Application of Convolutional Neural Network for Decoding of 12-Lead Electrocardiogram from a Frequency-Modulated Audio Stream (Sonified ECG)
Research of novel biosignal modalities with application to remote patient monitoring is a subject of state-of-the-art developments. This study is focused on sonified ECG modality, which can be transmitted as an acoustic wave and received by GSM (Global System for Mobile Communications) microphones....
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/6/1883 |
| _version_ | 1827305055381880832 |
|---|---|
| author | Vessela Krasteva Ivo Iliev Serafim Tabakov |
| author_facet | Vessela Krasteva Ivo Iliev Serafim Tabakov |
| author_sort | Vessela Krasteva |
| collection | DOAJ |
| description | Research of novel biosignal modalities with application to remote patient monitoring is a subject of state-of-the-art developments. This study is focused on sonified ECG modality, which can be transmitted as an acoustic wave and received by GSM (Global System for Mobile Communications) microphones. Thus, the wireless connection between the patient module and the cloud server can be provided over an audio channel, such as a standard telephone call or audio message. Patients, especially the elderly or visually impaired, can benefit from ECG sonification because the wireless interface is readily available, facilitating the communication and transmission of secure ECG data from the patient monitoring device to the remote server. The aim of this study is to develop an AI-driven algorithm for 12-lead ECG sonification to support diagnostic reliability in the signal processing chain of the audio ECG stream. Our methods present the design of two algorithms: (1) a transformer (ECG-to-Audio) based on the frequency modulation (FM) of eight independent ECG leads in the very low frequency band (300–2700 Hz); and (2) a transformer (Audio-to-ECG) based on a four-layer 1D convolutional neural network (CNN) to decode the audio ECG stream (10 s @ 11 kHz) to the original eight-lead ECG (10 s @ 250 Hz). The CNN model is trained in unsupervised regression mode, searching for the minimum error between the transformed and original ECG signals. The results are reported using the PTB-XL 12-lead ECG database (21,837 recordings), split 50:50 for training and test. The quality of FM-modulated ECG audio is monitored by short-time Fourier transform, and examples are illustrated in this paper and supplementary audio files. The errors of the reconstructed ECG are estimated by a popular ECG diagnostic toolbox. They are substantially low in all ECG leads: amplitude error (quartile range RMSE = 3–7 μV, PRD = 2–5.2%), QRS detector (Se, PPV > 99.7%), P-QRS-T fiducial points’ time deviation (<2 ms). Low errors generalized across diverse patients and arrhythmias are a testament to the efficacy of the developments. They support 12-lead ECG sonification as a wireless interface to provide reliable data for diagnostic measurements by automated tools or medical experts. |
| first_indexed | 2024-04-24T17:50:43Z |
| format | Article |
| id | doaj.art-8dab8259599648d7b55251949488c4b5 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-24T17:50:43Z |
| publishDate | 2024-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-8dab8259599648d7b55251949488c4b52024-03-27T14:04:02ZengMDPI AGSensors1424-82202024-03-01246188310.3390/s24061883Application of Convolutional Neural Network for Decoding of 12-Lead Electrocardiogram from a Frequency-Modulated Audio Stream (Sonified ECG)Vessela Krasteva0Ivo Iliev1Serafim Tabakov2Institute of Biophysics and Biomedical Engineering, Bulgarian Academy of Sciences, Acad. G. Bonchev Str. Bl 105, 1113 Sofia, BulgariaDepartment of Electronics, Faculty of Electronic Engineering and Technologies, Technical University of Sofia, 8 Kliment Ohridski Blvd., 1000 Sofia, BulgariaDepartment of Electronics, Faculty of Electronic Engineering and Technologies, Technical University of Sofia, 8 Kliment Ohridski Blvd., 1000 Sofia, BulgariaResearch of novel biosignal modalities with application to remote patient monitoring is a subject of state-of-the-art developments. This study is focused on sonified ECG modality, which can be transmitted as an acoustic wave and received by GSM (Global System for Mobile Communications) microphones. Thus, the wireless connection between the patient module and the cloud server can be provided over an audio channel, such as a standard telephone call or audio message. Patients, especially the elderly or visually impaired, can benefit from ECG sonification because the wireless interface is readily available, facilitating the communication and transmission of secure ECG data from the patient monitoring device to the remote server. The aim of this study is to develop an AI-driven algorithm for 12-lead ECG sonification to support diagnostic reliability in the signal processing chain of the audio ECG stream. Our methods present the design of two algorithms: (1) a transformer (ECG-to-Audio) based on the frequency modulation (FM) of eight independent ECG leads in the very low frequency band (300–2700 Hz); and (2) a transformer (Audio-to-ECG) based on a four-layer 1D convolutional neural network (CNN) to decode the audio ECG stream (10 s @ 11 kHz) to the original eight-lead ECG (10 s @ 250 Hz). The CNN model is trained in unsupervised regression mode, searching for the minimum error between the transformed and original ECG signals. The results are reported using the PTB-XL 12-lead ECG database (21,837 recordings), split 50:50 for training and test. The quality of FM-modulated ECG audio is monitored by short-time Fourier transform, and examples are illustrated in this paper and supplementary audio files. The errors of the reconstructed ECG are estimated by a popular ECG diagnostic toolbox. They are substantially low in all ECG leads: amplitude error (quartile range RMSE = 3–7 μV, PRD = 2–5.2%), QRS detector (Se, PPV > 99.7%), P-QRS-T fiducial points’ time deviation (<2 ms). Low errors generalized across diverse patients and arrhythmias are a testament to the efficacy of the developments. They support 12-lead ECG sonification as a wireless interface to provide reliable data for diagnostic measurements by automated tools or medical experts.https://www.mdpi.com/1424-8220/24/6/1883artificial intelligencemachine learningtelemedicineremote ECG monitoringECG-to-sound transformationacoustic ECG |
| spellingShingle | Vessela Krasteva Ivo Iliev Serafim Tabakov Application of Convolutional Neural Network for Decoding of 12-Lead Electrocardiogram from a Frequency-Modulated Audio Stream (Sonified ECG) Sensors artificial intelligence machine learning telemedicine remote ECG monitoring ECG-to-sound transformation acoustic ECG |
| title | Application of Convolutional Neural Network for Decoding of 12-Lead Electrocardiogram from a Frequency-Modulated Audio Stream (Sonified ECG) |
| title_full | Application of Convolutional Neural Network for Decoding of 12-Lead Electrocardiogram from a Frequency-Modulated Audio Stream (Sonified ECG) |
| title_fullStr | Application of Convolutional Neural Network for Decoding of 12-Lead Electrocardiogram from a Frequency-Modulated Audio Stream (Sonified ECG) |
| title_full_unstemmed | Application of Convolutional Neural Network for Decoding of 12-Lead Electrocardiogram from a Frequency-Modulated Audio Stream (Sonified ECG) |
| title_short | Application of Convolutional Neural Network for Decoding of 12-Lead Electrocardiogram from a Frequency-Modulated Audio Stream (Sonified ECG) |
| title_sort | application of convolutional neural network for decoding of 12 lead electrocardiogram from a frequency modulated audio stream sonified ecg |
| topic | artificial intelligence machine learning telemedicine remote ECG monitoring ECG-to-sound transformation acoustic ECG |
| url | https://www.mdpi.com/1424-8220/24/6/1883 |
| work_keys_str_mv | AT vesselakrasteva applicationofconvolutionalneuralnetworkfordecodingof12leadelectrocardiogramfromafrequencymodulatedaudiostreamsonifiedecg AT ivoiliev applicationofconvolutionalneuralnetworkfordecodingof12leadelectrocardiogramfromafrequencymodulatedaudiostreamsonifiedecg AT serafimtabakov applicationofconvolutionalneuralnetworkfordecodingof12leadelectrocardiogramfromafrequencymodulatedaudiostreamsonifiedecg |