Additional file 1 of Logistic regression has similar performance to optimised machine learning algorithms in a clinical setting: application to the discrimination between type 1 and type 2 diabetes in young adults

Additional file 1: Figure S1. Flow diagram of participants through the model development stages. T1D: type 1 diabetes, T2D: type 2 diabetes. Figure S2. ROC AUC plots obtained using external validation dataset for seven prediction models. Legend: Solid lines: black = Support Vector Machine, dark grey = Logistic Regression, light grey = Random Forest. Dotted lines: black = Neural Network, dark grey = K-Nearest Neighbours, light grey = Gradient Boosting Machine. Figure S3. Correlation coefficient matrix and scatter plot of model predictions obtained from external test validation data.

附加文件1：图S1。受试者历经模型开发各阶段的流程图。T1D：1型糖尿病（type 1 diabetes），T2D：2型糖尿病（type 2 diabetes）。图S2：基于7种预测模型的外部验证数据集生成的受试者工作特征曲线下面积（ROC AUC）绘图。图例：实线：黑色=支持向量机（Support Vector Machine），深灰色=逻辑回归（Logistic Regression），浅灰色=随机森林（Random Forest）；虚线：黑色=神经网络（Neural Network），深灰色=K近邻（K-Nearest Neighbours），浅灰色=梯度提升机（Gradient Boosting Machine）。图S3：基于外部测试验证数据得到的模型预测结果相关系数矩阵与散点图。