L. Sornmo

Politecnico di Milano, Milano, Lombardy, Italy

Are you L. Sornmo?

Claim your profile

Publications (97)40.46 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: The atrioventricular (AV) node plays a central role during atrial fibrillation (AF). We have recently proposed a statistical AV node model defined by parameters characterizing the arrival rate of atrial impulses, the probability of an impulse choosing either one of the dual AV nodal pathways, the refractory periods of the pathways, and the prolongation of refractory periods. All model parameters are estimated from the RR series using maximum likelihood (ML) estimation, except for the mean arrival rate of atrial impulses which is estimated by the AF frequency derived from the f-waves. The aim of this study is to present a unified approach to ML estimation which also involves the shorter refractory period, thus avoiding our previous Poincaré plot analysis which becomes biased. In addition, the number of RR intervals required for accurate parameter estimation is presented. The results show that the shorter refractory period can be accurately estimated, and that the resulting estimates converge to the true values when about 500 RR intervals are available.
    Conference proceedings: ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference 07/2013; 2013:2567-2570.
  • [Show abstract] [Hide abstract]
    ABSTRACT: The goal of the study is to determine whether photoplethysmography (PPG) can replace ECG-based detection of heart rate turbulence. Using the PPG, classification of ventricular premature beats (VPBs) is accomplished with a linear classifier. The two conventional parameters turbulence onset and slope are studied together with a recently introduced parameter characterizing turbulence shape. Performance is studied on a dataset with 4131 VPBs, recorded from a total of 27 patients in different clinical contexts (hemodialysis treatment, intensive care monitoring, and electrophysiological study). The sensitivity/ specificity of VPB classification was found to be 90.5/99.9%, with an accuracy of 99.3%, suggesting that classification of VPBs can be reliable made from the PPG. The main difference between the two types of turbulence analysis stems from the fact that the pulse transit time varies largely immediately after the VPB. Out of the 22 patients which had a sufficient number of VPBs, the outcome of ECG- and PPG-based analysis was identical in 21. It is concluded that the PPG may serve as a surrogate technique for the ECG in turbulence analysis.
    IEEE transactions on bio-medical engineering 06/2013; · 2.15 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Determination of heart status during dialysis can improve patient monitoring. Pressure sensors in the dialysis machine measures the heart pulses that propagates in the body and enter the extracorporeal blood circuit. A peristaltic blood pump, located in the same circuit, introduces strong periodic pressure pulses that interfere with the much weaker cardiac component. These signal characteristics make the extraction of the heart activity challenging. In the present study, we explore the possibility to extract and analyze the cardiac component using simulated data. The accuracy of the timing of each heartbeat is analyzed. Additionally, the heart component is extracted from patient pressure recordings, and compared to the heart rate computed from a photoplethysmogram. The results show that heart timings can be accurately determined using the pressure sensors of a dialysis machine.
    Computing in Cardiology Conference (CinC), 2013; 01/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this work is to develop a method for detection of brief episode paroxysmal atrial fibrillation (PAF). The proposed method utilizes four different features: RR interval irregularity, absence of P waves, presence of f-waves and noise level. The obtained features are applied to the Mamdani-type fuzzy inference method for decisionmaking. The performance was evaluated on one hundred 90 s long surrogate ECG signals with brief PAF episodes (5–30 beats). The robustness to noise in ECGs where noise level in each set is incremented in steps of 0.01 mV from 0 to 0.2 mV was examined as well. When compared to the coefficient of sample entropy, our method showed considerably better performance for low and moderate noise levels (< 0.06 mV) with an area under the receiver operating characteristic curve of 0.9 and 0.94, respectively. Similar performance is expected for higher noise levels as atrial activity is less used in the detection process. Finally, the results suggest that our method is more robust to false alarms due to ectopic beats or other irregular rhythms than the method under comparison.
    Computing in Cardiology Conference (CinC), 2013; 01/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: Intradialytic hypotension (IDH) is the most common complication during hemodialysis; early prediction and prevention of IDH would dramatically improve the living conditions for patients with end stage renal disease. A recently published study suggests that a decrease in the envelope of the photoplethysmograpy (PPG) signal can be used for predicting acute symptomatic IDH. In the present study, the PPG based method is extended by introducing a patient dependent detection threshold, which involves information on heart rate variability (HRV) and heart rate turbulence (HRT) from the current dialysis session. This is motivated since several studies have found significant differences in HRV and HRT between hypotension-prone and hypotension-resistant patients. Recordings from 15 patients during 38 hemodialysis sessions were used to evaluate the method. Symptomatic IDH was correctly predicted in 9 out of 14 cases, while 5 out of 24 cases were falsely predicted. The performance was better for acute symptomatic IDH, 5 out of 5 cases were correctly predicted. The present method represents a novel approach to combining information derived from ECG and PPG signals.
    Computing in Cardiology Conference (CinC), 2013; 01/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of the present study is to evaluate the effect of rate control drugs on the AV node characteristics during atrial fibrillation (AF) using a model-based approach. A statistical model of the AV nodal function is employed, defined by parameters which characterize the arrival rate of atrial impulses, the refractoriness of the fast and the slow AV-nodal pathway and the probability of atrial impulse to pass through either of the two pathways. The RATAF (RATe control in Atrial Fibrillation) study database consists of recordings from 60 patients with permanent AF at baseline and on treatment with metoprolol, verapamil, diltiazem and carvedilol, respectively. The resulting model parameter estimates indicate that the refractory period of the slow pathway as well as that of the fast pathway increased significantly during treatment with all four drugs. The results suggest that the proposed AV-node model can be used for non-invasive evaluation of the effect of rate control drugs.
    Computing in Cardiology Conference (CinC), 2013; 01/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: The goal of this study is to determine whether peripheral blood volume fluctuations triggered by ventricular premature beats (VPBs) are significantly related to hypotensive symptoms during dialysis treatment. Patients treated with hemodialysis often suffer from cardiovascular disorders and uremic neuropathy, increasing the propensity to homeostatic imbalance that, in turn, may result in intradialytic hypotension, cramps, nausea, dizziness, headache and other complications. VPBs, being abundant in hemodialysis patients, can be viewed as an internal disturbance leading to imbalance through acute blood pressure drop and prolonged tissue deoxygenation. The present study investigates and quantifies VPB-induced relative peripheral blood volume changes, measured from the fingertip photoplethysmographic (PPG) waveform, and their significance for characterization of physiological recovery of a disturbed circulatory state. The mean decrease in PPG amplitude, corresponding to an initial post-ectopic drop in blood volume delivered to the periphery, was 4 ± 3% in asymptomatic treatments, whereas 17 ± 3% in symptomatic dialysis treatments. This result indicates that significant differences exist between the two groups of treatment, providing a potential for development of intradialytic risk predictors.
    Computing in Cardiology Conference (CinC), 2013; 01/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: A novel method for QRST cancellation during atrial fibrillation (AF) is introduced for use in recordings with two or more leads. The method is based on an echo state neural network which estimates the time-varying, nonlinear transfer function between two leads, one lead with atrial activity and another lead without, for the purpose of canceling ventricular activity. The network has different sets of weights that define the input, hidden, and output layers, of which only the output set is adapted for every new sample to be processed. The performance is evaluated on ECG signals, with simulated f-waves added, by determining the root mean square error between the true f-wave signal and the estimated signal, as well as by evaluating the dominant AF frequency. When compared to average beat subtraction (ABS), being the most widely used method for QRST cancellation, the performance is found to be significantly better with an error reduction factor of 0.24-0.43, depending on f-wave amplitude. The estimates of dominant AF frequency are considerably more accurate for all f-wave amplitudes than the AF estimates based on ABS. The novel method is particularly well suited for implementation in mobile health systems where monitoring of AF during extended time periods is of interest.
    IEEE transactions on bio-medical engineering 08/2012; 59(10):2950-7. · 2.15 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A novel method for QRST cancellation during atrial fibrillation (AF) is introduced for use in recordings with two or more leads. The method is based on an echo state neural network (ESN) which estimates the time-varying, nonlinear transfer function between two leads, one lead with atrial activity and another lead without, for the purpose of canceling ventricular activity. The performance is evaluated on ECG signals, with simulated f-waves of low amplitude added, by determining the root mean square error P between the true f-wave signal and the estimated signal, as well as by evaluating the dominant AF frequency. When compared to average beat subtraction (ABS), being the most widely used method for QRST cancellation, the performance is found to be significantly better with equal to mean and standard deviation of PESN 24.8±7.3 and PABS 34.2±17.9 μV (p
    Computing in Cardiology (CinC), 2012; 01/2012
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Episodes of hypotension during hemodialysis treatment constitutes an important clinical problem which has received considerable attention in recent years. Despite the fact that numerous approaches to reducing the frequency of intradialytic hypotension (IDH) have been proposed and evaluated, the problem has not yet found a definitive solution-an observation which, in particular, applies to episodes of acute, symptomatic hypotension. This overview covers recent advances in methodology for predicting and preventing IDH. Following a brief overview of well-established hypotension-related variables, including blood pressure, blood temperature, relative blood volume, and bioimpedance, special attention is given to electrocardiographic and photoplethysmographic (PPG) variables and their significance for IDH prediction. It is concluded that cardiovascular variables which reflect heart rate variability, heart rate turbulence, and baroreflex sensitivity are important to explore in feedback control hemodialysis systems so as to improve their performance. The analysis of hemodialysis-related changes in PPG pulse wave properties hold considerable promise for improving prediction.
    IEEE reviews in biomedical engineering. 01/2012; 5:45-59.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Advanced analysis of atrial fibrillation (AF) intra — cardiac electrograms aims to establish clinical targets for ablation. Frequency domain approaches, such as Dominant Frequency Analysis (DFA), estimate the dominant frequency to identify cardiac sites with high activation rates as ablation targets. However, they often discard relevant information in the spectrum, such as the harmonic structure or the spectral envelope. Moreover, these methods do not provide a complete characterization of complex atrial fibrillation signals. We propose to use the correntropy function to estimate the fundamental frequency, instead of dominant frequency, in order to quantify the activation rate of AF signals. We also propose to use the correntropy method combined with Fourier Organization Analysis (FOA), a method previously proposed to model fibril-latory signals, in order to estimate more than one fundamental frequency in complex signals which can be the result of the interaction of multiple wavefronts. The proposed approach was benchmarked by using pseudo-real AF signals for assessing their performance in a controlled environment. Two datasets of AF real signals were assembled, one with regular and simple signals, and another one with complex signals. Correntropy estimation of fundamental frequency, combined with FOA, allowed not only to characterize the periodicity of AF signals, but also to study more complex signals by modelling them with more than one component.
    Cognitive Information Processing (CIP), 2012 3rd International Workshop on; 01/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper introduces a model of the atrioventricular node function during atrial fibrillation (AF), and describes the related ECG-based estimation method. The proposed model is defined by parameters that characterize the arrival rate of atrial impulses, the probability of an impulse choosing either one of the two atrioventricular nodal pathways, the refractory periods of these pathways, and the prolongation of the refractory periods. These parameters are estimated from the RR intervals using maximum likelihood estimation, except for the shorter refractory period which is estimated from the RR interval Poincaré plot, and the mean arrival rate of atrial impulses by the AF frequency. Simulations indicated that 200-300 RR intervals are generally needed for the estimates to be accurate. The model was evaluated on 30-min ECG segments from 36 AF patients. The results showed that 88% of the segments can be accurately modeled when the estimated probability density function (PDF) and an empirical PDF were at least 80% in agreement. The model parameters were estimated during head-up tilt test to assess differences caused by sympathetic stimulation. Both refractory periods decreased as a result of stimulation, and the likelihood of an impulse choosing the pathway with the shorter refractory period increased.
    IEEE transactions on bio-medical engineering 08/2011; 58(12):3386-95. · 2.15 Impact Factor
  • Source
    E Gil, L Sornmo, P Laguna
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study, alterations in the cardiovascular system caused by ventricular premature beats (VPBs) are investigated by analyzing the photoplethysmographic (PPG) signal. A simple algorithm for PPG-based detection of VPBs is devised and evaluated, and then employed for the analysis of heart rate turbulence (HRT), here labelled “pulse rate turbulence” (PRT). The pulse transit time is also studied as it constitutes the main difference between HRT and PRT. The data sets included a total of 3872 VPBs and 13169 normal beats. The results showed that VPBs can be detected from the PPG signal with a sensitivity of 92.8%, a specificity of 99.8% and an accuracy of 99.3%, using six features and a simple linear classifier. The shape of PRT was found to resemble that of HRT, the latter type of turbulence resulting from ECG-based analysis, suggesting that PRT analysis can be used as a replacement for HRT analysis when the ECG is not available.
    01/2011;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Acute hypotensive episodes are common during dialysis sessions, and represent a serious problem. Spectral analysis of heart rate variability (HRV) and barorrefle x sensitivity (BRS) is performed to study the behaviour of the autonomic nervous system (ANS) during the hemodialysis. The ratio between the low frequency (LF) and high frequency (HF) power of HRV, as well as BRS in the HF band, are significantly different in patients being prone and resistant to hypotension (p
    01/2011;
  • [Show abstract] [Hide abstract]
    ABSTRACT: We study a model of the atrioventricular node function during atrial fibrillation (AF), for which the model parameters can be estimated from the ECG. The proposed model is defined by parameters which characterize the arrival rate of atrial impulses, the probability of an impulse choosing either one of the two atrioventricular nodal pathways, the refractory periods of these pathways, and the prolongation of the refractory periods. The parameters are estimated from the RR intervals using maximum likelihood estimation, except for the shorter refractory period which is estimated from the RR interval Poincare plot, and the mean arrival rate of atrial impulses by the AF frequency estimated from the ECG. The model was evaluated on 30-min ECG segments from 36 AF patients. The results showed that 88% of the segments can be accurately modeled when the estimated probability density function (PDF) and an empirical PDF were at least 80% in agreement.
    Computing in Cardiology, 2011; 01/2011
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A generalized likelihood ratio test (GLRT) statistic is proposed for detection of heart rate turbulence (HRT), where a set of Karhunen-LoE??ve basis functions models HRT. The detector structure is based on the extended integral pulse frequency modulation model that accounts for the presence of ectopic beats and HRT. This new test statistic takes a priori information regarding HRT shape into account, whereas our previously presented GLRT detector relied solely on the energy contained in the signal subspace. The spectral relationship between heart rate variability (HRV) and HRT is investigated for the purpose of modeling HRV ??noise?? present during the turbulence period, the results suggesting that the white noise assumption is feasible to pursue. The performance was studied for both simulated and real data, leading to results which show that the new GLRT detector is superior to the original one as well as to the commonly used parameter turbulence slope (TS) on both types of data. Averaging ten ventricular ectopic beats, the estimated detection probability of the new detector, the previous detector, and TS were found to be 0.83, 0.35, and 0.41, respectively, when the false alarm probability was held fixed at 0.1.
    IEEE Transactions on Biomedical Engineering 03/2010; · 2.35 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study we propose the correntropy function as a discriminative measure for detecting nonlinearities in the respiratory pattern of chronic heart failure (CHF) patients with periodic or nonperiodic breathing pattern (PB or nPB, respectively). The complexity seems to be reduced in CHF patients with higher risk level. Correntropy reflects information on both, statistical distribution and temporal structure of the underlying dataset. It is a suitable measure due to its capability to preserve nonlinear information. The null hypothesis considered is that the analyzed data is generated by a Gaussian linear stochastic process. Correntropy is used in a statistical test to reject the null hypothesis through surrogate data methods. Various parameters, derived from the correntropy and correntropy spectral density (CSD) to characterize the respiratory pattern, presented no significant differences when extracted from the iteratively refined amplitude adjusted Fourier transform (IAAFT) surrogate data. The ratio between the powers in the modulation and respiratory frequency bands R was significantly different in nPB patients, but not in PB patients, which reflects a higher presence of nonlinearities in nPB patients than in PB patients.
    Conference proceedings: ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference 01/2010; 2010:2399-402.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a user-friendly interface in Matlab<sup>®</sup>, called BioSigBrowser, that aims to facilitate the use of algorithms in biomedical signal processing. It includes methods related with cardiovascular signal processing, namely some multimodal analysis. This platform can treat a single signal or work in a batch mode on a given database as is usual in research. Furthermore, its modular characteristic allows easy incorporation of new methods.
    Information Technology and Applications in Biomedicine, 2009. ITAB 2009. 9th International Conference on; 12/2009
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Heart rate turbulence (HRT) is commonly assessed by two parameters: turbulence onset (TO) and turbulence slope (TS), both obtained by averaging RR tachograms following a ventricular premature beat (VPB). It has been recently shown that a model-based detection-theoretical approach results in HRT indices outperforming TO/TS in identifying the presence or absence of HRT. The aim of this work is to evaluate the risk stratification ability of this approach, as compared with the classical parameters TO and TS, in a population of 96 ischemic patients with mild to moderate congestive heart failure. We found significant differences (Mann-Withney U test) between survivors and cardiac death groups for TS and the new parameter T<sub>¿</sub>(x). Survival analysis showed that T<sub>¿</sub>(x) is the HRT index with highest association to risk of cardiac death (hazard ratio=2.8, p =0.008). Results show improved risk stratification of the new description of HRT with respect to classical parameters.
    Computers in Cardiology, 2009; 10/2009
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this work, a method for non-invasive assessment of AF organization has been applied to discriminating between paroxysmal and long-term persistent AF episodes. Following extraction of the atrial activity (AA) signal, the dominant atrial frequency (DAF) of the AA was computed based on a hidden Markov model. Finally, the main atrial wave (MAW) was obtained by bandpass filtering centered on the DAF, thus providing a time series suitable for AF organization estimation with sample entropy (SampEn). The performance of the method was evaluated on 24-h Holter recordings with long-term changes in AF organization. The results showed that episodes of paroxysmal AF (0.0693 ± 0.0147) were consistently associated with lower SampEn than episodes with persistent AF (0.1056 ± 0.0146). Moreover, 94.2% of 1-min segments with paroxysmal AF and 88.6% of 1-min segments with persistent AF could be correctly classified based on SampEn information, thus making it possible to classify longterm recordings of patients without AF history.
    Computers in Cardiology, 2009; 10/2009

Publication Stats

356 Citations
40.46 Total Impact Points

Institutions

  • 2008–2011
    • Politecnico di Milano
      • Department of Bioengineering
      Milano, Lombardy, Italy
  • 1992–2010
    • Lund University
      • • Department of Electrical and Information Technology
      • • Department of Cardiology
      • • Department of Electroscience
      Lund, Skane, Sweden
  • 1997–2009
    • University of Zaragoza
      • • Engineering Research Institute of Aragon
      • • Department of Electrical Engineer and Communications
      Zaragoza, Aragon, Spain
    • University of Florence
      Florens, Tuscany, Italy
  • 2006
    • Otto-von-Guericke-Universität Magdeburg
      Magdeburg, Saxony-Anhalt, Germany
  • 1998–2006
    • Kaunas University of Technology
      • Department of Telecommunications
      Caunas, Kauno Apskritis, Lithuania
  • 2005
    • University Hospital Magdeburg
      Magdeburg, Saxony-Anhalt, Germany
  • 2004
    • Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center
      Torrance, California, United States
    • Good Samaritan Hospital Los Angeles
      Los Angeles, California, United States