Algorithm for exterior defibrillators that correctly identifies both adult and child arrhythmias
- Date:
- March 21, 2011
- Source:
- Elhuyar Fundazioa
- Summary:
- Ventricular fibrillation (VF) is a series of intense and disordered contractions of the ventricles (they are located in the lower part of the heart). In 40 percent of respiratory arrests suffered outside a hospital, it was shown that the first cardiac rhythm corresponds to VF, and the only manner to combat it is with electric discharges provided by defibrillation.
- Share:
Ventricular fibrillation (VF) is a series of intense and disordered contractions of the ventricles (they are located in the lower part of the heart). In 40 % of respiratory arrests suffered outside a hospital, it was shown that the first cardiac rhythm corresponds to VF, and the only manner to combat it is with electric discharges provided by defibrillation. This is why the presence of Automated External Defibrillators (AED) in public places is increasingly common. Telecommunications engineer Unai Irusta has carried out an enhancement-optimisation of these machines, having drawn up an innovative algorithm for them.
His PhD thesis, presented at the University of the Basque Country (UPV/EHU), is entitled New signal processing algorithms for automated external defibrillators.
When an AED is activated, the first thing to be done by it is to examine the heart rhythm (electrocardiogram) of the patient, before applying the first discharge. That is, before anything else, the arrhythmias that can be defibrillated or cannot have to be identified, because the discharges are not useful except in the former. This is precisely where Dr Irusta has carried out his greatest contribution. The arrhythmias of adults and children who suffer these kinds of arrests are not the same, and this researcher has worked on fusing the data from each. With this, he has created an innovative algorithm for AEDs that correctly distinguish the rhythms that can be defibrillated and those that cannot, both with adults and with children.
2,782 records in total
In order to design the algorithm, Dr Irusta started by creating a database of arrhythmias. On the one hand, he gathered and classified 1,090 child arrhythmias; according to the thesis, this is a children's database equal to the most important ones undertaken in the field of AEDs. As regards adults, ha reanalysed an already existing database, to which 928 new records were added. With the sum of both, he completed a database of 2,782 records, of which 1,270 were used for developing the algorithm, and 1,512 for its validation.
With all this, Dr Irusta presented his thesis on the new algorithm made up of four sub-algorithms. These are based on new parameters which were calculated thanks to new records in various zones of the signal, in order to thus detect the arrhythmias; these involve parameters such as time, frequency, inclination and autocorrelation function. The algorithm was verified to have surpassed the minimum laid down by the American Heart Association as regards capacity to detect rhythms which can or cannot be defibrillated, both with adults and with children.
Problems with interference
Dr Irusta also studied the capacity of the new algorithm for identifying data for cardiac arrests recorded by AED electrocardiogram. To deal with the arrest, cardiopulmonary resuscitation (CPR) is employed just up to the moment when an AED can be accessed, given that the impulses on the chest during resuscitation cause interferences in the electrocardiogram. However, in reality it would be more effective to be able to apply the two techniques simultaneously, which is why Dr Irusta has investigated if the new algorithm could read the recording well, even while continuing with CPR.
The new algorithm identified heart rhythms with great precision in those cases in which there are no interferences due to CPR, and surpassing the conditions imposed by the American Heart Association. However, the results were not the ones desired when there was interference from RCP. Dr Irusta has designed a method for obviating the interferences of those recordings which have been contaminated. Thanks to this, it has been possible to precisely identify rhythms which can be defibrillated, but not those which cannot be defibrillated.
About the author
Dr Unai Irusta Zarandona (Bilbao, 1973) is a telecommunications engineer. He drew up his PhD thesis under the direction of Jesús María Ruiz Ojeda, Professor of the Department of Electronics and Telecommunications at the Higher School of Engineering in Bilbao (UPV/EHU). Irusta, Ruiz and other members of this department have published several articles related to this line of research, including high impact journals such as IEEE Transactions on Biomedical Engineering and Resuscitation.
Irusta's research was mainly undertaken at the UPV/EHU. Apart from this, he spent three months in the University of Stavanger in Norway. In the Basque Country he worked with Osatu S. Coop, a company making defibrillators. Dr Irusta is currently lecturer at the Department of Electronics and Telecommunications at the Higher School of Engineering in Bilbao (UPV/EHU).
Story Source:
Materials provided by Elhuyar Fundazioa. Note: Content may be edited for style and length.
Cite This Page: