Evaluation of respiratory properties by means of fractional order models

The goal of this paper is to model and analyze the properties of the respiratory system by means of fractional calculus. A linear fractional order system of commensurate order is obtained using the real and the imaginary parts of the measured respiratory impedance through an identification technique. In this context, the features used for the classification of some respiratory diseases are the identified parameters of the linear fractional order system of commensurate order. These features are then classified using the K-Nearest Neighbors (KNN) classifier. The proposed method has achieved an accuracy of 40% using only the first feature, however by using all the features the accuracy has increased up to 100%. The proposed classification technique is validated on 15 patients: healthy, asthma and chronic obstructive pulmonary disease (COPD).