Abstract: A time series of face images of a human during a human activity are captured. A first artificial neural network (ANN) processing pipeline processes the captured time series of face images to provide a first attention level indicator signal. An electrophysiological signal indicative of the level of attention of the human during the activity is also captured. A second ANN processing pipeline processes the sensed electrophysiological signal to providing a second attention level indicator signal. A risk indicator signal is then generated based on at least one of the first attention level indicator and second attention level indicator. A user circuit is then triggered as a result of the risk indicator reaching or failing to reach at least one attention level threshold.

Abstract: A time series of face images of a human during a human activity are captured. A first artificial neural network (ANN) processing pipeline processes the captured time series of face images to provide a first attention level indicator signal. An electrophysiological signal indicative of the level of attention of the human during the activity is also captured. A second ANN processing pipeline processes the sensed electrophysiological signal to providing a second attention level indicator signal. A risk indicator signal is then generated based on at least one of the first attention level indicator and second attention level indicator. A user circuit is then triggered as a result of the risk indicator reaching or failing to reach at least one attention level threshold.