Least Squares Support Vector Regression for Spirometric Forced Expiratory Volume (FEV1) Values

Spirometry test is an inevitable, essential screening test in the case of respiratory and lung related diseases. This work focuses on predicting FEV1, which is the most significant and one of the deciding value in making the conclusion on respiratory related disorders by Least Squares Support Vector Machine (LS SVM) regression. This prediction of FEV1 values will enhance the spirometric method, when the data is incomplete or poorly recorded, accuracy of diagnosis of the abnormalities can be improved using SVM based methods. In this paper, an attempt is made to predict FEV1 values by LS SVM. The results show that the predication accuracy is very high.