Automated positioning of extracellular multi electrodes for online spike sorting

Extracellular recordings are a key tool to study the activity of neurons in vivo. Especially in the case of experiments with behaving animals, however, the tedious procedure of electrode placement can take a considerable amount of expensive and restricted experimental time. Furthermore, due to tissue drifts and other sources of variability in the recording setup, the position of the electrodes with respect to the neurons under study can change, causing low recording quality. Here, we developed a system for automatic electrode placement and its evaluation. The main features of this study are: First, we introduce a quality measure for the recording position of the electrode which should be maximized during recordings and is particularly suitable for the use of multi-trodes, e.g. tetrodes. This quality measure is based on the detected spike waveforms only and does not rely on computationally expensive clustering or spike sorting. It is also invariant for the discharge rate of the neurons. Second, an automated positioning system based on this quality measure is proposed. It estimates the recording quality at the current recording position. It is able to find favorable recording positions by locally maximizing the quality measure and adopts the electrode position smoothly to changes of the neuron positions. If the quality decreases, it will either try to maximize the quality by accounting for the changes in neuron positions or to find a new recording position where action potentials can be measured with a high signal to noise ratio. Third, we evaluate the system using a new simulator for extracellular recordings based on realistically reconstructed 3D neurons. The shape of the extracellular waveform is estimated from their morphology for every point on a 3D grid around the neurons. If a recording device is close to a neuron, the corresponding waveform for its spikes is calculated from that grid by interpolating the waveforms of the adjacent grid positions. This way we can simulate a realistic recording environment in which an unconstrained movement of electrodes and neurons and an interaction with the positioning system is possible