Nowadays, neurophysiological equipment is, in most cases, a set of closed systems designed to perform specific processing very efficiently. Tasks such as displaying signals, measuring times or amplitudes, quantifying sleep and many others are usually done properly. Why would one be interested in doing similar things with tools that are more difficult to use? I will try to show that they are not exactly similar things.
Let's begin with an example: When you take photographs with a simple camera, the result is good in a sunny day, when the object is not moving and if its distance to the camera is appropriate. More complex cameras allow taking photographs in more difficult circumstances at the price of having to make decisions and, consequently, at the price of having to make much more effort to understand the mechanics of the device. In almost any kind of tool there is a balance between versatility and simplicity. What is not always appreciated is that delegating decisions to automatic devices by hiding the details does not always produce the best results.
In the present tutorial we are going to face the problem of the analysis of the heart rate in the context of a neurophysiological recording. Usually, electrocardiographic signals are systematically recorded in sleep recordings and routine EEGs. We will describe the use of the tools contained in PhysioNet in this context.
The analysis of heart rate in a neurophysiological environment is more and more important. Heart rate is a window to the autonomous nervous system, showing changes associated with sleep and wakefulness, providing a useful means for detecting arousals, and showing specific changes in association with apneas or seizures. As a matter of fact, heart rate is another signal similar to EEG, EOG or EMG. It is a signal that contains a lot of information to link with the remaining signals of our recordings.
The procedure is very similar to the procedure described in Scilab tutorials (also in our site) but in this case we are not going to center on the use of Scilab but in the use of the command interface present in Windows and Unix systems. Since I think that Linux is the best choice we will center on this operating system, although in most cases the tools have been adapted to Windows and you can apply the same concepts with minor modifications.
PhysioNet contains a lot of biological signals coming from different fields. Signals are stored in WFDB signal file formats and they can be very precisely annotated, analyzed and displayed. To do this, PhysioNet also provides free tools to handle the signals. Taken together, the huge collections of data contained in PhysioNet, the fully specified formats and the tools to handle them form an unbeatable team. The goal of this tutorial is to present some of these tools to clinical neurophysiologists.
In Clinical Neurophysiology, most equipment does not include a save as option in formats directly compatible with PhysioNet tools. ASCII (text) files and EDF format are the most widely available options to store data in non-native format for neurophysiological devices. In this tutorial we will use ASCII files as the input of the procedure.
And let us begin. To follow the tutorial, all the tools that you need can be downloaded from the Internet: