Key result
A back propagation artificial neural network model predicted diabetic retinopathy with a significantly higher area under the ROC curve compared to multivariable logistic regression (0.84 vs 0.77, P<0.001).
Why the study?
Monitoring and predicting diabetic retinopathy in patients with diabetes is necessary for early discovery and timely treatment.
Does a BP-ANN model improve the prediction of diabetic retinopathy compared to multivariable logistic regression in patients with type 2 diabetes?
Cross-Sectional (n=530)
Does a BP-ANN model improve the prediction of diabetic retinopathy compared to multivariable logistic regression in patients with type 2 diabetes?
Absolute Event Rate: 0.84% vs 0.77%
p-value: p=<0.001
A back propagation artificial neural network model using four clinical parameters significantly improved the prediction of diabetic retinopathy compared to traditional logistic regression.
BP-ANN outperforms MLR for diabetic retinopathy prediction in type 2 diabetes; hypothesis-generating and requires prospective validation before clinical adoption.
Background: Monitoring and prediction of diabetic retinopathy (DR) is necessary in patients with diabetes for early discovery and timely treatment of disease. We aimed to analyze the association between DR and biochemical and metabolic parameters, and develop a predictive model for DR. Methods: A total of 530 Chinese residents including 423 with type 2 diabetes (T2D) aged 18 years or older participated in this study. The association between DR and biochemical and metabolic parameters was analyzed by the univariate and multivariable logistic regression (MLR). According to the MLR results, we developed a back propagation artificial neural network (BP-ANN) model by selecting tan-sigmoid as the transfer function of the hidden layers nodes, and pure-line of the output layer nodes, with training goal of 0.5×10 −5 . Results: There were 51 (9.6%) diabetic participants with DR. After univariate and MLR analysis, duration of diabetes, waist to hip ratio, HbA 1 c and family history of diabetes were independently associated with the presence of DR (all P < 0.05). Based on these parameters, the area under the receiver operating characteristic (ROC) curve for the BP-ANN model was significantly higher than that by MLR (0.84 vs. 0.77, P < 0.001). Conclusion: Our evaluation demonstrated the potential role of BP-ANN model to identify DR in screening practice. The presence of DR was well predictable using the proposed BP-ANN model based on four related parameters (duration of diabetes, waist to hip ratio, HbA 1 c and family history of diabetes).
No takes yet. Share an insight, caveat, or question.
Yao et al. (2019) conducted a cross-sectional in Diabetic retinopathy (n=530). Back propagation artificial neural network (BP-ANN) model vs. Multivariable logistic regression (MLR) was evaluated on Area under the receiver operating characteristic (ROC) curve for predicting diabetic retinopathy (p=<0.001). A back propagation artificial neural network model predicted diabetic retinopathy with a significantly higher area under the ROC curve compared to multivariable logistic regression (0.84 vs 0.77, P<0.001).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: