Method and system for automatically generating diagnosis result and computer equipment

Abstract

The invention belongs to the technical field of computers, and relates to a method and system for automatically generating a diagnosis result and computer equipment. The method comprises the steps: collecting medical history notes of a case, and carrying out the preprocessing of the medical history notes; constructing a pre-training word vector model Word2Vec by using the preprocessed corpus to obtain a corpus represented by a vector; constructing a neural network structure, and inputting the expectation of the vector representation into the neural network, wherein the neural network structurecomprises a bidirectional gating cycle unit BiGRU, a convolutional neural network (CNN) and an attention mechanism Att; setting a threshold value to select a label in the multi-classification probability matrix output by the neural network, wherein the selected label is a diagnosis result. Compared with a traditional bag-of-words model, the method has the advantages that more features can be extracted and are more effective; finally, a threshold value is set to process the multi-label problem, and correlation constraints are added, so that the conflict of diagnosis results is solved, and theresult generation is more reasonable.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to View More

AI Technical Summary

Benefits of technology

The technical effect of this patented method overcomes previous methods for making diagnoses by analyzing data from multiple sources instead of just one source like bags or words alone. This makes it easier than current models because they have many extra dimensions - such as attributes (such as age) and symptoms associated with diseases. Additionally, there may be correlations between different variables within each dataset which helps explain how these relationships affects their outcomes. Overall, this new approach provides better accuracy compared to existing techniques.

Problems solved by technology

This patented describes how experts use techniques like interventional therapy (IT) and intravenous injection(IVI). They study different ways to improve their ability to reproduce children without causing harm during childbirth. One way they work involves studying them under controlled conditions before giving birth. By doing these tests, practitioners may identify issues early enough to make informed decisions about which treatments should next take place.

Images