Method of establishing blood pressure condition prediction model, method of predicting blood pressure condition and non-transitory computer readable medium

A method of establishing blood pressure condition prediction model includes performing, by a processor: obtaining a training set and a validation set, wherein the training set and the validation set are associated with a plurality of variables, using the training set to generate an initial model, sorting the variables with a combined set and the trained model to obtain an importance ranking list, generating a plurality of candidate prediction models respectively corresponding to different combinations of the variables with the importance ranking list and the training set, calculating a plurality of performance indications respectively corresponding to the candidate prediction models according to the validation set, and according to the candidate prediction model with the highest performance indication among the plurality of candidate prediction models, generating and storing a blood pressure condition prediction model, wherein the blood pressure condition prediction model is configured to provide a predicted