Figure 1

The BP-ANN model used in present study included an input layer with 5 values, which included h, height; m, body weight; y, age; Z, impedances; s, gender; a hidden layer with 5 neurons and an output layer with one neuron. The solid circles represent individual neurons and the lines represent the inputs, outputs and weighted connections between the neurons. n, net input; b ⁱ _j, bias vector; W ⁱ _j,k, weight matrix; f ⁱ, transfer functions; a ⁱ _j, neuron output. The superscript i represents the serial number of layers and the subscripts j and k represent the serial number of the neuron and the input.