首页>
外国专利>
METHOD FOR VISUALIZING CLASSIFICATION RESULT OF DEEP NEURAL NETWORK FOR PREDICTION OF DISEASE PROGNOSIS THROUGH TIME SERIES MEDICAL DATA AND SYSTEM THEREOF
METHOD FOR VISUALIZING CLASSIFICATION RESULT OF DEEP NEURAL NETWORK FOR PREDICTION OF DISEASE PROGNOSIS THROUGH TIME SERIES MEDICAL DATA AND SYSTEM THEREOF
展开▼
机译:基于时间序列医学数据的深层神经网络预测结果的可视化预测方法及系统
展开▼
页面导航
摘要
著录项
相似文献
摘要
Disclosed are a method for visualizing a classification result of a deep neural network for prediction of disease prognosis through time series medical data and a system thereof. The method which is implemented with a computer includes the steps of: generating sequence data by expressing medical data of a patient in a sequence form that is time series data in chronological order; predicting the patients disease from the sequence data through a disease prediction model by using a recurrent neural network (RNN) and an attention network; and visualizing an attention weight of the sequence data for the patients disease prediction by using a gradient-weighted class activation map (Grad-CAM).;COPYRIGHT KIPO 2020
展开▼