In most cases, we record the electric potential difference between two points. The instrument used to measure the potential difference is called voltmeter. A voltmeter is not like a thermometer that obtains the temperature of a point. A voltmeter relates the potential of two points. So, the scale of the kind of the voltmeter in which we are interested is calibrated in positive and negative values. Let's see the scale of a voltmeter 2.1
Usually, the voltages in the body are too small to be expressed in Volts. Generally speaking, in Medicine we record signals in mV or 
V. But usually we are not only interested in the absolute value of the potential difference between two points: we are especially interested in its variation with time. If we introduce a chart, we have the main elements of a recording. We can see the variation of electric potential (voltage) with time and how the recording was born. Let's see a recording 2.2
In Medicine, we use many kinds of recordings. Some of them are called electro-something-gram: the electrocardiogram, the electromyogram, the electroencephalogram, the electroretinogram and many others. Besides its differences, they share a lot of properties.
Since our recordings measure very small potential differences, we need another device to amplify the values: the amplifier. An amplifier introduces some subtle conditioning to our recordings. Sometimes, it does not allow the amplification of very fast or very slow changes. It is somewhat more difficult to use than a voltmeter (we have to take more decisions). At this point we have our first definition of a recording:
A recording is a draw of the variation of the potential difference between two points of the body
We are also interested in measuring magnitudes that are not originally potential differences. We want to measure temperatures, pressures, positions... and many other parameters of our body that are not electrical. The solution is easy -we need to convert these non-electrical activities into electrical ones. Imagine that we want to record a sound (e.g., to evaluate whether a patient is snoring while sleeping). Could we imagine a device that hears the sound and translates the variation of the pressure of the air into voltages? Of course, we can use a microphone. We continuously face similar problems: in general these "translators" are called transducers.
A transducer is an electronic device that converts energy from one form to another. We can handle voltages. Since our signal uses volts, we have to calibrate the signal in relation to the original magnitude (degrees, dB or any other magnitude).
Until now, we used only isolated signals. Plotting several signals together has a lot of advantages: "the whole is more than the sum of its parts". Let's us consider the recording of this figure 2.3
Imagine an alien coming to the earth trying to understand something about humans by looking at this recording. Immediately, our visitor detects that the signal marked as BP and the signal marked as ECG have some kind of relation. Much more subtle relations can be seen in this recording but now. I would like to emphasize that the synchronous plotting of several signals adds a lot of information. The relations between different signals can be studied. At this point:
A recording is a draw of one or several signals (originally electrical or converted to electrical by using transducers) whose temporal relations are known
Drawings of this kind are very suggestive to humans. They can be very nice (at first, it is a little bit difficult to catch their beauty, but you can get it with a little of practice). However, these kinds of drawings also have some inconveniences, the worst of them being that they cannot be easily modified once plotted. Once we have a recording, this is frozen. We can make measurements but the recording itself is difficult to modify. In this field, our culture has evolved in the last centuries by converting the physical magnitudes to voltages, but in the last decades voltages have become numbers: analog devices became digital ones.
Do you remember Captain Nemo surrounded by manometers? In the last decades, he would have been surrounded by displays showing the same information. A voltage became a number, the tape recorder became a CD and so on. Biological recordings also changed. Let's see how we can convert our beautiful recordings into numbers at the figure 2.4
The blue signal is an analog signal (the signal given by the amplifier). At short intervals, we measure the voltage and store the numbers. Then, we can reconstruct the whole signal. The nice thing with this procedure is that if we choose properly the interval at which we measure voltages, then there is no loss of information. Do you see the analogy between sampling signals and the resolution of the screen of your computer? A photograph taken with small points cannot be distinguished from the vision of the model.
Converting the signal to numbers is pregnant with many more possibilities that it could seem at first sight. Let's see an example.
We want to extract the changes induced in the cerebral activity by a visual stimulus. To study them we give several flashes and record the EEG. Typically, the changes are very small and they could not be seen in the raw EEG. We need a procedure to separate the changes related to the stimulus from the remainder electroencephalographic activity. The next figure is a simulation of this process. If we give to our subject a lot of stimuli the basal activity is more or less random but the changes induced by the stimuli are constant across sweeps. The beginning of the sweep is coincident with the stimulus. By "averaging" several sweeps, we can extract the change (the right frame). As you can see, the evoked potential is not evident in the raw recording (the left traces). Since we are interested in the Visual Evoked Potential, the electroencephalogram becomes noise. It's life. 2.5
Do not pay too much attention to this point. It is a technical point included to demonstrate that since numbers have their own rules (averaging and so on), we can use these rules to get our goals (understanding and analyzing the recordings to improve the world).
Having a set of numbers instead of a drawing on paper is a very important step forward: our signals can be stored and analyzed by computers. There is a huge amount of knowledge waiting to be applied to our signals. A very large set of methods can be applied to our recordings.
And, as a practical consequence, this feature changed the laboratories of Clinical Neurophysiology. Twenty years ago, a Clinical Neurophysiologist was a person carrying heavy blocks of paper. Nowadays, he or she is a person looking at the screen of a computer. Computers are powerful, cheap and easy to repair, so specific instruments are disappearing: now, the Clinical Neurophysiology laboratory is full of computers. We can summarize the process as:
A recording is a set of numbers stored in the hard disk of our computer derived from the sampling of biological signals
The numbers that constitute the recording are stored in some strange way that our software is able to understand, and the Clinical Neurophysiologist does not need only being a clever person who detect small changes in curves but a person understanding sampling, edition of digital files, and similar issues. And believe it or not, the Clinical Neurophysiologist does not need only to know about the diseases and the shape of the curves: he or she also needs to understand something about mathematics and computers 2.6
Our culture does not stop. In the last decade, a new phenomenon appeared: computers became connected. Firstly through the phone line, then across wired networks, a new phenomenon appeared: the Internet. Since our recording is a file, it can be transferred from one site to another. It is not necessary to use a physical support such as a floppy disk or a magnetic tape record. As a consequence, the methods of analysis developed in one place can be accessible worldwide. Once more our recording changed:
A recording is a file located somewhere in the network derived from the sampling of biological signals
Let me introduce a very interesting resource in the Internet: PhysioNet. 2.6
PhysioNet is a combination of three different kinds of things:
Although it is mainly directed to Cardiology it can be extremely useful for a wider audience.
Would you not like to have a polygraph in you own computer?. This screen was taken from a home computer some days ago (Figure 2.8)
The ECG that you see is a file that has also been downloaded from PhysioNet. It contains a set of numbers coming from an analog ECG. You can download the tools necessary to do it from PhysioNet
Let me say that sampling is not only used by machines. When we see a graph with the blood pressure of a patient taken each 4 hours, this is also a sampled curve.
Converting curves into files is not an exotic phenomenon that belongs only to Clinical Neurophysiology or Psychophysiology. It is used to monitor patients in the Intensive Care Unit, in the Surgery Ward, and in many other places inside and outside a hospital. But even more, probably health care will evolve to predictive maintenance. If you hear a strange noise while driving a car it can be a sign of a failure, which you can correct before the breakdown. We will try to hear the noises of our body by using distant transducers and eventually we will correct the malfunctioning prior to its appearance. We have to be ready to do that.
Since this is a practical seminar, those people wanting to receive some aid to install PhysioNet tools in their computer, please sign the paper and you will receive some aid to do it, although you can do it by yourselves entering in PhysioNet
