Quick intelligent diagnosis method for infant pneumonia on the basis of hybrid deep learning model

Abstract

The invention disclose a quick intelligent diagnosis method for the infant pneumonia on the basis of a hybrid deep learning model, and solves the problem that the body of an infant is injured since existing infant pneumonia diagnosis time is overlong or a misdiagnosis happens. The quick intelligent diagnosis method comprises the following steps of: S1: measuring and collecting each piece of physiological index data of the infant, and according to whether the infant suffers from the pneumonia or not, marking the data; S2: carrying out data cleaning and abnormal data rejection on non-breathing audio data, and constructing a dataset used for training a pneumonia diagnosis model; S3: constructing a dataset used for training a rale identification model; S4: carrying out long short-term memory (LSTM) training; and S5: carrying out deep neural network (DNN) training for interpreting each piece of physiological index data of a patient so as to judge whether the patient suffers from the pneumonia or not. According to the quick intelligent diagnosis method disclosed by the invention, the diagnosis rate and the accuracy of the infant pneumonia can be effectively improved, and serious injuriescaused for the body of the infant due to overlong diagnosis waiting time or the misdiagnosis can be avoided.

Description

technical field [0001] The invention belongs to the field of intelligent medical treatment, and relates to a rapid and intelligent diagnosis method for infant pneumonia based on a hybrid deep learning model. Background technique [0002] Pneumonia refers to the inflammation of the terminal airways, alveoli and pulmonary interstitium, which can be caused by disease microorganisms, physical and chemical factors, immune damage, allergies and drugs. Bacterial pneumonia is the most common pneumonia and one of the most common infectious diseases. Pneumonia in daily life mainly refers to pneumonia caused by bacterial infection. Before the application of antibiotics, bacterial pneumonia was a great threat to human health. The emergence and development of antibiotics have significantly reduced the mortality rate of pneumonia. In daily life, pneumonia is a common disease in infants and young children. Pneumonia ranks first among hospitalized children, and it is also one of the disea...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problem that infants and young children are prone to pneumonia, and it will cause great damage to the body and it is difficult to quickly diagnose. In addition, infants and young children cannot express their illness in their own words, and pneumonia is more harmful to the body and prone to misdiagnosis. Therefore, it is more necessary rapid diagnosis

Images