Respiratory Rate Estimation Combining Autocorrelation Function-Based Power Spectral Feature Extraction with Gradient Boosting Algorithm

Various machine learning models have been used in the biomedical engineering field, but only a small number of studies conducted on respiratory rate estimation. Unlike ensemble using simple averages basic learners such as bagging, random forest, and boosting, gradient boosting algorithm is based effective iteration strategies. This just beginning to be for Based this, we propose novel methodology combining an autocorrelation function-based power spectral feature extraction process with estimate since acquire respiration frequency that finds time domain’s periodicity. The proposed solves overfitting training datasets because obtain data dimension by applying then split long-resampled wave signal increase input samples. model provides accurate estimates offers solution reliably managing estimation uncertainty. In addition, method presents more precise than conventional measurement techniques.