TY - JOUR
T1 - Predicting Ventricular Defibrillation Results Using Learning Models
T2 - A Design Practice and Performance Analysis
AU - Ling, Dean Chang
AU - Tsai, Min Shan
AU - Ling, Dean An
AU - Tsai, Shang Ho
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2021
Y1 - 2021
N2 - This work proposes a learning model to predict the outcome of electrical defibrillation from ECG signals in ventricular fibrillation (VF) periods, which is a lethal situation happening when a patient is suffering cardiac arrest. An animal experiment of rats is conducted to obtain the ECG signals and necessary information for this study. This proposed model only extracts one feature from the ECG signals and enjoys low computational complexity at both training and testing stages. The statistics of this extracted single feature is further analyzed, and mathematical closed-form formulas for several interesting performance indices including the sensitivity, specificity, accuracy, precision and Area Under the Curve (AUC) are obtained to gain more insights of the proposed system. Moreover, the extracted feature can be treated as a linear combination of individual frequency components of the ECG signal, where the combining coefficients of the linear combination may show informative clinical inference. Frequencies corresponding to large trained combining coefficients imply that they contribute more in distinguishing the defibrillation outcome, and vice versa. As a result, important frequencies of the ECG signals can be identified and insignificant frequencies can also be filtered out by the proposed training. Simulation results corroborate the analytical results, and show that the proposed scheme greatly outperforms several competitive learning models and traditional methods in terms of testing accuracy and computational complexity.
AB - This work proposes a learning model to predict the outcome of electrical defibrillation from ECG signals in ventricular fibrillation (VF) periods, which is a lethal situation happening when a patient is suffering cardiac arrest. An animal experiment of rats is conducted to obtain the ECG signals and necessary information for this study. This proposed model only extracts one feature from the ECG signals and enjoys low computational complexity at both training and testing stages. The statistics of this extracted single feature is further analyzed, and mathematical closed-form formulas for several interesting performance indices including the sensitivity, specificity, accuracy, precision and Area Under the Curve (AUC) are obtained to gain more insights of the proposed system. Moreover, the extracted feature can be treated as a linear combination of individual frequency components of the ECG signal, where the combining coefficients of the linear combination may show informative clinical inference. Frequencies corresponding to large trained combining coefficients imply that they contribute more in distinguishing the defibrillation outcome, and vice versa. As a result, important frequencies of the ECG signals can be identified and insignificant frequencies can also be filtered out by the proposed training. Simulation results corroborate the analytical results, and show that the proposed scheme greatly outperforms several competitive learning models and traditional methods in terms of testing accuracy and computational complexity.
KW - clinical inference via learning
KW - electrical defibrillation
KW - linear discriminant analysis
KW - machine learning
KW - outcome prediction
KW - principal component analysis
KW - statistical model
KW - Ventricular fibrillation
UR - http://www.scopus.com/inward/record.url?scp=85193278692&partnerID=8YFLogxK
U2 - 10.1109/OJCAS.2021.3127270
DO - 10.1109/OJCAS.2021.3127270
M3 - Article
AN - SCOPUS:85193278692
SN - 2644-1225
VL - 2
SP - 686
EP - 699
JO - IEEE Open Journal of Circuits and Systems
JF - IEEE Open Journal of Circuits and Systems
ER -