Machine learning regression algorithms to predict short-term efficacy after anti-VEGF treatment in diabetic macular edema based on real-world data.

The objective of this retrospective study was to predict short-term efficacy of anti-vascular endothelial growth factor (VEGF) treatment in diabetic macular edema (DME) using machine learning regression models. Real-world data from 279 DME patients who received anti-VEGF treatment at Ineye Hospital of Chengdu University of TCM between April 2017 and November 2022 were analyzed. Eight machine learning regression models were established to predict four clinical efficacy indicators. The accuracy of the models was evaluated using mean absolute error (MAE), mean square error (MSE) and coefficient of determination score (R2). Multilayer perceptron had the highest R2 and lowest MAE among all models. Regression tree and lasso regression had similar R2, with lasso having lower MAE and MSE. Ridge regression, linear regression, support vector machines and polynomial regression had lower R2 and higher MAE. Support vector machine had the lowest MSE, while polynomial regression had the highest MSE. Stochastic gradient descent had the lowest R2 and high MAE and MSE. The results indicate that machine learning regression algorithms are valuable and effective in predicting short-term efficacy in DME patients through anti-VEGF treatment, and the lasso regression is the most effective ML algorithm for developing predictive regression models.

Deep learning for detection of age-related macular degeneration: A systematic review and meta-analysis of diagnostic test accuracy studies.

OBJECTIVE: To evaluate the diagnostic accuracy of deep learning algorithms to identify age-related macular degeneration and to explore factors impacting the results for future model training. METHODS: Diagnostic accuracy studies published in PubMed, EMBASE, the Cochrane Library, and ClinicalTrails.gov before 11 August 2022 which employed deep learning for age-related macular degeneration detection were identified and extracted by two independent researchers. Sensitivity analysis, subgroup, and meta-regression were performed by Review Manager 5.4.1, Meta-disc 1.4, and Stata 16.0. The risk of bias was assessed using QUADAS-2. The review was registered (PROSPERO CRD42022352753). RESULTS: The pooled sensitivity and specificity in this meta-analysis were 94% (P = 0, 95% CI 0.94-0.94, I2 = 99.7%) and 97% (P = 0, 95% CI 0.97-0.97, I2 = 99.6%), respectively. The pooled positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and the area under the curve value were 21.77(95% CI 15.49-30.59), 0.06 (95% CI 0.04-0.09), 342.41 (95% CI 210.31-557.49), and 0.9925, respectively. Meta-regression indicated that types of AMD (P = 0.1882, RDOR = 36.03) and layers of the network (P = 0.4878, RDOR = 0.74) contributed to the heterogeneity. CONCLUSIONS: Convolutional neural networks are mostly adopted deep learning algorithms in age-related macular degeneration detection. Convolutional neural networks, especially ResNets, are effective in detecting age-related macular degeneration with high diagnostic accuracy. Types of age-related macular degeneration and layers of the network are the two essential factors that impact the model training process. Proper layers of the network will make the model more reliable. More datasets established by new diagnostic methods will be used to train deep learning models in the future, which will benefit for fundus application screening, long-range medical treatment, and reducing the workload of physicians.

The effectiveness of AGV, Ex-PRESS, or trabeculectomy in the treatment of primary and secondary glaucoma: a systematic review and network meta-analysis.

BACKGROUND: In this study, we compared the efficacy of Ahmed, Ex-PRESS, and trabeculectomy to provide a reference for determining surgical schemes for glaucoma patients undergoing external drainage surgery in clinical practice. METHODS: We performed a literature search for studies on the treatment of primary and secondary glaucoma with three types of external drainage surgery (Ahmed, Ex-PRESS, and trabeculectomy). As at April 24, 2021, seven electronic databases were searched for randomized controlled trials comparing any two of Ahmed, Ex-PRESS, and trabeculectomy in the treatment of glaucoma. The Cochrane tool was also adopted to evaluate the risk of bias in these trials. The relative risk (RR) with 95% confidence interval (CI), and weighted mean difference (WMD) were determined and compared indirectly using R software. RESULTS: A total of 14 randomized controlled trials were included in this study, involving 866 eyes of 808 patients. As for the intraocular pressure (IOP) after 3 months, trabeculectomy did not contribute to better improvement than Ahmed (WMD =0.014; 95% CI: -0.14-0.18) and Ex-PRESS (WMD =0.014; 95% CI: -0.072-0.097). However, there was a significant difference in the IOP 1 year between trabeculectomy and Ex-PRESS (WMD =0.097; 95% CI: 0.0080-0.18), with the latter achieving a favorable improvement effect. Meanwhile, the complete success (CS) of trabeculectomy was significantly lower than that of Ex-PRESS (RR =0.73; 95% CI: 0.57-0.93). In addition, Ex-PRESS was superior to Ahmed (WMD =-0.48; 95% CI: -0.89 to -0.084) in terms of a decreased number of post-operative medications. DISCUSSION: For glaucoma patients who are required to receive external drainage surgery, Ex-PRESS could achieve a significant effect on the IOP 1 year and CS, as well as a marked decrease in the number of post-operative medications used, compared with the other two types of surgery. In terms of the efficacy at least 1 year after surgery, Ex-PRESS should be one of the preferred methods for external drainage.

Intense pulsed light for meibomian gland dysfunction: a systematic review and meta-analysis.

BACKGROUND: Intense pulsed light therapy (IPL) is a new method being used to treat meibomian gland dysfunction (MGD) globally. With an increasing number of studies being published, it is necessary to consider additional factors related to treatment. This review aims to investigate the efficacy and safety of IPL for the treatment of MGD. METHODS: The PubMed, EMBASE, Web of Science, Cochrane Library, Google Scholar, China National Knowledge Infrastructure (CNKI), Wanfang, VIP, and SinoMed databases were searched through February 24, 2020. Randomized clinical trials and cohort studies comparing IPL+ meibomian gland expression (MGX) or IPL alone with control groups were included. The weighted mean difference (WMD) was calculated to analyze the Ocular Surface Disease Index (OSDI) score and Standard Patient Evaluation of Eye Dryness (SPEED) score, and the standard mean difference (SMD) was calculated to analyze the tear breakup time (TBUT). Heterogeneity was quantified by the I2 statistic ranging from 0 to 100%, and a random effects model was used in this meta-analysis. All analyses were performed by RevMan 5.3. All p values were calculated by the t test, and p values were regarded as statistically significant at p < 0.05. The Cochrane Collaboration's tool for assessing risk of bias was used to identify and evaluate bias in the literature. RESULTS: Nine studies with a total of 539 patients were included. Eight studies examined TBUT, six examined OSDI scores, and four examined SPEED scores. IPL combined with MGX showed superiority regarding the TBUT (SMD 2.33, 95% CI 1.04-3.61), and OSDI scores (WMD 11.93, 95% CI - 17.10 to - 6.77), with high heterogeneity. The SPEED scores were not significantly different. CONCLUSIONS: IPL combined with MGX may be an effective and safe treatment for MGD, but it cannot improve all symptoms. IPL alone is not superior to MGX. The efficacy is also affected by the number and average frequency of treatments. The efficacy of IPL may decrease within 6 months after the last treatment, so it should be considered a long-term adjuvant therapy combined with MGX. When patients receive 3 or 4 treatments (once every 3-4 weeks), a return visit at 6 months after the last treatment is required.

XEN gel implant with or without phacoemulsification for glaucoma: a systematic review and meta-analysis.

BACKGROUND: XEN gel implant is an alternative method of treating glaucoma by connecting the anterior chamber and the subconjunctival space. The efficacy of the XEN gel implant and whether to combine with phacoemulsification is what most concerned. This review aims to test the efficacy and safety of the XEN gel implant compared with trabeculectomy and to test the efficacy between XEN combined with phacoemulsification and XEN alone. METHODS: The PubMed, EMBASE, the Cochrane Library, and ClinicalTrials.gov were searched through July 8, 2020. Studies comparing XEN and trabeculectomy or comparing XEN + phacoemulsification and XEN alone were included. The standard mean differences (SMD) were calculated to analyze the lowered intraocular pressure (IOP) and the number of anti-glaucoma medications. All data were measured from baseline to endpoint. The I2 statistic quantified heterogeneity ranged from 0% to 100%, and a random effects model was used in this meta-analysis. Review Manager 5.3 performed all analyses. The t-test calculated all P values, and P values were regarded as statistically significant at P<0.05. The methodological index for non-randomized studies was used to find and test bias in the literature. RESULTS: Twelve studies with 1,602 eyes were included. Five studies compared the XEN gel implant with trabeculectomy. Eight compared XEN + phacoemulsification with XEN alone. There was no significant correlation between XEN and trabeculectomy groups in lowering IOP (SMD 0.30, 95% CI, 0.00 to 0.60, I2=60%) and reduced the number of IOP lowering drugs (SMD 0.01, 95% CI, -0.16 to 0.17, I2=0%). There was a significant difference between XEN + phacoemulsification and XEN alone in lowering IOP (1,034 eyes, SMD 0.22, 95% CI, 0.05 to 0.40, I2=38%) and reducing the number of medications (729 eyes, SMD 0.20, 95% CI, -0.06 to 0.46, I2=62%). CONCLUSIONS: XEN gel implant, although not effective as trabeculectomy, is a safe operation for open-angle and some other types of glaucoma. Meanwhile, XEN alone is more effective than XEN combined with phacoemulsification within 1 week after operations. After three months, the two groups are similar in lowering IOP. More extensive, better-designed, strictly blinded, multicenter randomized clinical trials are needed to confirm our findings.