Table 1 Related work.
From: An artificial neural network approach for predicting hypertension using NHANES data
Author | Input features | n Total | Type of model | AUC (%) |
|---|---|---|---|---|
Artifical neural network models comparison | ||||
LaFreniere et al.21 | Age, gender, BMI, sys/diast BP, high and low density lipoproteins, triglycerides, cholesterol, microalbumin, and urine albumin creatinine ratio | 379,027 | Backpropagation neural network | 82 |
Polak and Mendyk22 | Age, sex, diet, smoking and drinking habits, physical activity level and BMI | 159,989 | backpropagation (BP) and fuzzy network | 75 |
Tang et al.23 | Sys/diast BP, fasting plasma glucose, age, BMI, heart rate, gender, WC, diabetes, renal profile | 2,092 | Feed-forward, back-propagation neural network | 76 |
Ture et al.24 | Age, sex, hypertension, smoking, lipoprotein, triglyceride, uric acid, total cholesterol, BMI | 694 | Feed-forwardneural network | 81 |
Lynn et al.25 | Sixteen genes, age, BMI, fasting blood sugar, hypertension medication, no history of cancer, kidney, liver or lung | 22,184 genes, 159 cases | One-hidden-layer neural network | 96.72 |
Sakr et al.6 | Age, gender, race, reason for test, stress, medical history | 23,095 | Backpropagation neural network | 64 |
López-Martínez et al.12 | Age, gender, ethnicity, BMI, smoking history, kidney disease, diabetes | 24,434 | Three-hidden layer neural network | 77 |
