Physiological variation of the interval between consecutive heartbeats is known as the heart rate variability (HRV). HRV analysis is traditionally performed on electrocardiograms (ECG signals) and has become a useful tool in the diagnosis of different clinical and functional conditions. The progress in the sensor technique encouraged the development of alternative methods of analyzing cardiac activity: Seismocardiography and gyrocardiography. In our study we performed HRV analysis on ECG, seismocardiograms (SCG signals) and gyrocardiograms (GCG signals) using the PhysioNet Cardiovascular Toolbox. The heartbeats in ECG were detected using the Pan–Tompkins algorithm and the heartbeats in SCG and GCG signals were detected as peaks within 100 ms from the occurrence of the ECG R waves. The results of time domain, frequency domain and nonlinear HRV analysis on ECG, SCG and GCG signals are similar and this phenomenon is confirmed by very strong linear correlation of HRV indices. The differences between HRV indices obtained on ECG and SCG and on ECG and GCG were statistically insignificant and encourage using SCG or GCG for HRV estimation. Our results of HRV analysis confirm stronger correlation of HRV indices computed on ECG and GCG signals than on ECG and SCG signals because of greater tolerance to inter-subject variability and disturbances.

Gyrocardiography (GCG) is a non-invasive technique of analyzing cardiac vibrations by a MEMS (microelectromechanical system) gyroscope placed on a chest wall. Although its history is short in comparison with seismocardiography (SCG) and electrocardiography (ECG), GCG becomes a technique which may provide additional insight into the mechanical aspects of the cardiac cycle. In this review, we describe the summary of the history, definition, measurements, waveform description and applications of gyrocardiography. The review was conducted on about 55 works analyzed between November 2016 and September 2020. The aim of this literature review was to summarize the current state of knowledge in gyrocardiography, especially the definition, waveform description, the physiological and physical sources of the signal and its applications. Based on the analyzed works, we present the definition of GCG as a technique for registration and analysis of rotational component of local cardiac vibrations, waveform annotation, several applications of the gyrocardiography, including, heart rate estimation, heart rate variability analysis, hemodynamics analysis, and classification of various cardiac diseases.

Heart rate variability (HRV) is the physiological variation in the intervals between consecutive heartbeats that reflects the activity of the autonomic nervous system. This parameter is traditionally evaluated based on electrocardiograms (ECG signals). Seismocardiography (SCG) and/or gyrocardiography (GCG) are used to monitor cardiac mechanical activity; therefore, they may be used in HRV analysis and the evaluation of valvular heart diseases (VHDs) simultaneously. The purpose of this study was to compare the time domain, frequency domain and nonlinear HRV indices obtained from electrocardiograms, seismocardiograms (SCG signals) and gyrocardiograms (GCG signals) in healthy volunteers and patients with valvular heart diseases. An analysis of the time domain, frequency domain and nonlinear heart rate variability was conducted on electrocardiograms and gyrocardiograms registered from 29 healthy male volunteers and 30 patients with valvular heart diseases admitted to the Columbia University Medical Center (New York City, NY, USA). The results of the HRV analysis show a strong linear correlation with the HRV indices calculated from the ECG, SCG and GCG signals and prove the feasibility and reliability of HRV analysis despite the influence of VHDs on the SCG and GCG waveforms.

Analyzing biomedical data is a complex task that requires specialized knowledge. The development of knowledge and technology in the field of deep machine learning creates an opportunity to try and transfer human knowledge to the computer. In turn, this fact influences the development of systems for the automatic evaluation of the patient’s health based on data acquired from sensors. Electrocardiography (ECG) is a technique that enables visualizing the electrical activity of the heart in a noninvasive way, using electrodes placed on the surface of the skin. This signal carries a lot of information about the condition of heart muscle. The aim of this work is to create a system for semantic segmentation of the ECG signal. For this purpose, we used a database from Lobachevsky University available on Physionet, containing 200, 10-second, and 12-lead ECG signals with annotations, and applied one-dimensional U-Net with the addition of squeeze-excitation blocks. The created model achieved a set of parameters indicating high performance (for the test set: accuracy—0.95, AUC—0.99, specificity—0.95, sensitivity—0.99) in extracting characteristic parts of ECG signal such as P and T-waves and QRS complex, regardless of the lead.

Heart rate variability (HRV) has become a useful tool of assessing the function of the heart and of the autonomic nervous system. Over the recent years, there has been interest in heart rate monitoring without electrodes. Seismocardiography (SCG) is a non-invasive technique of recording and analyzing vibrations generated by the heart using an accelerometer. In this study, we compare HRV indices obtained from SCG and ECG on signals from combined measurement of ECG, breathing and seismocardiogram (CEBS) database and determine the influence of heart beat detector on SCG signals. We considered two heart beat detectors on SCG signals: reference detector using R waves from ECG signal to detect heart beats in SCG and a heart beat detector using only SCG signal. We performed HRV analysis and calculated time and frequency features. Beat detection performance of tested algorithm on all SCG signals is quite good on 85,954 beats ( $$\text {Se}=0.930$$ , $$\text {PPV}=0.934$$ ) despite lower performance on noisy signals. Correlation between HRV indices was calculated as coefficient of determination ( $$R^2$$ ) to determine goodness of fit to linear model. The highest $$R^2$$ values were obtained for mean interbeat interval ( $$R^2 = 1.000$$ for reference algorithm, $$R^2 = 0.9249$$ in the worst case), $${{\text{PSD}}}_{{\text{LF}}}$$ and $${{\text{PSD}}}_{{\text{HF}}}$$ ( $$R^2 = 1.000$$ for the best case, $$R^2 = 0.9846$$ for the worst case) and the lowest were obtained for $${{\text{PSD}}}_{{\text{VLF}}}$$ ( $$R^2 = 0.0009$$ in the worst case). Using robust model improved achieved correlation between HRV indices obtained from ECG and SCG signals except the $$R^2$$ values of pNN50 values in signals p001–p020 and for all analyzed signals. Calculated HRV indices derived from ECG and SCG are similar using two analyzed beat detectors, except SDNN, RMSSD, NN50, pNN50, and $${{\text{PSD}}}_{{\text{VLF}}}$$ . Relationship of HRV indices derived from ECG and SCG was influenced by used beat detection method on SCG signal.

Due to redundancy of over-dimensioned information, observed often in originally recorded biomedical signals, feature extraction and selection has become focus of much researches connected with biomedical signal processing and classification. Mixed new feature vector combined from time-frequency signal representation (obtained after wavelet transform) and Independent Component Analysis (ICA) applied for non-stationary signals is proposed as a preliminary stage in ECG waveform classification for patients with Atrial Fibrillation (AF). Discrete fast wavelet transform coefficients parameters including energy and entropy measures and components extracted as a result of FastICA algorithm implementation after optimization gave the best classifier performance of whole AF ECG classifier system. System was positively verified on the set of clinically classified ECG signals for control and atrial fibrillation (AF) disease patients taken from MITBIH data base. The measures of specificity and sensitivity computed for the set of 20 AF and 20 patients from control group divided into learning and verifying subsets were used to evaluate presented pattern recognition structure. Different types of wavelet basic functions for feature extraction stage and kernels for SVM classifier structure calculation were tested to find the best system architecture. Obtained results showed, that the ability of generalization and separation for enriched feature extraction based system increased, due to selectively choosing only the most representative features for analyzed AF detection problem.