Development of an innovative tool for automatic analysis of cardiac arrhythmias and conduction for pediatric patients
This project aimed to develop an innovative tool for the automatic analysis of cardiac arrhythmias and conduction patterns in pediatric patients. Traditional ECG signal analysis tools designed for adults may not be suitable for children due to age-dependent features of the circulatory system and the correlation between signal morphology, amplitude, and age. Using deep neural network algorithms the team successfully built a pioneering tool capable of fast and effective electrocardiological diagnostics in children.
There are several tools for automatic ECG signal analysis in adults, but these solutions may not be reasonable/applicable for the ECG analyses in paediatric population neither in children with normal sinus rhythm nor with various kinds of heart rhythm disturbances. This is the result of age dependent distinct features of the circulatory system such as: strong correlation between morphology, signal amplitude, and the age as well as relevant respiratory sinus arrhythmia or sinus tachycardia.
To develop an innovative tool for the automatic analysis of cardiac arrhythmias and conduction patterns in pediatric patients. To date, developing such a solution for this category has been problematic due to the: strong correlation of signal morphology with patient age, higher physiological heart rate, sinus rhythm irregularity.
Algorithms that enable high-quality ECG signal analysis in children were built. The algorithms were built using deep neural network architectures (e.g., ResNet) and used on filtered ECG signals. The basic method to “train” the algorithm used a database of signals from pediatric patients.
A pioneering tool for performing fast and effective electrocardiological diagnostics in children was created. This tool can be applied in: pediatric cardiology, general pediatrics, sports medicine.
Coordinating Center: Medical University of Warsaw
Project value: PLN 5 363 925.00
Co-financing value: PLN 3,995,230.00
Project co-financed by the European Union through the European Regional Development Fund under the Smart Growth Operational Programme. The project is carried out as part of the National Center for Research and Development: Fast Track: “6 / 1.1.1 / 2020 SS Big/MSP/JN 4”.