abstract |
The present invention provides a method for diagnosing, screening or prognosing human or animal diseases, and a method for determining the severity and cause of the disease. Collect patient data, including biological, physical, demographic, racial, environmental and medical history data, screen data for disease diagnosis, digitize data and test Analyze the judgment of the data, group individual data values, preprocess these data to create preprocessed values, and input the selected data to the neural network to input individual data input values Analyzing the contribution to the neural network, selecting the optimal training neural network trained to generate a diagnostic index based on performance, accuracy and cost, and other patient data into the trained neural network An input value is generated that indicates whether the patient is or is likely to suffer from the disease. [Selection] Figure 1 |