SamplePage

This software detects electrophysiological spikes from extracellular recordings.

The following pictures show a typical extracelluar recording (above: x = time (ms); y = current (A)) and my spike detector output (below: x = time (ms); y = 1 on a spike position, 0 elsewhere):

Or if you zoom out:

The prevailing problem with this kind of analysis is that there is a mild baseline fluctuation and the sampling rate may vary among data files. This spike detecting application (recount2.exe) is not bothered by such unwanted fluctuations. It takes a running average (of any size) and compares the average with a 4 points after the current position. If it satisfies a primary threshold condition, it checks another 4 points forward in time and checks if a second threshold condition is met. This way, the detector picks up only a spike of a desired shape. In the 20-minute recording example from the rat suprachiasmatic nucleus, as presented above, it worked with 100% success rate. But the success rate may vary depending on the seal resistance.