An intelligent diagnosis system and device for new coronary pneumonia based on lung sounds

Abstract

The invention discloses an intelligent diagnosis system and device for new coronary pneumonia based on lung sounds, belonging to the technical field of biomedical signal processing. This invention proposes the intelligent diagnosis of new coronary pneumonia based on lung sound data for the first time. Since the lung sound data is passed through a stethoscope-type acoustic sensor, it is non-radioactive, low in cost, non-invasive, easy to operate, and highly comfortable. It is suitable for primary hospitals and community health Center; the present invention uses a two-level neural network for intelligent diagnosis of new coronary pneumonia, the first-level neural network is used to classify normal and abnormal, and the second-level neural network is used to start only when the first-level classification result is abnormal, and classify the abnormal as Different clinical classification results, because the vast majority of people are non-COVID-19 patients, that is, there are far more non-COVID-19 patients than COVID-19 patients, so the second-level neural network is only turned on in fewer cases. Compared with a single-stage neural network, the invention has higher group detection efficiency for one multi-classification.

Description

technical field [0001] The invention belongs to the technical field of biomedical signal processing, and more specifically relates to an intelligent diagnosis system and device for new coronary pneumonia based on lung sounds. Background technique [0002] "Clinical classification" refers to the classification of a disease into several types based on the specific conditions of clinical onset characteristics or pathological changes, which is important information for treatment and mortality. At present, the Chinese Center for Disease Control and Prevention classifies the clinical types of new coronary pneumonia into "light, common, severe, and critical" to strengthen clinical classification guidance, better identify cases, and improve the level of treatment. [0003] The existing novel coronavirus clinical classification technology mainly relies on the detection of patients' clinical symptoms, lung CT images and blood inflammatory markers. For example, the "Intelligent Auxili...

AI Technical Summary

Problems solved by technology

[0004] However, these methods are radioactive, expensive, invasive, and the group detection efficiency is not high enough, and the operation is complicated, the comfort is low, and the applicable scenarios are limited.

Based on the current trend of increasing number of COVID-19 infections and the limitations of current clinical classification methods, a novel coronavirus with high group detection efficiency and high energy efficiency has been developed, which is suitable for energy-constrained application scenarios such as wearables. Pneumonia clinical classification system is necessary

Images