In the present review we will describe and discuss the physiological and technological background necessary in understanding the dynamic parameters of fluid responsiveness and how they relate to recent softwares and algorithms' applications. We will also discuss the potential clinical applications of these parameters in the management of patients under general anesthesia and mechanical ventilation along with the potential improvements in the computational algorithms.

Biomedical signals are nonstationary in nature, namely, their statistical properties are time-dependent. Such changes in the underlying statistical properties of the signal and the effects of external noise often affect the performance and applicability of automatic signal processing methods that require stationarity. A number of methods have been proposed to address the problem of finding stationary signal segments within larger nonstationary signals. In this framework, processing and analysis are applied to each resulting locally stationary segment separately. The method proposed in this paper addresses the problem of finding locally quasi-stationary signal segments. Particularly, our proposed algorithm is designed to solve the specific problem of segmenting semiperiodic biomedical signals corrupted with broadband noise according to the various degrees of external noise power. It is based on the sample entropy and the relative sensitivity of this signal regularity metric to changes in the underlying signal properties and broadband noise levels. The assessment of the method was carried out by means of experiments on ECG signals drawn from the MIT-BIH arrhythmia database. The results were measured in terms of false alarms based on the changepoint detection bias. In summary, the results achieved were a sensitivity of 97%, and an error of 16% for records corrupted with muscle artifacts.

We describe an improved automatic algorithm to estimate the pulse-pressure-variation (PPV) index from arterial blood pressure (ABP) signals. This enhanced algorithm enables for PPV estimation during periods of abrupt hemodynamic changes. Numerous studies have shown PPV to be one of most specific and sensitive predictors of fluid responsiveness in mechanically ventilated patients. The algorithm uses a beat detection algorithm to perform beat segmentation, kernel smoothers for envelope detection, and a suboptimal Kalman filter for PPV estimation and artifact removal. In this paper, we provide a detailed description of the algorithm and assess its performance on over 40 h of ABP signals obtained from 18 mechanically ventilated crossbred Yorkshire swine. The subjects underwent grade V liver injury after splenectomy, while receiving mechanical ventilation, and general anesthesia with isoflurane. All subjects in the database underwent a period of abrupt hemodynamic change after an induced grade V liver injury involving severe blood loss resulting in hemorrhagic shock, followed by fluid resuscitation with either 0.9% normal saline or lactated ringers solutions. Trained experts manually calculated PPV at five time instances during the period of abrupt hemodynamic changes. We report validation results comparing the proposed algorithm against a commercial system (pulse contour cardiac output, PICCO) with continuous PPV monitoring capabilities. Both systems were assessed during periods of abrupt hemodynamic changes against the "gold-standard" PPV, calculated and manually annotated by experts. Our results indicate that the proposed algorithm performs considerably better than the PICCO system during regions of abrupt hemodynamic changes.

We present a novel method to iteratively calculate discrete Fourier transforms for discrete time signals with sample time intervals that may be widely nonuniform. The proposed recursive Fourier transform (RFT) does not require interpolation of the samples to uniform time intervals, and each iterative transform update of N frequencies has computational order N. Because of the inherent non-uniformity in the time between successive heart beats, an application particularly well suited for this transform is power spectral density (PSD) estimation for heart rate variability. We compare RFT based spectrum estimation with Lomb-Scargle Transform (LST) based estimation. PSD estimation based on the LST also does not require uniform time samples, but the LST has a computational order greater than Nlog(N). We conducted an assessment study involving the analysis of quasi-stationary signals with various levels of randomly missing heart beats. Our results indicate that the RFT leads to comparable estimation performance to the LST with significantly less computational overhead and complexity for applications requiring iterative spectrum estimations.

We present a novel parametric power spectral density (PSD) estimation algorithm for nonstationary signals based on a Kalman filter with variable number of measurements (KFVNM). The nonstationary signals under consideration are modeled as time-varying autoregressive (AR) processes. The proposed algorithm uses a block of measurements to estimate the time-varying AR coefficients and obtains high-resolution PSD estimates. The intersection of confidence intervals (ICI) rule is incorporated into the algorithm to generate a PSD with adaptive window size from a series of PSDs with different number of measurements. We report the results of a quantitative assessment study and show an illustrative example involving the application of the algorithm to intracranial pressure signals (ICP) from patients with traumatic brain injury (TBI).

In this communication, we estimated the Lempel-Ziv complexity (LZC) on over 40 h of arterial blood pressure (ABP) recordings corresponding to 18 mechanically ventilated animal subjects. In this study, all subjects underwent a period of abrupt hemodynamic changes after an induced injury involving severe blood loss leading to hemorrhagic shock, followed by fluid resuscitation using either lactated ringers or 0.9% normal saline. The LZC metric experienced a statistically significant increase (p < 0.01) immediately following the induced injury and a statistically significant reduction following the administration of fluid therapy (p < 0.01). These results indicate that LZC of ABP may be useful as a dynamic metric to assess fluid responsiveness.

Cardiovascular signals such as arterial blood pressure (ABP), pulse oximetry (POX), and intracranial pressure (ICP) contain useful information such as heart rate, respiratory rate, and pulse pressure variation (PPV). We present a novel state-space model of cardiovascular signals and describe how it can be used with the extended Kalman filter (EKF) to simultaneously estimate and track many cardiovascular parameters of interest using a unified statistical approach. We analyze data from four databases containing cardiovascular signals and present representative examples intended to illustrate the versatility, accuracy, and robustness of the algorithm. Our results demonstrate the ability of the algorithm to estimate and track several clinically relevant features of cardiovascular signals. We illustrate how the algorithm can be used to elegantly solve several actively researched and clinically significant problems including heart and respiratory rate estimation, artifact removal, pulse morphology characterization, and PPV estimation.

T-wave alternans (TWA) are beat-to-beat amplitude oscillations in the T-waves of electrocardiograms (ECGs). Numerous clinical studies have demonstrated the link between these oscillations and ventricular arrhythmias. Several methods have been developed in recent years to detect and quantify this important feature. Most methods estimate the amplitude differences between pairs of consecutive T-waves. One such method is known as modified moving average (MMA) analysis. The TWA magnitude is obtained by means of the maximum absolute difference of even and odd heartbeat series averages computed at T-waves or ST-T complexes. This method performs well for different levels of TWA, noise, and phase shifts, but it is sensitive to the alignment of the T-waves. In this paper we propose a preprocessing stage for the MMA method to ensure an optimal alignment of such averages. The alignment is performed by means of a continuous time warping technique. Our assessment study demonstrates the improved performance of the proposed algorithm.

Physiological signal simulators are often used to conduct validation studies of commercially available devices such as oscillometric non-invasive blood pressure (NIBP) monitors. Numerous assessment studies have been conducted using simulators to validate commercial NIBP monitors. While there are several simulators commercially available to evaluate oscillometric NIBP devices, currently there are no simulators designed to validate invasive pressure signal devices. A statistical model and simulator for invasive cardiovascular pressure signals such as arterial blood pressure and intracranial pressure are described. The model incorporates the effects of respiration on pressure signals and can be used to generate synthetic signals with time and frequency domain characteristics matching any desired subject population. Additionally, the way that noise and artefacts typically present in real pressure signals should be modelled is described. The proposed statistical model is a useful tool for validation of algorithms designed to process or analyse biomedical pressure signals to estimate parameters of clinical interest such as the cardiac frequency, heart rate variability, respiratory frequency, and pulse pressure variation in the presence of noise. The model can be used to simulate signals in order to validate commercial devices that process and analyse invasive pressure signals.

OBJECTIVE: To describe and report the reliability of a portable, laptop-based, real-time, continuous physiologic data acquisition system (PDAS) that allows for synchronous recording of physiologic data, clinical events, and event markers at the bedside for physiologic research studies in the intensive care unit. DESIGN: Descriptive report of new research technology. SETTING: Adult and pediatric intensive care units in three tertiary care academic hospitals. PATIENTS: Sixty-four critically ill and injured patients were studied, including 34 adult (22 males and 12 females) and 30 pediatric (19 males and 11 females). INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Data transmission errors during bench and field testing were measured. The PDAS was used in three separate research studies, by multiple users, and for repeated recordings of the same set of signals at various intervals for different lengths of time.Both parametric (1 Hz) and waveform (125-500 Hz) signals were recorded and analyzed. Details of the PDAS components are explained and examples are given from the three experimental physiology-based protocols. Waveform data include electrocardiogram, respiration, systemic arterial pressure (invasive and noninvasive), oxygen saturation, central venous pressure, pulmonary arterial pressure, left and right atrial pressures, intracranial pressure, and regional cerebral blood flow. Bench and field testing of the PDAS demonstrated excellent reliability with 100% accuracy and no data transmission errors. The key feature of simultaneously capturing physiologic signal data and clinical events (e.g., changes in mechanical ventilation, drug administration, clinical condition) is emphasized. CONCLUSIONS: The PDAS provides a reliable tool to record physiologic signals and associated clinical events on a second-to-second basis and may serve as an important adjunctive research tool in designing and performing clinical physiologic studies in critical illness and injury.

We present a new analysis and visualization method for studying the functional relationship between the pulse morphology of pressure signals and time or signal metrics such as heart rate, pulse pressure, and means of pressure signals, such as arterial blood pressure and central venous pressure. The pulse morphology is known to contain potentially useful clinical information, but it is difficult to study in the time domain without the aid of a tool such as the method we present here. The primary components of the method are established signal processing techniques, nonparametric regression, and an automatic beat detection algorithm. Some of the insights that can be gained from this are demonstrated through the analysis of intracranial pressure signals acquired from patients with traumatic brain injuries. The analysis indicates the point of transition from low-pressure morphology consisting of three distinct peaks to a high-pressure morphology consisting of a single peak. In addition, we demonstrate how the analysis can reveal distinctions in the relationship between morphology and several signal metrics for different patients.

We analyzed intracranial pressure (ICP) signals during periods of acute intracranial hypertension (ICH) using the Lempel-Ziv (LZ) complexity measure. Our results indicate the LZ complexity of ICP decreases during periods of ICH. The mean LZ complexity before ICH was 0.20+/-0.04, while the mean LZ complexity during ICH was 0.16+/-0.03 (p0.05). The mean decrease of the LZ complexity values during the ICH episodes was 19.5 Additionally, we present preliminary evidence suggesting that periods of ICH may be detectable from non-invasive signals coupled with ICP, such as pulse oximetry (SpO2).

Body temperature is a classical diagnostic tool for a number of diseases. However, it is usually employed as a plain binary classification function (febrile or not febrile), and therefore its diagnostic power has not been fully developed. In this paper, we describe how body temperature regularity can be used for diagnosis. Our proposed methodology is based on obtaining accurate long-term temperature recordings at high sampling frequencies and analyzing the temperature signal using a regularity metric (approximate entropy). In this study, we assessed our methodology using temperature registers acquired from patients with multiple organ failure admitted to an intensive care unit. Our results indicate there is a correlation between the patient's condition and the regularity of the body temperature. This finding enabled us to design a classifier for two outcomes (survival or death) and test it on a dataset including 36 subjects. The classifier achieved an accuracy of 72

Ventricular extrasystoles (VE) are ectopic heartbeats involving irregularities in the heart rhythm. VEs arise in response to impulses generated in some part of the heart different from the sinoatrial node. These are caused by the premature discharge of a ventricular ectopic focus. VEs after myocardial infarction are associated with increased mortality. Screening of VEs is typically a manual and time consuming task that involves analysis of the heartbeat morphology, QRS duration, and variations of the RR intervals using long-term electrocardiograms. We describe a novel algorithm to perform automatic classification of VEs and report the results of our validation study. The proposed algorithm makes use of bounded clustering algorithms, morphology matching, and RR interval length to perform automatic VE classification without prior knowledge of the number of classes and heartbeat features. Additionally, the proposed algorithm does not need a training set.

Lempel-Ziv complexity (LZ) and derived LZ algorithms have been extensively used to solve information theoretic problems such as coding and lossless data compression. In recent years, LZ has been widely used in biomedical applications to estimate the complexity of discrete-time signals. Despite its popularity as a complexity measure for biosignal analysis, the question of LZ interpretability and its relationship to other signal parameters and to other metrics has not been previously addressed. We have carried out an investigation aimed at gaining a better understanding of the LZ complexity itself, especially regarding its interpretability as a biomedical signal analysis technique. Our results indicate that LZ is particularly useful as a scalar metric to estimate the bandwidth of random processes and the harmonic variability in quasi-periodic signals.

Extracellular microelectrode recordings (MER) often contain artifact from a variety of sources that confound traditional signal-processing techniques that require stationary signal segments. We designed an algorithm to locate the longest stationary segment of MER signals. In this paper we provide a description of the segmentation algorithm and its performance assessment. Simulation results demonstrate that the automatic segmentation algorithm we proposed is capable of accurately identifying the boundaries of the longest stationary segments in MER signals. In our simulation study the segmentation algorithm correctly identified the boundaries of the longest MER stationary segments in 99.5% of the cases.

Current indices used in the evaluation of antihypertensive treatment duration and homogeneity such as the trough-peak, smoothness index, and normalized smoothness index were designed to be applied to ambulatory blood pressure monitoring recordings from individual participants. Evaluation of antihypertensive treatment in populations is often carried out by calculating these individual indices for each of the participants and providing summarizing statistics about the population, such as the mean and median. We describe a new population vector index and graphical method for the statistical assessment of antihypertensive treatment reduction, duration, and homogeneity (RDH) from ambulatory blood pressure monitoring. The population (RDH) was specifically designed as a tool to evaluate and compare blood pressure coverage offered by antihypertensive drugs over 24 h in populations. The population RDH is a three-component vector index that incorporates information about the reduction, duration, and homogeneity of antihypertensive treatment, as well as their statistical significance over the 24 h period. In addition to defining the RDH index, in this paper we also demonstrate its usefulness and advantages as an index and graphical method for antihypertensive treatment duration and homogeneity assessment by using it to analyze two data sets.

Beat detection algorithms have many clinical applications including pulse oximetry, cardiac arrhythmia detection, and cardiac output monitoring. Most of these algorithms have been developed by medical device companies and are proprietary. Thus, researchers who wish to investigate pulse contour analysis must rely on manual annotations or develop their own algorithms. We designed an automatic detection algorithm for pressure signals that locates the first peak following each heart beat. This is called the percussion peak in intracranial pressure (ICP) signals and the systolic peak in arterial blood pressure (ABP) and pulse oximetry (SpO2) signals. The algorithm incorporates a filter bank with variable cutoff frequencies, spectral estimates of the heart rate, rank-order nonlinear filters, and decision logic. We prospectively measured the performance of the algorithm compared to expert annotations of ICP, ABP, and SpO2 signals acquired from pediatric intensive care unit patients. The algorithm achieved a sensitivity of 99.36% and positive predictivity of 98.43% on a dataset consisting of 42,539 beats.

We analyzed time series generated by 20 schizophrenic patients and 20 sex- and age-matched control subjects using three nonlinear methods of time series analysis as test statistics: central tendency measure (CTM) from the scatter plots of first differences of data, approximate entropy (ApEn), and Lempel-Ziv (LZ) complexity. We divided our data into a training set (10 patients and 10 control subjects) and a test set (10 patients and 10 control subjects). The training set was used for algorithm development and optimum threshold selection. Each method was assessed prospectively using the test dataset. We obtained 80% sensitivity and 90% specificity with LZ complexity, 90% sensitivity, and 60% specificity with ApEn, and 70% sensitivity and 70% specificity with CTM. Our results indicate that there exist differences in the ability to generate random time series between schizophrenic subjects and controls, as estimated by the CTM, ApEn, and LZ. This finding agrees with most previous results showing that schizophrenic patients are characterized by less complex neurobehavioral and neuropsychologic measurements.

Despite the exponential growth in heart rate variability (HRV) research, the reproducibility and reliability of HRV metrics continues to be debated. We estimated the reliability of 11 metrics calculated from 5 min records. We also compared the accuracy of the HRV metrics calculated from ECG records spanning 10 s to 10 min as compared with the metrics calculated from 5 min records. The mean heart rate was more reproducible and could be more accurately estimated from very short segments (1 min) than any of the other HRV metrics. HRV metrics that effectively highpass filter the R-R interval series were more reliable than the other metrics and could be more accurately estimated from very short segments. This indicates that most of the HRV is caused by drift and nonstationary effects. Metrics that are sensitive to low frequency components of HRV have poor repeatability and cannot be estimated accurately from short segments (10 min).

OBJECTIVE: To determine whether decomplexification of intracranial pressure dynamics occurs during periods of severe intracranial hypertension (intracranial pressure >25 mm Hg for >5 mins in the absence of external noxious stimuli) in pediatric patients with intracranial hypertension. DESIGN: Retrospective analysis of clinical case series over a 30-month period from April 2000 through January 2003. SETTING: Multidisciplinary 16-bed pediatric intensive care unit. PATIENTS: Eleven episodes of intracranial hypertension from seven patients requiring ventriculostomy catheter for intracranial pressure monitoring and/or cerebral spinal fluid drainage. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We measured changes in the intracranial pressure complexity, estimated by the approximate entropy (ApEn), as patients progressed from a state of normal intracranial pressure (25 mm Hg) to intracranial hypertension. We found the ApEn mean to be lower during the intracranial hypertension period than during the stable and recovering periods in all the 11 episodes (0.5158 +/- 0.0089, 0.3887 +/- 0.077, and 0.5096 +/- 0.0158, respectively, p .01). Both the mean reduction in ApEn from the state of normal intracranial pressure (stable region) to intracranial hypertension (-0.1271) and the increase in ApEn from the ICH region to the recovering region (0.1209) were determined to be statistically significant (p .01). CONCLUSIONS: Our results indicate that decreased complexity of intracranial pressure coincides with periods of intracranial hypertension in brain injury. This suggests that the complex regulatory mechanisms that govern intracranial pressure may be disrupted during acute periods of intracranial hypertension. This phenomenon of decomplexification of physiologic dynamics may have important clinical implications for intracranial pressure management.

We propose a new vector index for the statistical assessment of antihypertensive treatment duration and homogeneity from ambulatory blood pressure monitoring. We termed this approach for evaluating and comparing blood pressure coverage offered by antihypertensive drugs over 24 h as the reduction-duration-homogeneity index. The reduction-duration-homogeneity index is a three-component vector index that incorporates information about the reduction, duration, and homogeneity of antihypertensive treatment, as well as their statistical significance. The advantages of the reduction-duration-homogeneity index are demonstrated by several comparative examples.

We describe an algorithm to estimate the instantaneous power spectral density (PSD) of nonstationary signals. The algorithm is based on a dual Kalman filter that adaptively generates an estimate of the autoregressive model parameters at each time instant. The algorithm exhibits superior PSD tracking performance in nonstationary signals than classical nonparametric methodologies, and does not assume local stationarity of the data. Furthermore, it provides better time-frequency resolution, and is robust to model mismatches. We demonstrate its usefulness by a sample application involving PSD estimation of intracranial pressure signals (ICP) from patients with traumatic brain injury (TBI).

We studied changes in intracranial pressure (ICP) complexity, estimated by the approximate entropy (ApEn) of the ICP signal, as subjects progressed from a state of normal ICP ( 20-25 mmHg) to acutely elevated ICP (an ICP "spike" defined as ICP > 25 mmHg for or = 5 min). We hypothesized that the measures of intracranial pressure (ICP) complexity and irregularity would decrease during acute elevations in ICP. To test this hypothesis we studied ICP spikes in pediatric subjects with severe traumatic brain injury (TBI). We conclude that decreased complexity of ICP coincides with episodes of intracranial hypertension (ICH) in TBI. This suggests that the complex regulatory mechanisms that govern intracranial pressure are disrupted during acute rises in ICP. Furthermore, we carried out a series of experiments where ApEn was used to analyze synthetic signals of different characteristics with the objective of gaining a better understanding of ApEn itself, especially its interpretation in biomedical signal analysis.

We review the potential limitations of the two current methodologies for evaluating the duration of action of antihypertensive therapy: the smoothness index (SI) and the trough : peak ratio (TP). We propose a simple correction factor for the SI. The correction factor prevents the SI from reaching erroneous high values in situations in which the reduction in blood pressure (BP) is inadequate but very homogeneous. We refer to the corrected index as the SIn (normalized SI).

We designed a new methodology to estimate the pulse pressure variation index (deltaPP) in arterial blood pressure (ABP). The method uses automatic detection algorithms, kernel smoothing, and rank-order filters to continuously estimate deltaPP. The technique can be used to estimate deltaPP from ABP alone, eliminating the need for simultaneously acquiring airway pressure.

Currently, no reliable method exists to predict the onset of paroxysmal atrial fibrillation (PAF). We propose a predictor that includes an analysis of the R-R time series. The predictor uses three criteria: the number of premature atrial complexes (PAC) not followed by a regular R-R interval, runs of atrial bigeminy and trigeminy, and the length of any short run of paroxysmal atrial tachycardia. An increase in activity detected by any of these three criteria is an indication of an imminent episode of PAF. Using the Physionet database of the Computers in Cardiology 2001 Challenge, the predictor achieved a sensitivity of 89% and a specificity of 91

We describe a low cost portable Holter design that can be implemented with off-the-shelf components. The recorder is battery powered and includes a graphical display and keyboard. The recorder is capable of acquiring up to 48 hours of continuous electrocardiogram data at a sample rate of up to 250 Hz.