Weaning From Mechanical Ventilation: A Retrospective Analysis Leading to a Multimodal Perspective

Analysis of the Cardiorespiratory Pattern of Patients Undergoing Weaning Using Artificial Intelligence

The optimal extubating moment is still a challenge in clinical practice. Respiratory pattern variability analysis in patients assisted through mechanical ventilation to identify this optimal moment could contribute to this process. This work proposes the analysis of this variability using several time series obtained from the respiratory flow and electrocardiogram signals, applying techniques based on artificial intelligence. 154 patients undergoing the extubating process were classified in three groups: successful group, patients who failed during weaning process, and patients who after extubating failed before 48 hours and need to reintubated. Power Spectral Density and time-frequency domain analysis were applied, computing Discrete Wavelet Transform. A new Q index was proposed to determine the most relevant parameters and the best decomposition level to discriminate between groups. Forward selection and bidirectional techniques were implemented to reduce dimensionality. Linear Discriminant Analysis and Neural Networks methods were implemented to classify these patients. The best results in terms of accuracy were, 84.61 ± 3.1% for successful versus failure groups, 86.90 ± 1.0% for successful versus reintubated groups, and 91.62 ± 4.9% comparing the failure and reintubated groups. Parameters related to Q index and Neural Networks classification presented the best performance in the classification of these patients.

Biosignal-Based Digital Biomarkers for Prediction of Ventilator Weaning Success

We evaluated new features from biosignals comprising diverse physiological response information to predict the outcome of weaning from mechanical ventilation (MV). We enrolled 89 patients who were candidates for weaning from MV in the intensive care unit and collected continuous biosignal data: electrocardiogram (ECG), respiratory impedance, photoplethysmogram (PPG), arterial blood pressure, and ventilator parameters during a spontaneous breathing trial (SBT). We compared the collected biosignal data's variability between patients who successfully discontinued MV (n = 67) and patients who did not (n = 22). To evaluate the usefulness of the identified factors for predicting weaning success, we developed a machine learning model and evaluated its performance by bootstrapping. The following markers were different between the weaning success and failure groups: the ratio of standard deviations between the short-term and long-term heart rate variability in a Poincaré plot, sample entropy of ECG and PPG, α values of ECG, and respiratory impedance in the detrended fluctuation analysis. The area under the receiver operating characteristic curve of the model was 0.81 (95% confidence interval: 0.70-0.92). This combination of the biosignal data-based markers obtained during SBTs provides a promising tool to assist clinicians in determining the optimal extubation time.

Automatic sleep staging by cardiorespiratory signals: a systematic review

BACKGROUND

Because of problems with the recording and analysis of the EEG signal, automatic sleep staging using cardiorespiratory signals has been employed as an alternative. This study reports on certain critical points which hold considerable promise for the improvement of the results of the automatic sleep staging using cardiorespiratory signals.

METHODS

A systematic review.

RESULTS

The review and analysis of the literature in this area revealed four outstanding points: (1) the feature extraction epoch length, denoting that the standard 30-s segments of cardiorespiratory signals do not carry enough information for automatic sleep staging and that a 4.5-min length segment centering on each 30-s segment is proper for staging, (2) the time delay between the EEG signal extracted from the central nervous system activity and the cardiorespiratory signals extracted from the autonomic nervous system activity should be considered in the automatic sleep staging using cardiorespiratory signals, (3) the information in the morphology of ECG signals can contribute to the improvement of sleep staging, and (4) applying convolutional neural network (CNN) and long short-term memory network (LSTM) deep structures simultaneously to a large PSG recording database can lead to more reliable automatic sleep staging results.

CONCLUSIONS

Considering the above-mentioned points simultaneously can improve automatic sleep staging by cardiorespiratory signals. It is hoped that by considering the points, staging sleep automatically using cardiorespiratory signals, which does not have problems with the recording and analysis of EEG signals, yields results acceptably close to the results of automatic sleep staging by EEG signals.

Sleep-wake classification via quantifying heart rate variability by convolutional neural network

OBJECTIVE

Fluctuations in heart rate are intimately related to changes in the physiological state of the organism. We exploit this relationship by classifying a human participant's wake/sleep status using his instantaneous heart rate (IHR) series.

APPROACH

We use a convolutional neural network (CNN) to build features from the IHR series extracted from a whole-night electrocardiogram (ECG) and predict every 30 s whether the participant is awake or asleep. Our training database consists of 56 normal participants, and we consider three different databases for validation; one is private, and two are public with different races and apnea severities.

MAIN RESULTS

On our private database of 27 participants, our accuracy, sensitivity, specificity, and [Formula: see text] values for predicting the wake stage are [Formula: see text], 52.4%, 89.4%, and 0.83, respectively. Validation performance is similar on our two public databases. When we use the photoplethysmography instead of the ECG to obtain the IHR series, the performance is also comparable. A robustness check is carried out to confirm the obtained performance statistics.

SIGNIFICANCE

This result advocates for an effective and scalable method for recognizing changes in physiological state using non-invasive heart rate monitoring. The CNN model adaptively quantifies IHR fluctuation as well as its location in time and is suitable for differentiating between the wake and sleep stages.

Nonlinear dynamic analysis of the cardiorespiratory system in patients undergoing the weaning process

In this work, the cardiorespiratory pattern of patients undergoing extubation process is studied. First, the respiratory and cardiac signals were resampled, next the Symbolic Dynamics (SD) technique was implemented, followed of a dimensionality reduction applying Forward Selection (FS) and Moving Window with Variance Analysis (MWVA) methods. Finally, the Linear Discriminant Analysis (LDA) and Support Vector Machines (SVM) classifiers were used. The study analyzed 153 patients undergoing weaning process, classified into 3 groups: Successful Group (SG: 94 patients), Failed Group (FG: 39 patients), and patients who had been successful during the extubation and had to be reintubated before 48 hours, Reintubated Group (RG: 21 patients). According to the results, the best classification present an accuracy higher than 88.98 ± 0.013% in all proposed combinations.

Analysis of blood pressure signal in patients with different ventricular ejection fraction using linear and non-linear methods

Changes in the left ventricle function produce alternans in the hemodynamic and electric behavior of the cardiovascular system. A total of 49 cardiomyopathy patients have been studied based on the blood pressure signal (BP), and were classified according to the left ventricular ejection fraction (LVEF) in low risk (LR: LVEF>35%, 17 patients) and high risk (HR: LVEF≤35, 32 patients) groups. We propose to characterize these patients using a linear and a nonlinear methods, based on the spectral estimation and the recurrence plot, respectively. From BP signal, we extracted each systolic time interval (STI), upward systolic slope (BPsl), and the difference between systolic and diastolic BP, defined as pulse pressure (PP). After, the best subset of parameters were obtained through the sequential feature selection (SFS) method. According to the results, the best classification was obtained using a combination of linear and nonlinear features from STI and PP parameters. For STI, the best combination was obtained considering the frequency peak and the diagonal structures of RP, with an area under the curve (AUC) of 79%. The same results were obtained when comparing PP values. Consequently, the use of combined linear and nonlinear parameters could improve the risk stratification of cardiomyopathy patients.

Kölliker-Fuse nuclei regulate respiratory rhythm variability via a gain-control mechanism

Respiration varies from breath to breath. On the millisecond timescale of spiking, neuronal circuits exhibit variability due to the stochastic properties of ion channels and synapses. Does this fast, microscopic source of variability contribute to the slower, macroscopic variability of the respiratory period? To address this question, we modeled a stochastic oscillator with forcing; then, we tested its predictions experimentally for the respiratory rhythm generated by the in situ perfused preparation during vagal nerve stimulation (VNS). Our simulations identified a relationship among the gain of the input, entrainment strength, and rhythm variability. Specifically, at high gain, the periodic input entrained the oscillator and reduced variability, whereas at low gain, the noise interacted with the input, causing events known as "phase slips", which increased variability on a slow timescale. Experimentally, the in situ preparation behaved like the low-gain model: VNS entrained respiration but exhibited phase slips that increased rhythm variability. Next, we used bilateral muscimol microinjections in discrete respiratory compartments to identify areas involved in VNS gain control. Suppression of activity in the nucleus tractus solitarii occluded both entrainment and amplification of rhythm variability by VNS, confirming that these effects were due to the activation of the Hering-Breuer reflex. Suppressing activity of the Kölliker-Fuse nuclei (KFn) enhanced entrainment and reduced rhythm variability during VNS, consistent with the predictions of the high-gain model. Together, the model and experiments suggest that the KFn regulates respiratory rhythm variability via a gain control mechanism.

Classification of patients undergoing weaning from mechanical ventilation using the coherence between heart rate variability and respiratory flow signal

Weaning from mechanical ventilation is still one of the most challenging problems in intensive care. Unnecessary delays in discontinuation and weaning trials that are undertaken too early are both undesirable. This study investigated the contribution of spectral signals of heart rate variability (HRV) and respiratory flow, and their coherence to classifying patients on weaning process from mechanical ventilation. A total of 121 candidates for weaning, undergoing spontaneous breathing tests, were analyzed: 73 were successfully weaned (GSucc), 33 failed to maintain spontaneous breathing so were reconnected (GFail), and 15 were extubated after the test but reintubated within 48 h (GRein). The power spectral density and magnitude squared coherence (MSC) of HRV and respiratory flow signals were estimated. Dimensionality reduction was performed using principal component analysis (PCA) and sequential floating feature selection. The patients were classified using a fuzzy K-nearest neighbour method. PCA of the MSC gave the best classification with the highest accuracy of 92% classifying GSucc versus GFail patients, and 86% classifying GSucc versus GRein patients. PCA of the respiratory flow signal gave the best classification between GFail and GRein patients (79% accuracy). These classifiers showed a good balance between sensitivity and specificity. Besides, the spectral coherence between HRV and the respiratory flow signal, in patients on weaning trial process, can contribute to the extubation decision.

Power index of the inspiratory flow signal as a predictor of weaning in intensive care units

Disconnection from mechanical ventilation, called the weaning process, is an additional difficulty in the management of patients in intensive care units (ICU). Unnecessary delays in the discontinuation process and a weaning trial that is undertaken too early are undesirable. In this study, we propose an extubation index based on the power of the respiratory flow signal (Pi). A total of 132 patients on weaning trials were studied: 94 patients with successful trials (group S) and 38 patients who failed to maintain spontaneous breathing and were reconnected (group F). The respiratory flow signals were processed considering the following three stages: a) zero crossing detection of the inspiratory phase, b) inflection point detection of the flow curve during the inspiratory phase, and c) calculation of the signal power on the time instant indicated by the inflection point. The zero crossing detection was performed using an algorithm based on thresholds. The inflection points were marked considering the zero crossing of the second derivative. Finally, the inspiratory power was calculated from the energy contained over the finite time interval (between the instant of zero crossing and the inflection point). The performance of this parameter was evaluated using the following classifiers: logistic regression, linear discriminant analysis, the classification and regression tree, Naive Bayes, and the support vector machine. The best results were obtained using the Bayesian classifier, which had an accuracy, sensitivity and specificity of 87%, 90% and 81% respectively.

Evaluating physiological dynamics via synchrosqueezing: prediction of ventilator weaning

Oscillatory phenomena abound in many types of signals. Identifying the individual oscillatory components that constitute an observed biological signal leads to profound understanding about the biological system. The instantaneous frequency (IF), the amplitude modulation (AM), and their temporal variability are widely used to describe these oscillatory phenomena. In addition, the shape of the oscillatory pattern, repeated in time for an oscillatory component, is also an important characteristic that can be parametrized appropriately. These parameters can be viewed as phenomenological surrogates for the hidden dynamics of the biological system. To estimate jointly the IF, AM, and shape, this paper applies a novel and robust time-frequency analysis tool, referred to as the synchrosqueezing transform (SST). The usefulness of the model and SST are shown directly in predicting the clinical outcome of ventilator weaning. Compared with traditional respiration parameters, the breath-to-breath variability has been reported to be a better predictor of the outcome of the weaning procedure. So far, however, all these indices normally require at least 20 min of data acquisition to ensure predictive power. Moreover, the robustness of these indices to the inevitable noise is rarely discussed. We find that based on the proposed model, SST and only 3 min of respiration data, the ROC area under curve of the prediction accuracy is 0.76. The high predictive power that is achieved in the weaning problem, despite a shorter evaluation period, and the stability to noise suggest that other similar kinds of signal may likewise benefit from the proposed model and SST.

Temperature multiscale entropy analysis: a promising marker for early prediction of mortality in septic patients

A few studies estimating temperature complexity have found decreased Shannon entropy, during severe stress. In this study, we measured both Shannon and Tsallis entropy of temperature signals in a cohort of critically ill patients and compared these measures with the sequential organ failure assessment (SOFA) score, in terms of intensive care unit (ICU) mortality. Skin temperature was recorded in 21 mechanically ventilated patients, who developed sepsis and septic shock during the first 24 h of an ICU-acquired infection. Shannon and Tsallis entropies were calculated in wavelet-based decompositions of the temperature signal. Statistically significant differences of entropy features were tested between survivors and non-survivors and classification models were built, for predicting final outcome. Significantly reduced Tsallis and Shannon entropies were found in non-survivors (seven patients, 33%) as compared to survivors. Wavelet measurements of both entropy metrics were found to predict ICU mortality better than SOFA, according to a combination of area under the curve, sensitivity and specificity values. Both entropies exhibited similar prognostic accuracy. Combination of SOFA and entropy presented improved the outcome of univariate models. We suggest that reduced wavelet Shannon and Tsallis entropies of temperature signals may complement SOFA in mortality prediction, during the first 24 h of an ICU-acquired infection.

Characterization of the respiratory pattern variability of patients with different pressure support levels

One of the most challenging problems in intensive care is still the process of discontinuing mechanical ventilation, called weaning process. Both an unnecessary delay in the discontinuation process and a weaning trial that is undertaken too early are undesirable. In this study, we analyzed respiratory pattern variability using the respiratory volume signal of patients submitted to two different levels of pressure support ventilation (PSV), prior to withdrawal of the mechanical ventilation. In order to characterize the respiratory pattern, we analyzed the following time series: inspiratory time, expiratory time, breath duration, tidal volume, fractional inspiratory time, mean inspiratory flow and rapid shallow breathing. Several autoregressive modeling techniques were considered: autoregressive models (AR), autoregressive moving average models (ARMA), and autoregressive models with exogenous input (ARX). The following classification methods were used: logistic regression (LR), linear discriminant analysis (LDA) and support vector machines (SVM). 20 patients on weaning trials from mechanical ventilation were analyzed. The patients, submitted to two different levels of PSV, were classified as low PSV and high PSV. The variability of the respiratory patterns of these patients were analyzed. The most relevant parameters were extracted using the classifiers methods. The best results were obtained with the interquartile range and the final prediction errors of AR, ARMA and ARX models. An accuracy of 95% (93% sensitivity and 90% specificity) was obtained when the interquartile range of the expiratory time and the breath duration time series were used a LDA model. All classifiers showed a good compromise between sensitivity and specificity.

Forecasting respiratory collapse: theory and practice for averting life-threatening infant apneas

Apnea of prematurity is a common disorder of respiratory control among preterm infants, with potentially serious adverse consequences on infant development. We review the capability for automatically assessing apnea risk and predicting apnea episodes from multimodal physiological measurements, and for using this knowledge to provide timely therapeutic intervention. We also review other, similar clinical domains of respiratory distress assessment and prediction in the hope of gaining useful insights. We propose an algorithmic framework for constructing discriminative feature vectors from physiological measurements, and for building robust and effective statistical models for apnea assessment and prediction.

Analysis of roots in ARMA model for the classification of patients on weaning trials

One objective of mechanical ventilation is the recovery of spontaneous breathing as soon as possible. Remove the mechanical ventilation is sometimes more difficult that maintain it. This paper proposes the study of respiratory flow signal of patients on weaning trials process by autoregressive moving average model (ARMA), through the location of poles and zeros of the model. A total of 151 patients under extubation process (T-tube test) were analyzed: 91 patients with successful weaning (GS), 39 patients that failed to maintain spontaneous breathing and were reconnected (GF), and 21 patients extubated after the test but before 48 hours were reintubated (GR). The optimal model was obtained with order 8, and statistical significant differences were obtained considering the values of angles of the first four poles and the first zero. The best classification was obtained between GF and GR, with an accuracy of 75.3% on the mean value of the angle of the first pole.

Automatic diagnosis of ADHD based on multichannel nonlinear analysis of actimetry registries

Attention-Deficit Hyperactivity Disorder (ADHD) is the most common mental health problem in childhood and adolescence. It is commonly diagnosed by means of subjective methods which tend to overestimate the severity of the pathology. A number of objective methods also exist, but they are either expensive or time-consuming. Some recent proposals based on nonlinear processing of activity registries have deserved special attention. Since they rely on actigraphy measurements, they are both inexpensive and non-invasive. Among these methods, those shown to have higher reliability are based on single-channel complexity assessment of the activity patterns. This way, potentially useful information related to the interaction between the different channels is discarded. In this paper we propose a new methodology for ADHD diagnosis based on joint complexity assessment of multichannel activity registries. Results on real data show that the proposed method constitute a useful diagnostic aid tool reaching 87:10% sensitivity and 84.38% specificity. The combination of ADHD indicators extracted with the proposed method with single-channel complexity-based indices previously proposed lead to sensitivity and specifity values above 90%.

Isoflurane and ketamine anesthesia have different effects on ventilatory pattern variability in rats

We hypothesize that isoflurane and ketamine impact ventilatory pattern variability (VPV) differently. Adult Sprague-Dawley rats were recorded in a whole-body plethysmograph before, during and after deep anesthesia. VPV was quantified from 60-s epochs using a complementary set of analytic techniques that included constructing surrogate data sets that preserved the linear structure but disrupted nonlinear deterministic properties of the original data. Even though isoflurane decreased and ketamine increased respiratory rate, VPV as quantified by the coefficient of variation decreased for both anesthetics. Further, mutual information increased and sample entropy decreased and the nonlinear complexity index (NLCI) increased during anesthesia despite qualitative differences in the shape and period of the waveform. Surprisingly mutual information and sample entropy did not change in the surrogate sets constructed from isoflurane data, but in those constructed from ketamine data, mutual information increased and sample entropy decreased significantly in the surrogate segments constructed from anesthetized relative to unanesthetized epochs. These data suggest that separate mechanisms modulate linear and nonlinear variability of breathing.

Recurrence quantification analysis of heart rate variability and respiratory flow series in patients on weaning trials

Autonomic nervous system regulates the behavior of cardiac and respiratory systems. Its assessment during the ventilator weaning can provide information about physio-pathological imbalances. This work proposes a non linear analysis of the complexity of the heart rate variability (HRV) and breathing duration (T(Tot)) applying recurrence plot (RP) and their interaction joint recurrence plot (JRP). A total of 131 patients on weaning trials from mechanical ventilation were analyzed: 92 patients with successful weaning (group S) and 39 patients that failed to maintain spontaneous breathing (group F). The results show that parameters as determinism (DET), average diagonal line length (L), and entropy (ENTR), are statistically significant with RP for T(Tot) series, but not with HRV. When comparing the groups with JRP, all parameters have been relevant. In all cases, mean values of recurrence quantification analysis are higher in the group S than in the group F. The main differences between groups were found on the diagonal and vertical structures of the joint recurrence plot.

Nonlinear analysis of actigraphic signals for the assessment of the attention-deficit/hyperactivity disorder (ADHD)

Attention-deficit/hyperactivity disorder (ADHD) is the most common neurobehavioral disorder in children and adolescents; however, its etiology is still unknown, which hinders the existence of reliable, fast and inexpensive standard diagnostic methods. In this paper, we propose a novel methodology for automatic diagnosis of the combined type of ADHD based on nonlinear signal processing of 24h-long actigraphic registries. Since it relies on actigraphy measurements, it constitutes an inexpensive and non-invasive objective diagnostic method. Our results on real data reach 96.77% sensitivity and 84.38% specificity by means of multidimensional classifiers driven by combined features from different time intervals. Our analysis also reveals that, if features from a single time interval are used, the whole 24-h interval is the only one that yields classification figures with practical diagnostic capabilities. Overall, our figures overcome those obtained by actigraphy-based methods reported and are comparable with others based on more expensive (and not so convenient) adquisition methods.

Comparisons of predictive performance of breathing pattern variability measured during T-piece, automatic tube compensation, and pressure support ventilation for weaning intensive care unit patients from mechanical ventilation

OBJECTIVE

To investigate the influence of different ventilatory supports on the predictive performance of breathing pattern variability for extubation outcomes in intensive care unit patients.

DESIGN AND SETTING

A prospective measurement of retrospectively analyzed breathing pattern variability in a medical center.

PATIENTS

Sixty-eight consecutive and ready-for-weaning patients were divided into success (n=45) and failure (n=23) groups based on their extubation outcomes.

MEASUREMENTS

Breath-to-breath analyses of peak inspiratory flow, total breath duration, tidal volume, and rapid shallow breathing index were performed for three 30-min periods while patients randomly received T-piece, 100% inspiratory automatic tube compensation with 5 cm H2O positive end-expiratory pressure, and 5 cm H2O pressure support ventilation with 5 cm H2O positive end-expiratory pressure trials. Coefficient of variations and data dispersion (standard descriptor values SD1 and SD2 of the Poincaré plot) were analyzed to serve as breathing pattern variability indices.

MAIN RESULTS

Under all three trials, breathing pattern variability in extubation failure patients was smaller than in extubation success patients. Compared to the T-piece trial, 100% inspiratory automatic tube compensation with 5 cm H2O positive end-expiratory pressure and 5 cm H2O pressure support ventilation with 5 cm H2O positive end-expiratory pressure decreased the ability of certain breathing pattern variability indices to discriminate extubation success from extubation failure. The areas under the receiver operating characteristic curve of these breathing pattern variability indices were: T-piece (0.73-0.87)>100% inspiratory automatic tube compensation with 5 cm H2O positive end-expiratory pressure (0.60-0.79)>5 cm H2O pressure support ventilation with 5 cm H2O positive end-expiratory pressure (0.53-0.76). Analysis of the classification and regression tree indicated that during the T-piece trial, a SD1 of peak inspiratory flow>3.36 L/min defined a group including all extubation success patients. Conversely, the combination of a SD1 of peak inspiratory flow ≤3.36 L/min and a coefficient of variations of rapid shallow breathing index ≤0.23 defined a group of all extubation failure patients. The decision strategies using SD1 of peak inspiratory flow and coefficient of variations of rapid shallow breathing index measured during 100% inspiratory automatic tube compensation with 5 cm H2O positive end-expiratory pressure and 5 cm H2O pressure support ventilation with 5 cm H2O positive end-expiratory pressure trials achieved a less clear separation of extubation failure from extubation success.

CONCLUSIONS

Since 100% inspiratory automatic tube compensation with 5 cm H2O positive end-expiratory pressure and 5 cm H2O pressure support ventilation with 5 cm H2O positive end-expiratory pressure reduce the predictive performance of breathing pattern variability, breathing pattern variability measurement during the T-piece trial is the best choice for predicting extubation outcome in intensive care unit patients patients.

Lung and brainstem cytokine levels are associated with breathing pattern changes in a rodent model of acute lung injury

Acute lung injury evokes a pulmonary inflammatory response and changes in the breathing pattern. The inflammatory response has a centrally mediated component which depends on the vagi. We hypothesize that the central inflammatory response, complimentary to the pulmonary inflammatory response, is expressed in the nuclei tractus solitarii (nTS) and that the expression of cytokines in the nTS is associated with breathing pattern changes. Adult, male Sprague-Dawley rats (n=12) received intratracheal instillation of either bleomycin (3units in 120μl of saline) or saline (120μl). Respiratory pattern changed by 24h. At 48h, bronchoalveolar lavage fluid and lung tissue had increased IL-1β and TNF-α levels, but not IL-6. No changes in these cytokines were noted in serum. Immunocytochemical analysis of the brainstem indicated increased expression of IL-1β in the nTS commissural subnucleus that was localized to neurons. We conclude that breathing pattern changes in acute lung injury were associated with increased levels of IL-1β in brainstem areas which integrate cardio-respiratory sensory input.

Study of multiparameter respiratory pattern complexity in surgical critically ill patients during weaning trials

Background

Separation from mechanical ventilation is a difficult task, whereas conventional predictive indices have not been proven accurate enough, so far. A few studies have explored changes of breathing pattern variability for weaning outcome prediction, with conflicting results. In this study, we tried to assess respiratory complexity during weaning trials, using different non-linear methods derived from theory of complex systems, in a cohort of surgical critically ill patients.

Results

Thirty two patients were enrolled in the study. There were 22 who passed and 10 who failed a weaning trial. Tidal volume and mean inspiratory flow were analyzed for 10 minutes during two phases: 1. pressure support (PS) ventilation (15-20 cm H₂O) and 2. weaning trials with PS: 5 cm H₂O. Sample entropy (SampEn), detrended fluctuation analysis (DFA) exponent, fractal dimension (FD) and largest lyapunov exponents (LLE) of the two respiratory parameters were computed in all patients and during the two phases of PS. Weaning failure patients exhibited significantly decreased respiratory pattern complexity, reflected in reduced sample entropy and lyapunov exponents and increased DFA exponents of respiratory flow time series, compared to weaning success subjects (p < 0.001). In addition, their changes were opposite between the two phases of the weaning trials. A new model including rapid shallow breathing index (RSBI), its product with airway occlusion pressure at 0.1 sec (P_0.1), SampEn and LLE predicted better weaning outcome compared with RSBI, P_0.1 and RSBI* P_0.1 (conventional model, R² = 0.874 vs 0.643, p < 0.001). Areas under the curve were 0.916 vs 0.831, respectively (p < 0.05).

Conclusions

We suggest that complexity analysis of respiratory signals can assess inherent breathing pattern dynamics and has increased prognostic impact upon weaning outcome in surgical patients.

Consultation system for determining the patients' readiness for weaning from long-term mechanical ventilation

Determining the readiness of the patient for weaning from long-term mechanical ventilation with high degree of accuracy is a prerequisite for successful weaning attempt. A computer system for the determination of the moment for beginning the weaning process is proposed and realized as a program package WeaningMV. It is designed on the basis of four classification methods: stepwise discriminant analysis, stepwise logistic regression; intuitionistic fuzzy Voronoi diagrams and non-pulmonary weaning index, applied to 17 features (variables). The system is designed and preliminary tested on 151 patients with features in two classes "not ready for weaning" and "ready for weaning" and it has a high degree for sensitivity (0.99) and specificity (0.89). After the application of the system in the everyday physician's practice a set of 72 patients has been examined. For that set the sensitivity is 0.92 and the specificity 0.88. The system consults the medical team on the basis of objective processing of the information contained in the data and is suitable for use in the clinical practice as well as in the training of the respective specialists.

Changes of heart and respiratory rate dynamics during weaning from mechanical ventilation: a study of physiologic complexity in surgical critically ill patients

PURPOSE

The aim of the study was to investigate heart rate (HR) and respiratory rate (RR) complexity in patients with weaning failure or success, using both linear and nonlinear techniques.

MATERIALS AND METHODS

Forty-two surgical patients were enrolled in the study. There were 24 who passed and 18 who failed a weaning trial. Signals were analyzed for 10 minutes during 2 phases: (1) pressure support (PS) ventilation (15-20 cm H(2)O) and (2) weaning trials with PS (5 cm H(2)O). Low- and high-frequency (LF, HF) components of HR signals, HR multiscale entropy (MSE), RR sample entropy, cross-sample entropy between cardiorespiratory signals, Poincaré plots, and α1 exponent were computed in all patients and during the 2 phases of PS.

RESULTS

Weaning failure patients exhibited significantly decreased RR sample entropy, LF, HF, and α1 exponent, compared with weaning success subjects (P < .001). Their changes were opposite between the 2 phases, except for MSE that increased between and within groups (P < .001). A new model including rapid shallow breathing index (RSBI), α1 exponent, RR, and cross-sample entropies predicted better weaning outcome compared with RSBI, airway occlusion pressure at 0.1 second (P(0.1)), and RSBI × P(0.1) (conventional model, R(2) = 0.887 vs 0.463; P < .001). Areas under the curve were 0.92 vs 0.86, respectively (P < .005).

CONCLUSIONS

We suggest that nonlinear analysis of cardiorespiratory dynamics has increased prognostic impact upon weaning outcome in surgical patients.

Symbolic dynamic analysis of relations between cardiac and breathing cycles in patients on weaning trials

Traditional time-domain techniques of data analysis are often not sufficient to characterize the complex dynamics of the cardiorespiratory interdependencies during the weaning trials. In this paper, the interactions between the heart rate (HR) and the breathing rate (BR) were studied using joint symbolic dynamic analysis. A total of 133 patients on weaning trials from mechanical ventilation were analyzed: 94 patients with successful weaning (group S) and 39 patients that failed to maintain spontaneous breathing (group F). The word distribution matrix enabled a coarse-grained quantitative assessment of short-term nonlinear analysis of the cardiorespiratory interactions. The histogram of the occurrence probability of the cardiorespiratory words presented a higher homogeneity in group F than in group S, measured with a higher number of forbidden words in group S as well as a higher number of words whose probability of occurrence is higher than a probability threshold in group S. The discriminant analysis revealed the best results when applying symbolic dynamic variables. Therefore, we hypothesize that joint symbolic dynamic analysis provides enhanced information about different interactions between HR and BR, when comparing patients with successful weaning and patients that failed to maintain spontaneous breathing in the weaning procedure.

Optimal parameters study for sample entropy-based atrial fibrillation organization analysis

Sample entropy (SampEn) is a nonlinear regularity index that requires the a priori selection of three parameters: the length of the sequences to be compared, m, the patterns similarity tolerance, r, and the number of samples under analysis, N. Appropriate values for m, r and N have been recommended and widely used in the literature for the application of SampEn to some physiological time series, such as heart rate, hormonal data, etc. However, no guidelines exist for the selection of that values in other cases. Therefore, an optimal parameters study should be required for the application of SampEn to not previously analyzed biomedical signals. In the present work, a thorough analysis on the optimal values for m, r and N is presented within the context of atrial fibrillation (AF) organization estimation, computed from surface electrocardiogram recordings. Recently, the evaluation of AF organization through SampEn, has revealed clinically useful information that could be used for a better treatment of this arrhythmia. The present study analyzed optimal SampEn parameter values within two different scenarios of AF organization estimation, such as the prediction of paroxysmal AF termination and the electrical cardioversion outcome in persistent AF. As a result, interesting recommendations about the selection of m, r and N, together with the relationship between N and the sampling rate (f(s)) were obtained. More precisely, (i) the proportion between N and f(s) should be higher than 1s and f(s)>or=256 Hz, (ii) overlapping between adjacent N-length windows does not improve AF organization estimation with respect to the analysis of non-overlapping windows, and (iii) values of m and r maximizing successful classification for the analyzed AF databases should be considered within a range wider than the proposed in the literature for heart rate analysis, i.e. m=1 and m=2 and r between 0.1 and 0.25 times the standard deviation of the data.

Multi-parameter analysis of ECG and respiratory flow signals to identify success of patients on weaning trials

Statistical analysis, power spectral density, and Lempel Ziv complexity, are used in a multi-parameter approach to analyze four temporal series obtained from the Electrocardiographic and Respiratory Flow signals of 126 patients on weaning trials. In which, 88 patients belong to successful group (SG), and 38 patients belong to failure group (FG), i.e. failed to maintain spontaneous breathing during trial. It was found that mean values of cardiac inter-beat and breath durations give higher values for SG than for FG; Kurtosis coefficient of the spectrum of the rapid shallow breathing index is higher for FG; also Lempel Ziv complexity mean values associated with the respiratory flow signal are bigger for FG. Patients were then classified with a pattern recognition neural network, obtaining 80% of correct classifications (81.6% for FG and 79.5% for SG).

Spectral analysis of the RR series and the respiratory flow signal on patients in weaning process

A considerable number of patients in weaning process have problems to keep spontaneous breathing during the trial and after it. This study proposes to extract characteristic parameters of the RR series and respiratory flow signal according to the patients' condition in weaning test. Three groups of patients have been considered: 93 patients with successful trials (group S), 40 patients that failed to maintain spontaneous breathing (group F), and 21 patients who had successful weaning trials, but that had to be reintubated before 48 hours (group R). The characterization was performed using spectral analysis of the signals, through the power spectral density, cross power spectral density and Coherence method. The parameters were extracted on the three frequency bands (VLF, LF and HF), and the principal statistical differences between groups were obtained in bands of VLF and HF. The results show an accuracy of 76.9% in the classification of the groups S and F.

Biosignal analysis techniques for weaning outcome assessment

Discontinuation of mechanical ventilation in critically ill patients is a challenging task and involves a careful weighting of the benefits of early extubation and the risks of premature spontaneous breathing trial. Recently, apart from studying different physiological variables by means of descriptive statistical tests, breathing pattern variability analysis has been performed for the assessment of weaning readiness. A limited number of clinical studies implementing different weaning protocols in heterogeneous groups of patients and using a variable set of signal processing techniques have appeared in the critical care literature, with varying results. The purpose of this review article is 3-fold: (1) to describe the different signal processing techniques being implemented for the assessment of weaning readiness, (2) to provide insight into the pathophysiological mechanisms that may govern breath-to-breath variability/complexity in health and disease, and (3) to present results from the critical care literature derived from the application of biosignal analysis tools for the identification of possible weaning indices.

Time-frequency analysis of cardiac and respiratory parameters for the prediction of ventilator weaning

Mechanical ventilators are used to provide life support in patients with respiratory failure. Assessing autonomic control during the ventilator weaning provides information about physiopathological imbalances. Autonomic parameters can be derived and used to predict success in discontinuing from the mechanical support. Time-frequency analysis is used to derive cardiac and respiratory parameters, as well as their evolution in time, during ventilator weaning in 130 patients. Statistically significant differences have been observed in autonomic parameters between patients who are considered ready for spontaneous breathing and patients who are not. A classification based on respiratory frequency, heart rate and heart rate variability spectral components has been proposed and has been able to correctly classify more than 80% of the cases.