN
Objective@#To explore the risk factors of deep venous thrombosis(DVT) in lower limbs of orthopedic inpatients, and provide reference for clinical nursing and prevention of thrombosis.@*Methods@#A retrospective case-control approach was used in this study. A total of 148 orthopedic inpatients diagnosed with DVT in the First Affiliated Hospital of Soochow University from February 2014 to October 2018 were selected as the case group. At the same time, 148 orthopaedic inpatients with undiagnosed DVT in the same Department were randomly selected as the control group. The electronic medical history data and related laboratory examination indexes of the two groups of patients during hospitalization in Orthopedics Department were analyzed retrospectively. Gender, age, seven indicators of blood coagulation, as well as history of hypertension, diabetes, history of DVT and other common risk factors of DVT in the two groups were statistically analyzed.@*Results@#In the case group, 63 patients (42.5%) DVT developed in the right lower extremity. The differences in the clinical symptoms and characteristics of fever, pulmonary infection, cough, expectoration, chest distress and shortness of breath, pleural effusion and constipation between the two groups were statistically significant (χ2 value was 5.688-12.312, P<0.01 or 0.05). Single factor analysis of related risk factors showed the differences of age, type of injury, lower limb joint replacement, a plaster cast, hypertension, history of DVT, central venous catheter, D-dimer and preoperative albumin levels, fibrinogen degradation products, and red blood cell count between the two groups were statistically significant (t value was-7.275-3.998, χ2 value was 4.889-13.305, Z value was-3.500--3.454, P < 0.01). Multivariate Logistic regression analysis showed that lower limb joint replacement(OR=0.383, 95%CI0.190-0.773), cough (OR=0.085, 95%CI0.010-0.731), chest distress and shortness of breath (OR=0.240, 95%CI0.077-0.745), constipation (OR=0.312, 95%CI0.135-0.718), red blood cell count (OR=3.314, 95%CI 2.105-5.216), hypertension (OR=0.534, 95%CI0.292-0.976) were independent risk factors for DVT formation in orthopedic inpatients (P<0.01 or 0.05).@*Conclusions@#DVT in orthopedic inpatients is a common result of multiple factors, among which, lower limb joint replacement, cough, chest distress and shortness of breath, constipation, red blood cell count and hypertension are independent risk factor, which should be monitored and given preventive care.
RESUMEN
Objective:To explore the risk factors of deep venous thrombosis(DVT) in lower limbs of orthopedic inpatients, and provide reference for clinical nursing and prevention of thrombosis.Methods:A retrospective case-control approach was used in this study. A total of 148 orthopedic inpatients diagnosed with DVT in the First Affiliated Hospital of Soochow University from February 2014 to October 2018 were selected as the case group. At the same time, 148 orthopaedic inpatients with undiagnosed DVT in the same Department were randomly selected as the control group. The electronic medical history data and related laboratory examination indexes of the two groups of patients during hospitalization in Orthopedics Department were analyzed retrospectively. Gender, age, seven indicators of blood coagulation, as well as history of hypertension, diabetes, history of DVT and other common risk factors of DVT in the two groups were statistically analyzed.Results:In the case group, 63 patients (42.5%) DVT developed in the right lower extremity. The differences in the clinical symptoms and characteristics of fever, pulmonary infection, cough, expectoration, chest distress and shortness of breath, pleural effusion and constipation between the two groups were statistically significant ( χ2 value was 5.688-12.312, P<0.01 or 0.05). Single factor analysis of related risk factors showed the differences of age, type of injury, lower limb joint replacement, a plaster cast, hypertension, history of DVT, central venous catheter, D-dimer and preoperative albumin levels, fibrinogen degradation products, and red blood cell count between the two groups were statistically significant ( t value was-7.275-3.998, χ 2 value was 4.889-13.305, Z value was-3.500--3.454, P < 0.01). Multivariate Logistic regression analysis showed that lower limb joint replacement( OR=0.383, 95% CI0.190-0.773), cough ( OR=0.085, 95% CI0.010-0.731), chest distress and shortness of breath ( OR=0.240, 95% CI0.077-0.745), constipation ( OR=0.312, 95% CI0.135-0.718), red blood cell count ( OR=3.314, 95% CI 2.105-5.216), hypertension ( OR=0.534, 95% CI0.292-0.976) were independent risk factors for DVT formation in orthopedic inpatients ( P<0.01 or 0.05). Conclusions:DVT in orthopedic inpatients is a common result of multiple factors, among which, lower limb joint replacement, cough, chest distress and shortness of breath, constipation, red blood cell count and hypertension are independent risk factor, which should be monitored and given preventive care.