Triggering on individual charged particles is becoming more and more important in modern particle physics experiments for efficient background reduction. Studies for the Belle II detector using feed-forward artificial neural networks (Multi Layer Perceptrons, MLP) have shown that a significant part of the background can be rejected already at the first level trigger. The MLP uses the drift time information from the Central Drift Chamber (CDC) for a given track to estimate the z-vertex with a precision of 1 cm, which is a factor of 10 better than the conventional methods.