Training Deep Network Designer with feature vectors

Hello all, I have just begun using ANN and I feel so confused using the Deep Network Designer toolbox right now. I have a problem with a small dataset size of 20 images. I have 10 normal and 10 abnormal images. I have extracted two features from normal and abnormal images. Lets call the features F1 and F2. So now, F1 feature has 10 normal image feature values and 10 abnormal image feature values. Same for feature F2 it also has 10 normal and 10 abnormal values. F1normal, F1abnormal, F2normal, F2abnormal are all 1D features with 10 rows. Now I want to design a Neural Network which will consider both F1normal and F2normal feature vectors to classify as normal. Similarly for abnormal. The problem is I have 4 input vectors and i need 2 output vectors and I dont know how to design a network. I was thinking of using a Back Propagation Neural Network but how do I acheive this? I have attacthed the 4 input vectors. Please help me with a detailed answer or code.