METHOD FOR PREDICTING MEDICAL EVENT FROM ELECTRONIC MEDICAL RECORD BY USING PRE-TRAINED ARTIFICIAL NEURAL NETWORK, AND APPARATUS THEREFOR

摘要

Disclosed is a method by which a computing device predicts a medical event on the basis of a pre-trained artificial neural network according to various embodiments. Disclosed are a method for predicting a medical event on the basis of a pre-trained artificial neural network, and an apparatus therefor, the method comprising the steps of: receiving an electronic medical record vector comprising a plurality of vital sign components; and outputting the medical event corresponding to the electronic medical record vector by using the artificial neural network, wherein the artificial neural network is pre-trained on the basis of learning data, and the learning data includes augmented electronic medical record vectors reconstructed using original electronic medical record vectors acquired in advance of a first time point on the basis of a mask vector that loses at least one vital sign component among the plurality of vital sign components at the first time point.