Font Size: a A A

Research On The Judgment Of The Depth Of Anesthesia Of Patients Based On Integrated Classification Technology

Posted on:2019-01-11Degree:MasterType:Thesis
Country:ChinaCandidate:X Q CaoFull Text:PDF
GTID:2432330566473322Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
In nowadays surgery,an accurate and non-invasive monitoring of depth of anesthesia system is indispensable.Not only the DOA system can help anesthesiologist more accurately to give the anesthetic,but also help the patients at the suitable state for surgery.In this thesis,we use multiple vital signs,like heart rate,mean blood pressure,electromyography(EMG)as the input,and use the Bispectral index(BIS)as the target,through three different methods(artifical neural networks,adaptive neuro-fuzzy inference system,support vector machine)to train and build models.As the result,ANFIS perform best,the MAE is 9.2848±3.2653 and Correlation coefficient is 0.767±0.0874.Then,use ensemble technique to depose the outputs from the models to make the model more universality and accuracy.After that,compared the final results with BIS and single model,we can find the ensemble technique of ANFIS & SVM perform better than any single model,the MAE is 8.9419±3.3584 and Correlation coefficient is 0.797±0.0861,and prove the possibility of substituting ensemble model for BIS.We hope this method can be used in clinical field in the future.
Keywords/Search Tags:Multiple vital signs, artifical neural networks (ANN), adaptive neuro-fuzzy inference system (ANFIS), support vector machine (SVM), ensemble technique
PDF Full Text Request
Related items