A novel approach for features extraction in physiological signals

Cosoli, Gloria; Casacanditella, Luigi; Pietroni, Filippo; Calvaresi, Andrea; Revel, Gian Marco; Scalise, Lorenzo

doi:10.1109/memea.2015.7145232

Cited by 26 publications

(15 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…into everyday life. Wearable bio-sensors provide an opportunity for real-time monitoring of human vital signs, thus enabling the possibility for preventive, timely notification and real-time diagnosis [ 8 ]. Unlike commonly-used BP sensors, which demand a specific measurement procedure, modern wearable bio-sensors monitor vital signals on line and all day long, presenting no additional burden other than wearing the device.…”

Section: Introductionmentioning

confidence: 99%

Non-Invasive Blood Pressure Estimation from ECG Using Machine Learning Techniques

Simjanoska

Gjoreski

Gams

et al. 2018

Sensors

136

View full text Add to dashboard Cite

Background: Blood pressure (BP) measurements have been used widely in clinical and private environments. Recently, the use of ECG monitors has proliferated; however, they are not enabled with BP estimation. We have developed a method for BP estimation using only electrocardiogram (ECG) signals. Methods: Raw ECG data are filtered and segmented, and, following this, a complexity analysis is performed for feature extraction. Then, a machine-learning method is applied, combining a stacking-based classification module and a regression module for building systolic BP (SBP), diastolic BP (DBP), and mean arterial pressure (MAP) predictive models. In addition, the method allows a probability distribution-based calibration to adapt the models to a particular user. Results: Using ECG recordings from 51 different subjects, 3129 30-s ECG segments are constructed, and seven features are extracted. Using a train-validation-test evaluation, the method achieves a mean absolute error (MAE) of 8.64 mmHg for SBP, 18.20 mmHg for DBP, and 13.52 mmHg for the MAP prediction. When models are calibrated, the MAE decreases to 7.72 mmHg for SBP, 9.45 mmHg for DBP and 8.13 mmHg for MAP. Conclusion: The experimental results indicate that, when a probability distribution-based calibration is used, the proposed method can achieve results close to those of a certified medical device for BP estimation.

show abstract

Section: Introductionmentioning

confidence: 99%

Non-Invasive Blood Pressure Estimation from ECG Using Machine Learning Techniques

Simjanoska

Gjoreski

Gams

et al. 2018

Sensors

136

View full text Add to dashboard Cite

show abstract

“…valves closure, cardiac contraction, blood momentum changes, etc.). The characteristics of these signals were found to correlate with different cardiovascular pathologies [4]- [7]. These signals may also contain useful information that is complementary to other diagnostic methods [8]- [10].…”

Section: Introductionmentioning

confidence: 99%

Analysis of seismocardiographic signals using polynomial chirplet transform and smoothed pseudo Wigner-Ville distribution

Taebi

Mansy

2017

2017 IEEE Signal Processing in Medicine and Biology Symposium (SPMB)

View full text Add to dashboard Cite

Seismocardiographic (SCG) signals are chest surface vibrations induced by cardiac activity. These signals may offer a method for diagnosing and monitoring heart function. Successful classification of SCG signals in health and disease depends on accurate signal characterization and feature extraction. One approach of determining signal features is to estimate its time-frequency characteristics. In this regard, four different time-frequency distribution (TFD) approaches were used including short-time Fourier transform (STFT), polynomial chirplet transform (PCT), Wigner-Ville distribution (WVD), and smoothed pseudo Wigner-Ville distribution (SPWVD). Synthetic SCG signals with known time-frequency properties were generated and used to evaluate the accuracy of the different TFDs in extracting SCG spectral characteristics. Using different TFDs, the instantaneous frequency (IF) of each synthetic signal was determined and the error (NRMSE) in estimating IF was calculated. STFT had lower NRMSE than WVD for synthetic signals considered. PCT and SPWVD were, however, more accurate IF estimators especially for the signal with time-varying frequencies. PCT and SPWVD also provided better discrimination between signal frequency components. Therefore, the results of this study suggest that PCT and SPWVD would be more reliable methods for estimating IF of SCG signals. Analysis of actual SCG signals showed that these signals had multiple spectral components with slightly time-varying frequencies. More studies are needed to investigate SCG spectral properties for healthy subjects as well as patients with different cardiac conditions.

show abstract

“…In this work, the approach described previously by authors in (Cosoli et al, 2015) has been applied for both rest and motion tests, in order to compute the HR values for both the gold standard and the BH3 raw signals. The algorithm is based on the computation of the slope of the signal, which makes the extraction of relevant features possible, according to the morphology of the signal (Hu et al, 2014).…”

Section: Signals Processingmentioning

confidence: 99%

Smart Monitoring of User’s Health at Home: Performance Evaluation and Signal Processing of a Wearable Sensor for the Measurement of Heart Rate and Breathing Rate

Casaccia

Pietroni

Calvaresi

et al. 2016

Proceedings of the 9th International Joint Conference on Biomedical Engineering Systems and Technologies

View full text Add to dashboard Cite

Nowadays, the monitoring of users' health status is possible by means of smart sensing devices at low-cost and with high measuring capabilities. Wearable devices are able to acquire multiple physiological and physical waveforms and are equipped with on-board algorithms to process these signals and extract the required quantities. However, the performance of such processing techniques should be evaluated and compared to different approaches, e.g. processing of the raw waveforms acquired. In this paper, the authors have performed a metrological characterization of a commercial wearable monitoring device for the continuous acquisition of physiological quantities (e.g. Heart Rate-HR and Breathing Rate-BR) and raw waveforms (e.g. Electrocardiogram-ECG). The aim of this work is to compare the performance of the onboard processing algorithms for the calculation of HR and BR with a novel approach applied to the raw signals. Results show that the HR values provided by the device are accurate enough (±2.1 and ±2.8 bpm in static and dynamic tests), without the need of additional processing. On the contrary, the implementation of the dedicated processing technique for breathing waveform allows to compute accurate BR values (±2.1 bpm with respect to standard equipment).

show abstract

A novel approach for features extraction in physiological signals

Cited by 26 publications

References 16 publications

Non-Invasive Blood Pressure Estimation from ECG Using Machine Learning Techniques

Non-Invasive Blood Pressure Estimation from ECG Using Machine Learning Techniques

Analysis of seismocardiographic signals using polynomial chirplet transform and smoothed pseudo Wigner-Ville distribution

Smart Monitoring of User’s Health at Home: Performance Evaluation and Signal Processing of a Wearable Sensor for the Measurement of Heart Rate and Breathing Rate

Contact Info

Product

Resources

About