Resumen
Method for obtaining ECG data, determining if it contains noise patterns from an external assistance device like a Left Ventricular Assist Device (LVAD) and dynamically filtering them. The method comprises the steps of obtaining ECG data from a patient; and checking if the ECG data comprises noise patterns from the external assistance device by analyzing the ECG data using a time and frequency transformation and classifying the analyzed ECG data as with or without LVAD noise patterns, and if the ECG data comprises noise patterns: applying a Fast Fourier Transform for transforming the ECG data into a frequency domain; clustering the ECG data by frequency; searching for a principal rotation frequency and secondary rotation frequencies of the pump; filtering the principal and the secondary rotation frequencies from the ECG data; and obtaining filtered ECG data, and if the ECG data does not comprise noise patterns: outputting the obtained ECG data.
Reivindicaciones
1. A method for obtaining ECG data, comprising the steps of: - obtaining ECG data from a patient; - checking if the ECG data comprises noise patterns from a Left Ventricular Assist Device (LVAD), which comprises a pump, by analyzing the obtained ECG data using a time and frequency transformation and classifying the analyzed ECG data as with or without noise patterns, and - if the ECG data comprises LVAD noise patterns: ∘ applying a frequency transformation for transforming the ECG data into a frequency domain; ∘ search for a principal rotation frequency of the pump in a range previously known; ∘ clustering the ECG data by frequency, generating one or more clusters; ∘ determining secondary rotation frequencies of the pump from clusters generated; ∘ filtering the principal rotation frequency and the secondary rotation frequencies from the ECG data; and ∘ outputting filtered ECG data - if the ECG data does not comprise LVAD noise patterns: ∘ outputting the obtained ECG data as filtered ECG data. 2. The method according to claim 1, further comprising a step of filtering also an alternate current interference, with a previously known frequency. 3. The method according to any of previous claims, wherein the step of checking if the ECG data comprises noise patterns from the LVAD comprises the sub-steps of: - analyzing the ECG data in frequency and time by performing a Continuous Wavelet Transform (CWT), thus, obtaining a scalogram; and - classifying the analyzed ECG data for determining the presence of the LVAD. 4. The method according to claim 3, further comprising a step of determining and outputting actual operation information of the LVAD, being its working frequencies and time on each frequency. 5. The method according to claim 4, further comprising a step of determining anomalies in the LVAD by checking if the principal rotation frequency and the secondary rotation frequencies obtained corresponds to a range previously known. 6. The method according to claim 1, wherein the filtering step is performed by using sequentially dynamic filters. 7. The method according to claim 6, wherein the filtering step is performed by using Notch filters implemented with Infinite Impulse Response (IIR) filters. 8. The method according to claim 3, wherein the step of classifying the analyzed ECG data for determining the presence of a LVAD is performed using deep learning classification or using image processing techniques. 9. The method according to claim 1, wherein the step of applying a frequency transformation is performed using a Fast Fourier Transform (FFT). 10. The method according to claim 1, wherein the step of clustering the ECG data by frequency further comprises a stage of selecting a maximum value of each cluster as each secondary rotation frequency. 11. An ECG system comprising: - one or more electrodes intended to be attached to a patient and configured to obtain ECG measurements; - an ECG module, connected to the electrodes and configured to digitalize and process ECG measurements, and - a processing module connected to the ECG module and configured to perform the steps of the method according to any of claims 1 to 10. 12. The ECG system according to claim 11, wherein the processing module is placed between the electrodes and the ECG module. 13. The ECG system according to claim 11, wherein the processing module is placed after the ECG module. 14. The ECG system according to claim 11, wherein the processing module is placed in a cloud server connected to the ECG module and configured to receive ECG data from said ECG module. 15. The ECG system according to claim 11, wherein the processing module is placed in a mobile device connected to the ECG module and configured to receive ECG data from said ECG module.
