Rhythm Analysis during Cardiopulmonary Resuscitation Using Convolutional Neural Networks

Abstract

Chest compressions during cardiopulmonary resuscitation (CPR) induce artifacts in the ECG that may provoque inaccurate rhythm classification by the algorithm of the defibrillator. The objective of this study was to design an algorithm to produce reliable shock/no-shock decisions during CPR using convolutional neural networks (CNN). A total of 3319 ECG segments of 9 s extracted during chest compressions were used, whereof 586 were shockable and 2733 nonshockable. Chest compression artifacts were removed using a Recursive Least Squares (RLS) filter, and the filtered ECG was fed to a CNN classifier with three convolutional blocks and two fully connected layers for the shock/no-shock classification. A 5-fold cross validation architecture was adopted to train/test the algorithm, and the proccess was repeated 100 times to statistically characterize the performance. The proposed architecture was compared to the most accurate algorithms that include handcrafted ECG features and a random forest classifier (baseline model). The median (90% confidence interval) sensitivity, specificity, accuracy and balanced accuracy of the method were 95.8% (94.6–96.8), 96.1% (95.8–96.5), 96.1% (95.7–96.4) and 96.0% (95.5–96.5), respectively. The proposed algorithm outperformed the baseline model by 0.6-points in accuracy. This new approach shows the potential of deep learning methods to provide reliable diagnosis of the cardiac rhythm without interrupting chest compression therapy.

Einfluss verschiedener automatisierter externer Defibrillatoren auf die Hands-off-Intervalle von Ersthelfern

“…B. durch das Vorladen des Kondensators, bereits während der Herzrhythmusanalyse, könnten die HOI weiter verkürzen [ 21 ]. Einen neuen Ansatz bilden Algorithmen auf Basis von künstlichen neuronalen Netzwerken und Deep-Learning-Methoden, damit auch während der Herzrhythmusanalyse reanimiert werden kann, ohne die Qualität der Analyse zu beeinträchtigen [ 13 , 14 ]. Bei Asystolie kann zum Beispiel durch einen vorgeschalteten Asystoliedetektor die Abwesenheit von elektrischen Herzaktivitäten identifiziert werden, ohne den kompletten Algorithmus durchlaufen zu müssen [ 13 ].…”

A Machine Learning Model for the Prognosis of Pulseless Electrical Activity during Out-of-Hospital Cardiac Arrest

“…Second, the number of cases included in the study was substantial, but augmenting the dataset’s size would allow the development of more accurate models. In particular, advanced solutions based on deep learning algorithms could also be developed based on features extracted by neural network architectures [ 50 , 51 , 52 ].…”

Detection of shockable rhythms using convolutional neural networks during chest compressions provided by a load distributing band

“…Adding more data would allow the design of deeper networks leading to the extraction of more sophisticated features and so, to better performance. Futhermore, more accurate SAAs have been demonstrated during manual chest compressions when the diagnosis is made using a majority criterion on three shorter consecutive intervals [11]. In brief, there is still room to improve the accuracy of rhythm analysis with concurrent LDB compressions, which may lead to accurate enough solutions that bring uninterrupted mechanical CPR into real practice.…”

“…In the context of supervised learning, deep learning (DL) techniques have proven to be very efficient for biomedical signal classification tasks, contributing to important advances even in the field of OHCA. In particular, recent studies show that deep learning techniques outperform tradiditional machine learning algorithms in the shock/no-shock diagnosis of the rhythm during intervals free of chest compressions [7][8][9][10] and during manual CPR [11].…”