Prediction Of Complex Acute Appendicitis Based On PCA-HGS-MSVM

In the process of digital transformation in the medical field,the machine semester algorithm has been fully applied and has achieved excellent results,so the deeper integration of machine learning algorithms into the medical field has gradually begun to become the focus of research.In the medical field,acute appendicitis is one of the most common abdominal emergencies,and complicated acute appendicitis is difficult to treat and has a high mortality rate compared with simple acute appendicitis.Therefore,it is necessary to use machine learning to construct a prediction model for complex acute appendicitis and strengthen the screening and prediction of complex acute appendicitis.Existing algorithms have limitations in predicting complex acute appendicitis,and traditional machine learning algorithms have fast training time but low prediction accuracy.Deep learning models have complex structures,require a large amount of data for training,and require high computing resources and time overhead,making them difficult to deploy in real-world applications.The scoring system has a certain subjectivity and cannot make objective predictions.Single marker prediction performance is low.In view of the above problems,this thesis applies Mercer support vector machine to complex acute appendicitis prediction,which has better performance than traditional support vector machine,so a complex acute appendicitis prediction model with the optimization goal of improving the prediction accuracy of MSVM is proposed.The optimization of the prediction model for complex acute appendicitis is divided into two parts: optimizing feature processing and optimizing MSVM internal parameters by using swarm intelligence algorithm.In the first part,in order to solve the problem of a large number of features,this applies reduces the dimensionality of the data through the PCA algorithm.In the second part,in order to solve the problem of low prediction accuracy of MSVM,the HGS swarm algorithm is used to optimize the internal penalty factor and weight factor of MSVM,and comparative experiments are carried out to verify their effectiveness.In the third part,in order to verify the extensiveness of this research model,four different datasets are selected for verification.The experimental results show that on the dataset of 358 cases of acute appendicitis in the First Hospital of Jilin University in Changchun City in the past five years,the PCA-HGS-MSVM prediction model proposed in this applies shows excellent performance on four parity indicators.