Large network monitoring of electrogenic cells by means of micro electrode arrays (MEA) has become a valuable methodology in electrophysiology. In order to understand the mechanisms of learning (plasticity) and memory in fundamental neuroscience it is imperative to longterm interface large populations of interconnected neurons1. Currently, passive low spatial density MEAs integrate up to 50 electrodes/mm2 which leads to a substantial undersampling of the network since mammalian neuron sizes are in the order of a few tens of micrometers. One important technological bottleneck to increase the spatial resolution of passive MEAs is the routing of each electrode to its external amplifier. Our approach overcomes this limitation by integrating a high density (575 electrodes/mm2) active MEA based on both the active pixel sensor (APS) concept2 and high-speed imaging technologies3. This results in a real-time platform that was validated by acquiring the activity of cardiomyocyte cultures