Ensemble Learning-Based Pulse Signal Recognition: Classification Model Development Study

Abstract

Background In pulse signal analysis and identification, time domain and time frequency domain analysis methods can obtain interpretable structured data and build classification models using traditional machine learning methods. Unstructured data, such as pulse signals, contain rich information about the state of the cardiovascular system, and local features of unstructured data can be extracted and classified using deep learning. Objective The objective of this paper was to comprehensively use machine learning and deep learning classification methods to fully exploit the information about pulse signals. Methods Structured data were obtained by using time domain and time frequency domain analysis methods. A classification model was built using a support vector machine (SVM), a deep convolutional neural network (DCNN) kernel was used to extract local features of the unstructured data, and the stacking method was used to fuse the above classification results for decision making. Results The highest average accuracy of 0.7914 was obtained using only a single classifier, while the average accuracy obtained using the ensemble learning approach was 0.8330. Conclusions Ensemble learning can effectively use information from structured and unstructured data to improve classification accuracy through decision-level fusion. This study provides a new idea and method for pulse signal classification, which is of practical value for pulse diagnosis objectification.

An innovative approach for assessing coronary artery lesions: Fusion of wrist pulse and photoplethysmography using a multi-sensor pulse diagnostic device

“…The amplitudes and phases of these identified peaks and troughs provide essential information about physiological and pathological conditions [ 18 ]. By precisely these seven features points in Fig.…”

OEDL: an optimized ensemble deep learning method for the prediction of acute ischemic stroke prognoses using union features

“…Second, after considering bagging, boosting, stacking, and other methods (see the section Results), we chose the stacking algorithm as the ensemble method in this study, and the model constructed by it is named as the ensemble of deep learning (EDL). Stacking is a method that combines the outputs of multiple base learners according to a certain combination strategy ( 66 , 67 ). We chose classical and representative deep learning models such as DNN, LSTM-RNN, and DBN as the base learner ( 68 – 70 ) and random forest (RF) as the meta-learner.…”

New directions of digitally driven S&T evaluation

“…In the medical field, pulse signals contain rich information about the state of the cardiovascular system, which is also unstructured data. In the study of pulse type prediction, a DCNN kernel can extract the unstructured data's local features (Yan et al, 2021). Compared with the time domain and frequency domain features alone, the average accuracy of the DCNN-and SVM-based stacking network (DSSN) pulse signal ensemble classification model, which incorporates the unstructured data of pulse signals, is significantly improved.…”