Research on a Pre-diagnostic System for Lung Diseases Based on Multi-classification Model of Support Vector Machine

China has entered an aging society, and the elderly are a high-risk group for death from lung diseases. After the epidemic, although many pneumonia patients have recovered, lung damage still exists, and there is a greater demand for lung examination. In response to this demand, an intelligent home lung disease pre-diagnosis system is designed to predict a variety of lung diseases. The system uses the lung breath sound feature analysis technology to generate a training model through noise reduction processing and feature extraction and compares it with the actual collected breath sound to diagnose lung health. The system uses the double-threshold Empirical Mode Decomposition denoising algorithm and the feature extraction algorithm based on Hilbert Huang Transform to construct a support vector machine multi-classification model to realize the prediction of lung diseases. Such a model facilitates the early detection and timely treatment of lung diseases and provides patients with convenient home diagnostic tools.