Andalusian researchers work on a software to study heart activity during sleep

Málaga 21 May 2008Human beings are determined to control, change and understand many issues related to health, like the breathing rhythm and cardiac frequency. In order to study it in depth, experts need to have the tools that automatically analyse the data obtained. That's why diagnosis and therapies can show some light into this.


A group of researchers of the University of Málaga (UMA) have decided to improve one of such tools to apply it, for example, to the influence of the different sleep stages in the cardiac activity. These researchers of the Applied Physics department of the Telecommunication Engineering Faculty (UMA) collaborated a few years ago in a project of the National Institutes of Health (NIH) in the USA in which they tried to apply a segmentation algorithm of temporary, previously developed by the UMA, to identify the different cardiac activity schemes that follow one another while we sleep and try to relate them to the different sleep stages.

The idea behind these experiments was to represent the changes that take place, for example, in the cardiac frequency of healthy and ill people, as well as astronauts, while they were sleeping. The project did not reveal the expected results, probably because the designed computing algorithms were not the most appropriate ones to measure such a complex activity. This research group will now have three years to start to develop a method to face this challenge. The project has been funded with 67.800 euro as part of the order passed by the Andalusian Ministry of Innovation, Science and Enterprise within the project of excellence framework.

According to the leading researcher, Dr. Bernaola Galván, the difficulty in finding out the patterns of cardiac activity is the enormous complexity of these data. Contrary to what has been usually thought, in the normal activity states cardiac rhythm is very variable and this results in series of heterogeneous data at all levels. "These patterns have a fractal nature. However, if we isolated just a small part of them, we would observe that it is an identical copy of the same whole", he explained. For example, we can imagine a tree branch whose sprouts are new branches which, in turn, will come into bud of identical structure ad infinitum.

The value of these algorithms as a diagnosis tool in patients with sleep apnea, for example, could be very interesting. However, in order to reach that point, scientists would have to solve this complex Rubick's cube before. "The first step will consist of making many numeric simulations, which will result in a large number of artificial fractal sequences with which a computing "training" programme will be implemented, Dr. Bernaola Galván explained. Researchers hope that, with this "training" the software can automatically find the points where relevant changes in the analysed data series take place.

The research group does not rule out the possibility of studying new scenarios. "We would like to apply our techniques to encephalogram series while doing physical activities", Dr. Bernaola Galván explained. These scientists have not stopped widening their field of study ever since they started their research line in the 1990's. The segmentation algorithms initially developed by this group were applied to the search for areas of different proportions between nucleotides in DNA sequences. This led them to collaborate with Dr. Bernardi, author of the isochors theory, the first attempt to describe the structure of the human genome at a large scale.

Another relevant work carried out by this team consisted of measuring the heart rhythm in patients with cardiac congestive failure, in collaboration with Harvard University hospital. The research revealed that the cardiac activity of healthy people and ill people showed areas with the same variations, provided that they were subjected to the same stimulus. This indicated that both types of people react, or try to react, to stimulus in a similar way. However, the determining factor for this to happen was the extent of such changes, which was much smaller in the case of ill people.

Yet more of their project experience includes the application of their algorithms to study sun radiation in collaboration with the University of Málaga group, led by Dr. Sidrach de Cardona. Among other things, these studies aim to determine whether or not solar energy facilities are profitable, the frequency of sunny days, and how the atmosphere evolves in a particular place, to name but a few.

The research group of the Applied Physics department of the Telecommunication Engineering Faculty (UMA) that will be working in the current project of excellence is made up by five members. However, there will also be a researcher from the University of Boston involved as a result of the many international collaborations of the group.

The beginning of this sort of temporary series analysis dates back to the 1970's, and although it is still applied to several fields today, it actually developed as a result of the need to control and forecast changes in the coal production in mines, due to the possibility of running out of it.

Leslie Versweyveld

[Medical IT News][Calendar][Virtual Medical Worlds Community][News on Advanced IT]