US Patent 11089989 Shockable heart rhythm classifier for defibrillators

A variety of convolutional neural network based shockable heart rhythm classifiers are described. The neural network is configured to receive an electrocardiogram segment as an input and to generate an output indicative of whether the received electrocardiogram segment represents a heart rhythm that is suitable for treatment by a defibrillation shock. Preferably, the received electrocardiogram segment is not transformed or processed prior to its reception by the convolutional neural network and no features of the electrocardiogram are identified to the convolutional neural network. In some embodiments, the received electrocardiogram segment is the sole input to the convolutional neural network. The described classifier is well suited for use in defibrillators.