Non-invasive classification of severe sepsis and systemic inflammatory response syndrome using a nonlinear support vector machine: a preliminary study

Enhancing sepsis management through machine learning techniques: A review

Sepsis is a major public health problem and a leading cause of death in the world, where delay in the beginning of treatment, along with clinical guidelines non-adherence have been proved to be associated with higher mortality. Machine Learning is increasingly being adopted in developing innovative Clinical Decision Support Systems in many areas of medicine, showing a great potential for automatic prediction of diverse patient conditions, as well as assistance in clinical decision making. In this context, this work conducts a narrative review to provide an overview of how specific Machine Learning techniques can be used to improve sepsis management, discussing the main tasks addressed, the most popular methods and techniques, as well as the obtained results, in terms of both intelligent system accuracy and clinical outcomes improvement.

Enhancing sepsis management through machine learning techniques: A review

Sepsis is a major public health problem and a leading cause of death in the world, where delay in the beginning of treatment, along with clinical guidelines non-adherence have been proved to be associated with higher mortality. Machine Learning is increasingly being adopted in developing innovative Clinical Decision Support Systems in many areas of medicine, showing a great potential for automatic prediction of diverse patient conditions, as well as assistance in clinical decision making. In this context, this work conducts a narrative review to provide an overview of how specific Machine Learning techniques can be used to improve sepsis management, discussing the main tasks addressed, the most popular methods and techniques, as well as the obtained results, in terms of both intelligent system accuracy and clinical outcomes improvement.

Heart Rate Variability as a Biomarker of Neurocardiogenic Injury After Subarachnoid Hemorrhage

BACKGROUND

The objective of this study was to examine whether heart rate variability (HRV) measures can be used to detect neurocardiogenic injury (NCI).

METHODS

Three hundred and twenty-six consecutive admissions with aneurysmal subarachnoid hemorrhage (SAH) met criteria for the study. Of 326 subjects, 56 (17.2%) developed NCI which we defined by wall motion abnormality with ventricular dysfunction on transthoracic echocardiogram or cardiac troponin-I > 0.3 ng/mL without electrocardiogram evidence of coronary artery insufficiency. HRV measures (in time and frequency domains, as well as nonlinear technique of detrended fluctuation analysis) were calculated over the first 48 h. We applied longitudinal multilevel linear regression to characterize the relationship of HRV measures with NCI and examine between-group differences at baseline and over time.

RESULTS

There was decreased vagal activity in NCI subjects with a between-group difference in low/high frequency ratio (β 3.42, SE 0.92, p = 0.0002), with sympathovagal balance in favor of sympathetic nervous activity. All time-domain measures were decreased in SAH subjects with NCI. An ensemble machine learning approach translated these measures into a classification tool that demonstrated good discrimination using the area under the receiver operating characteristic curve (AUROC 0.82), the area under precision recall curve (AUPRC 0.75), and a correct classification rate of 0.81.

CONCLUSIONS

HRV measures are significantly associated with our label of NCI and a machine learning approach using features derived from HRV measures can classify SAH patients that develop NCI.

Machine learning for prediction of septic shock at initial triage in emergency department

BACKGROUND

We hypothesized utilizing machine learning (ML) algorithms for screening septic shock in ED would provide better accuracy than qSOFA or MEWS.

METHODS

The study population was adult (≥20 years) patients visiting ED for suspected infection. Target event was septic shock within 24 h after arrival. Demographics, vital signs, level of consciousness, chief complaints (CC) and initial blood test results were used as predictors. CC were embedded into 16-dimensional vector space using singular value decomposition. Six base learners including support vector machine, gradient-boosting machine, random forest, multivariate adaptive regression splines and least absolute shrinkage and selection operator and ridge regression and their ensembles were tested. We also trained and tested MLP networks with various setting.

RESULTS

A total of 49,560 patients were included and 4817 (9.7%) had septic shock within 24 h. All ML classifiers significantly outperformed qSOFA score, MEWS and their age-sex adjusted versions with their AUROC ranging from 0.883 to 0.929. The ensembles of the base classifiers showed the best performance and addition of CC embedding was associated with statistically significant increases in performance.

CONCLUSIONS

ML classifiers significantly outperforms clinical scores in screening septic shock at ED triage.

Critical Transitions in Intensive Care Units: A Sepsis Case Study

The progression of complex human diseases is associated with critical transitions across dynamical regimes. These transitions often spawn early-warning signals and provide insights into the underlying disease-driving mechanisms. In this paper, we propose a computational method based on surprise loss (SL) to discover data-driven indicators of such transitions in a multivariate time series dataset of septic shock and non-sepsis patient cohorts (MIMIC-III database). The core idea of SL is to train a mathematical model on time series in an unsupervised fashion and to quantify the deterioration of the model’s forecast (out-of-sample) performance relative to its past (in-sample) performance. Considering the highest value of the moving average of SL as a critical transition, our retrospective analysis revealed that critical transitions occurred at a median of over 35 hours before the onset of septic shock, which suggests the applicability of our method as an early-warning indicator. Furthermore, we show that clinical variables at critical-transition regions are significantly different between septic shock and non-sepsis cohorts. Therefore, our paper contributes a critical-transition-based data-sampling strategy that can be utilized for further analysis, such as patient classification. Moreover, our method outperformed other indicators of critical transition in complex systems, such as temporal autocorrelation and variance.

Machine learning in critical care: state-of-the-art and a sepsis case study

BACKGROUND

Like other scientific fields, such as cosmology, high-energy physics, or even the life sciences, medicine and healthcare face the challenge of an extremely quick transformation into data-driven sciences. This challenge entails the daunting task of extracting usable knowledge from these data using algorithmic methods. In the medical context this may for instance realized through the design of medical decision support systems for diagnosis, prognosis and patient management. The intensive care unit (ICU), and by extension the whole area of critical care, is becoming one of the most data-driven clinical environments.

RESULTS

The increasing availability of complex and heterogeneous data at the point of patient attention in critical care environments makes the development of fresh approaches to data analysis almost compulsory. Computational Intelligence (CI) and Machine Learning (ML) methods can provide such approaches and have already shown their usefulness in addressing problems in this context. The current study has a dual goal: it is first a review of the state-of-the-art on the use and application of such methods in the field of critical care. Such review is presented from the viewpoint of the different subfields of critical care, but also from the viewpoint of the different available ML and CI techniques. The second goal is presenting a collection of results that illustrate the breath of possibilities opened by ML and CI methods using a single problem, the investigation of septic shock at the ICU.

CONCLUSION

We have presented a structured state-of-the-art that illustrates the broad-ranging ways in which ML and CI methods can make a difference in problems affecting the manifold areas of critical care. The potential of ML and CI has been illustrated in detail through an example concerning the sepsis pathology. The new definitions of sepsis and the relevance of using the systemic inflammatory response syndrome (SIRS) in its diagnosis have been considered. Conditional independence models have been used to address this problem, showing that SIRS depends on both organ dysfunction measured through the Sequential Organ Failure (SOFA) score and the ICU outcome, thus concluding that SIRS should still be considered in the study of the pathophysiology of Sepsis. Current assessment of the risk of dead at the ICU lacks specificity. ML and CI techniques are shown to improve the assessment using both indicators already in place and other clinical variables that are routinely measured. Kernel methods in particular are shown to provide the best performance balance while being amenable to representation through graphical models, which increases their interpretability and, with it, their likelihood to be accepted in medical practice.

Machine Learning for Predicting Outcomes in Trauma

To date, there are no reviews on machine learning (ML) for predicting outcomes in trauma. Consequently, it remains unclear as to how ML-based prediction models compare in the triage and assessment of trauma patients. The objective of this review was to survey and identify studies involving ML for predicting outcomes in trauma, with the hypothesis that models predicting similar outcomes may share common features but the performance of ML in these studies will differ greatly. MEDLINE and other databases were searched for studies involving trauma and ML. Sixty-five observational studies involving ML for the prediction of trauma outcomes met inclusion criteria. In total 2,433,180 patients were included in the studies. The studies focused on prediction of the following outcome measures: survival/mortality (n = 34), morbidity/shock/hemorrhage (n = 12), hospital length of stay (n = 7), hospital admission/triage (n = 6), traumatic brain injury (n = 4), life-saving interventions (n = 5), post-traumatic stress disorder (n = 4), and transfusion (n = 1). Six studies were prospective observational studies. Of the 65 studies, 33 used artificial neural networks for prediction. Importantly, most studies demonstrated the benefits of ML models. However, algorithm performance was assessed differently by different authors. Sensitivity-specificity gap values varied greatly from 0.035 to 0.927. Notably, studies shared many features for model development. A common ML feature base may be determined for predicting outcomes in trauma. However, the impact of ML will require further validation in prospective observational studies and randomized clinical trials, establishment of common performance criteria, and high-quality evidence about clinical and economic impacts before ML can be widely accepted in practice.

Septic shock prediction for ICU patients via coupled HMM walking on sequential contrast patterns

BACKGROUND AND OBJECTIVE

Critical care patient events like sepsis or septic shock in intensive care units (ICUs) are dangerous complications which can cause multiple organ failures and eventual death. Preventive prediction of such events will allow clinicians to stage effective interventions for averting these critical complications.

METHODS

It is widely understood that physiological conditions of patients on variables such as blood pressure and heart rate are suggestive to gradual changes over a certain period of time, prior to the occurrence of a septic shock. This work investigates the performance of a novel machine learning approach for the early prediction of septic shock. The approach combines highly informative sequential patterns extracted from multiple physiological variables and captures the interactions among these patterns via coupled hidden Markov models (CHMM). In particular, the patterns are extracted from three non-invasive waveform measurements: the mean arterial pressure levels, the heart rates and respiratory rates of septic shock patients from a large clinical ICU dataset called MIMIC-II.

EVALUATION AND RESULTS

For baseline estimations, SVM and HMM models on the continuous time series data for the given patients, using MAP (mean arterial pressure), HR (heart rate), and RR (respiratory rate) are employed. Single channel patterns based HMM (SCP-HMM) and multi-channel patterns based coupled HMM (MCP-HMM) are compared against baseline models using 5-fold cross validation accuracies over multiple rounds. Particularly, the results of MCP-HMM are statistically significant having a p-value of 0.0014, in comparison to baseline models. Our experiments demonstrate a strong competitive accuracy in the prediction of septic shock, especially when the interactions between the multiple variables are coupled by the learning model.

CONCLUSIONS

It can be concluded that the novelty of the approach, stems from the integration of sequence-based physiological pattern markers with the sequential CHMM model to learn dynamic physiological behavior, as well as from the coupling of such patterns to build powerful risk stratification models for septic shock patients.

The efficacy of thymosin α1 as immunomodulatory treatment for sepsis: a systematic review of randomized controlled trials

BACKGROUND

Thymosin α1 (Tα1) as immunomodulatory treatment is supposed to be beneficial for the sepsis patients by regulating T cell subsets and inflammatory mediators. However, limited by the small sample size and the poor study design, the persuasive power of the single clinical studies is weak. This meta-analysis aimed to investigate the impact of Tα1 on the sepsis patients.

METHODS

We searched for the Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE, CBM, VIP, CNKI, WANFANG, Igaku Chuo Zasshi (ICHUSHI) and Korean literature databases reporting the effects of Tα1 on outcomes in sepsis patients.

RESULTS

Among 444 related articles, 19 randomized controlled trials (RCTs) met our inclusion criteria. Mortality events were reported in 10 RCTs included 530 patients, and the meta-analysis showed significant decrease in Tα1 group compared with control group (RR 0.59, 95 % CI 0.45 to 0.77, p = 0.0001). The subgroup analysis showed no difference between the two dosages (RR 0.59, 95 % CI 0.43 to 0.81; RR 0.59, 95 % CI 0.35 to 0.98, respectively). In 9 RCTs, with a total of 489 patients, Tα1 administered once per day decrease APACHE II score significantly (SMD -0.80, 95 % CI -1.14 to -0.47, p < 0.0001) while Tα1 twice per day showed no effect (SMD 0.30, 95 % CI-0.10 to 0.70, p = 0.14). However, the length of ICU stay, the incidence of multiple organ failure (MOF) and duration of mechanical ventilation were not significantly affected by Tα1 treatment (SMD -0.52, 95 % CI -1.06 to 0.11, p = 0.06; SMD -0.49, 95 % CI -1.09 to 0.11, p = 0.11; SMD -0.37, 95 % CI -0.90 to 0.17, p = 0.17, respectively). As to the immunological indicators, the level of HLA-DR were increased by Tα1 (SMD 1.23, 95 % CI 0.28 to 2.18, p = 0.01) according to the pooled analysis of 8 studies involving 721 patients. Lymphocyte subsets CD3, CD4 and cytokines IL-6, IL-10 and TNF-α were also beneficially affected by Tα1 treatment.

CONCLUSIONS

Tα1 may be beneficial to sepsis patients in reducing mortality and modulating inflammation reactions. However, the quality of evidence supporting the effectiveness is low considering the small sample sizes and inadequate adherence to standardized reporting guidelines for RCTs among the included studies.

Using demographic and time series physiological features to classify sepsis in the intensive care unit

Sepsis, a systemic inflammatory response to infection, is a major health care problem that affects millions of patients every year in the intensive care units (ICUs) worldwide. Despite the fact that ICU patients are heavily instrumented with physiological sensors, early sepsis detection remains challenging, perhaps because clinicians identify sepsis by (i) using static scores derived from bed-side measurements individually, and (ii) deriving these scores at a much slower rate than the rate for which patient data is collected. In this study, we construct a generalized linear model (GLM) for the probability that an ICU patient has sepsis as a function of demographics and bedside measurements. Specifically, models were trained on 29 patient recordings from the MIMIC II database and evaluated on a different test set including 8 patient recordings. A classification accuracy of 62.5% was achieved using demographic measures as features. Adding physiological time series features to the model increased the classification accuracy to 75%. Although very preliminary, these results suggest that using generalized linear models incorporating real time physiological signals may be useful for an early detection of sepsis, thereby improving the chances of a successful treatment.

Efficacy and safety of low-molecular-weight heparin in patients with sepsis: a meta-analysis of randomized controlled trials

Low-molecular-weight heparin (LMWH) is part of standard supportive care. We conducted a meta-analysis to investigate the efficacy and safety of LMWH in septic patients. We searched Pubmed, Embase, CKNI and Wanfang database prior to July 2015 for randomized controlled trials investigating treatment with LMWH in septic patients. We identified 11 trials involving 594 septic patients. Meta-analysis showed that LMWH significantly reduced prothrombin time (mean differences [MD] −0.88; 95% CI −1.47 to −0.29), APACHE II score (MD −2.50; 95% CI −3.55 to −1.46), and 28-day mortality (risk ratio [RR] 0.72; 95% CI 0.57–0.91) as well as increased the platelet counts (MD 18.33; 95% CI 0.73–35.93) than the usual treatment. However, LMWH did not reduce D-dimer (MD −0.34; 95% CI −0.85 to 0.18). LMWH also significantly increased the bleeding events (RR 3.82; 95% CI 1.81–8.08). LMWH appears to reduce 28-day mortality and APACHE II score among septic patients. Bleeding complications should be monitored during the LMWH treatment. As for limited data about LMWH and sepsis in the English literature, only trials published in the Chinese were included in the meta-analysis.

Using what you get: dynamic physiologic signatures of critical illness

The development and resolution of cardiopulmonary instability take time to become clinically apparent, and the treatments provided take time to have an impact. The characterization of dynamic changes in hemodynamic and metabolic variables is implicit in physiologic signatures. When primary variables are collected with high enough frequency to derive new variables, this data hierarchy can be used to develop physiologic signatures. The creation of physiologic signatures requires no new information; additional knowledge is extracted from data that already exist. It is possible to create physiologic signatures for each stage in the process of clinical decompensation and recovery to improve outcomes.

From data to optimal decision making: a data-driven, probabilistic machine learning approach to decision support for patients with sepsis

BACKGROUND

A tantalizing question in medical informatics is how to construct knowledge from heterogeneous datasets, and as an extension, inform clinical decisions. The emergence of large-scale data integration in electronic health records (EHR) presents tremendous opportunities. However, our ability to efficiently extract informed decision support is limited due to the complexity of the clinical states and decision process, missing data and lack of analytical tools to advice based on statistical relationships.

OBJECTIVE

Development and assessment of a data-driven method that infers the probability distribution of the current state of patients with sepsis, likely trajectories, optimal actions related to antibiotic administration, prediction of mortality and length-of-stay.

METHODS

We present a data-driven, probabilistic framework for clinical decision support in sepsis-related cases. We first define states, actions, observations and rewards based on clinical practice, expert knowledge and data representations in an EHR dataset of 1492 patients. We then use Partially Observable Markov Decision Process (POMDP) model to derive the optimal policy based on individual patient trajectories and we evaluate the performance of the model-derived policies in a separate test set. Policy decisions were focused on the type of antibiotic combinations to administer. Multi-class and discriminative classifiers were used to predict mortality and length of stay.

RESULTS

Data-derived antibiotic administration policies led to a favorable patient outcome in 49% of the cases, versus 37% when the alternative policies were followed (P=1.3e-13). Sensitivity analysis on the model parameters and missing data argue for a highly robust decision support tool that withstands parameter variation and data uncertainty. When the optimal policy was followed, 387 patients (25.9%) have 90% of their transitions to better states and 503 patients (33.7%) patients had 90% of their transitions to worse states (P=4.0e-06), while in the non-policy cases, these numbers are 192 (12.9%) and 764 (51.2%) patients (P=4.6e-117), respectively. Furthermore, the percentage of transitions within a trajectory that lead to a better or better/same state are significantly higher by following the policy than for non-policy cases (605 vs 344 patients, P=8.6e-25). Mortality was predicted with an AUC of 0.7 and 0.82 accuracy in the general case and similar performance was obtained for the inference of the length-of-stay (AUC of 0.69 to 0.73 with accuracies from 0.69 to 0.82).

CONCLUSIONS

A data-driven model was able to suggest favorable actions, predict mortality and length of stay with high accuracy. This work provides a solid basis for a scalable probabilistic clinical decision support framework for sepsis treatment that can be expanded to other clinically relevant states and actions, as well as a data-driven model that can be adopted in other clinical areas with sufficient training data.

Computing network-based features from physiological time series: application to sepsis detection

Sepsis is a systemic deleterious host response to infection. It is a major healthcare problem that affects millions of patients every year in the intensive care units (ICUs) worldwide. Despite the fact that ICU patients are heavily instrumented with physiological sensors, early sepsis detection remains challenging, perhaps because clinicians identify sepsis by using static scores derived from bed-side measurements individually, i.e., without systematically accounting for potential interactions between these signals and their dynamics. In this study, we apply network-based data analysis to take into account interactions between bed-side physiological time series (PTS) data collected in ICU patients, and we investigate features to distinguish between sepsis and non-sepsis conditions. We treated each PTS source as a node on a graph and we retrieved the graph connectivity matrix over time by tracking the correlation between each pair of sources' signals over consecutive time windows. Then, for each connectivity matrix, we computed the eigenvalue decomposition. We found that, even though raw PTS measurements may have indistinguishable distributions in non-sepsis and early sepsis states, the median /I of the eigenvalues computed from the same data is statistically different (p <; 0.001) in the two states and the evolution of /I may reflect the disease progression. Although preliminary, these findings suggest that network-based features computed from continuous PTS data may be useful for early sepsis detection.

Adjuvant treatment with crude rhubarb for patients with systemic inflammation reaction syndrome/sepsis: a meta-analysis of randomized controlled trials

OBJECTIVE

The objective of this study is to evaluate the benefits of adjuvant treatment with crude rhubarb in patients with systemic inflammation reaction syndrome/sepsis by conducting a meta-analysis.

METHODS

We conducted a systematic literature search of medical electronic databases (up to October 2013). Only randomized controlled trials (RCTs) assessing adjuvant treatment with crude rhubarb in septic patients were included.

RESULTS

A total of 15 RCTs with 869 patients were identified. Pooled analysis showed that interleukin 6 (standardized mean differences [SMDs], -1.30; 95% confidence intervals [CIs], -1.94 to -0.66), tumor necrosis factor α (SMD, -0.95; 95% CI, -1.55 to -0.36), procalcitonin (SMD, -1.50; 95% CI, -2.20 to -0.80), von Willebrand factor (mean differences [MDs], -144.11; 95% CI, -253.87 to -34.35), prothrombin time (MD, -2.38; 95% CI, -2.67 to -2.10), acute physiology and chronic health evaluation II scores (MD, -4.51; 95% CI, -5.30 to -3.73), and gastrointestinal dysfunction (risk ratio, 0.28; 95% CI, 0.16-0.49) were significantly reduced after treatment with crude rhubarb. Platelet number (MD, 58.16; 95% CI, 51.16-65.15) was significantly increased. However, crude rhubarb therapy did not significantly reduce 28-day mortality (risk ratio, 0.60; 95% CI, 0.36-1.00) compared with the usual treatment.

CONCLUSIONS

Adjuvant treatment with crude rhubarb appears to have additional benefits in septic patients. Antiinflammation and anticoagulant/antiaggregant properties may be its potential mechanism.

From vital signs to clinical outcomes for patients with sepsis: a machine learning basis for a clinical decision support system

OBJECTIVE

To develop a decision support system to identify patients at high risk for hyperlactatemia based upon routinely measured vital signs and laboratory studies.

MATERIALS AND METHODS

Electronic health records of 741 adult patients at the University of California Davis Health System who met at least two systemic inflammatory response syndrome criteria were used to associate patients' vital signs, white blood cell count (WBC), with sepsis occurrence and mortality. Generative and discriminative classification (naïve Bayes, support vector machines, Gaussian mixture models, hidden Markov models) were used to integrate heterogeneous patient data and form a predictive tool for the inference of lactate level and mortality risk.

RESULTS

An accuracy of 0.99 and discriminability of 1.00 area under the receiver operating characteristic curve (AUC) for lactate level prediction was obtained when the vital signs and WBC measurements were analysed in a 24 h time bin. An accuracy of 0.73 and discriminability of 0.73 AUC for mortality prediction in patients with sepsis was achieved with only three features: median of lactate levels, mean arterial pressure, and median absolute deviation of the respiratory rate.

DISCUSSION

This study introduces a new scheme for the prediction of lactate levels and mortality risk from patient vital signs and WBC. Accurate prediction of both these variables can drive the appropriate response by clinical staff and thus may have important implications for patient health and treatment outcome.

CONCLUSIONS

Effective predictions of lactate levels and mortality risk can be provided with a few clinical variables when the temporal aspect and variability of patient data are considered.

New approaches to the study of sepsis

Models of sepsis have been instructive in understanding the sequence of events in animals and, to an extent, in humans with sepsis. Events developing early in sepsis suggest that a hyperinflammatory state exists, accompanied by a buildup of oxidants in tissues reflective of a redox imbalance. Development of immunosuppression and degraded innate and adaptive immune responses are well-established complications of sepsis. In addition, there is robust activation of the complement system, which contributes to the harmful effects of sepsis. These events appear to be associated with development of multiorgan failure. The relevance of animal models of sepsis to human sepsis and the failure of human clinical trials are discussed, together with suggestions as to how clinical trial design might be improved.

Estimation of cardiac output and systemic vascular resistance using a multivariate regression model with features selected from the finger photoplethysmogram and routine cardiovascular measurements

Background

Cardiac output (CO) and systemic vascular resistance (SVR) are two important parameters of the cardiovascular system. The ability to measure these parameters continuously and noninvasively may assist in diagnosing and monitoring patients with suspected cardiovascular diseases, or other critical illnesses. In this study, a method is proposed to estimate both the CO and SVR of a heterogeneous cohort of intensive care unit patients (N=48).

Methods

Spectral and morphological features were extracted from the finger photoplethysmogram, and added to heart rate and mean arterial pressure as input features to a multivariate regression model to estimate CO and SVR. A stepwise feature search algorithm was employed to select statistically significant features. Leave-one-out cross validation was used to assess the generalized model performance. The degree of agreement between the estimation method and the gold standard was assessed using Bland-Altman analysis.

Results

The Bland-Altman bias ±precision (1.96 times standard deviation) for CO was -0.01 ±2.70 L min^-1 when only photoplethysmogram (PPG) features were used, and for SVR was -0.87 ±412 dyn.s.cm^-5 when only one PPG variability feature was used.

Conclusions

These promising results indicate the feasibility of using the method described as a non-invasive preliminary diagnostic tool in supervised or unsupervised clinical settings.

Identification of apnea during respiratory monitoring using support vector machine classifier: a pilot study

To determine the use of photoplethysmography (PPG) as a reliable marker for identifying respiratory apnea based on time-frequency features with support vector machine (SVM) classifier. The PPG signals were acquired from 40 healthy subjects with the help of a simple, non-invasive experimental setup under normal and induced apnea conditions. Artifact free segments were selected and baseline and amplitude variabilities were derived from each recording. Frequency spectrum analysis was then applied to study the power distribution in the low frequency (0.04-0.15 Hz) and high frequency (0.15-0.40 Hz) bands as a result of respiratory pattern changes. Support vector machine (SVM) learning algorithm was used to distinguish between the normal and apnea waveforms using different time-frequency features. The algorithm was trained and tested (780 and 500 samples respectively) and all the simulations were carried out using linear kernel function. Classification accuracy of 97.22 % was obtained for the combination of power ratio and reflection index features using SVM classifier. The pilot study indicates that PPG can be used as a cost effective diagnostic tool for detecting respiratory apnea using a simple, robust and non-invasive experimental setup. The ease of application and conclusive results has proved that such a system can be further developed for use in real-time monitoring under critical care conditions.

Photoplethysmographic variability analysis in critical care--current progress and future challenges

The concept of early goal-directed therapy emphasizes the need for early diagnosis and intervention to achieve better therapeutic outcomes in critical care. There has been rapidly growing interest in the use of the photoplethysmogram (PPG), also known as the "pulse oximetry waveform", as a noninvasive diagnostic tool in this clinical setting. The peripheral PPG exhibits beat-to-beat variability driven by physiological mechanisms such as respiration and sympathetic vascular activity. This paper provides an overview of the current progress towards the application of PPG waveform variability (PPGV) in emergency and intensive care. Studies to date have demonstrated the potential value of PPGV for assessing a range of pathophysiological conditions including blood loss, sepsis and low systemic vascular resistance. Translation of research findings into clinical practice poses several future challenges, including the need for large scale validation studies with appropriate measurement systems, more robust solutions to signal quality issues (such as motion artifacts), and better physiological understanding of the information-rich PPGV.

Multivariate classification of systemic vascular resistance using photoplethysmography

Systemic vascular resistance (SVR) classification is useful for the diagnosis and prognosis of critical pathophysiological conditions, with the ability to identify patients with abnormally high or low SVR of immense clinical value. In this study, a supervised classifier, based on Bayes' rule, is employed to classify a heterogeneous group of intensive care unit patients (N = 48) as being below (SVR < 900 dyn s cm(-5)), within (900 ⩽ SVR ⩽ 1200 dyn s cm(-5)) or above (SVR > 1200 dyn s cm(-5)) the clinically accepted range for normal SVR. Features derived from the finger photoplethysmogram (PPG) waveform and other routine cardiovascular measurements (heart rate and mean arterial pressure) were used as inputs to the classifier. In the construction of the classifier model, two techniques were used to approximate the class conditional probability densities--a single Gaussian distribution model (also known as discriminant analysis) and a non-parametric model using the Parzen window kernel density estimation method. An exhaustive feature search was performed to select a feature subset that maximized the performance indicator, Cohen's kappa coefficient (κ). The Gaussian model with multiple features achieved the best overall kappa coefficient (κ = 0.57), although the results from the non-parametric model were comparable (κ = 0.51). The optimum subset in the Gaussian model consisted of PPG waveform variability features, including the low-frequency to high-frequency ratio (LF/HF) and the normalized mid-frequency power (MF(NU)), in addition to the PPG pulse wave features, such as pulse width, peak-to-notch time, reflection index, and notch time ratio. The classifier performed particularly well in discriminating low SVR, with a sensitivity of 85%, specificity of 86%, positive predictive value of 88% and a negative predictive value of 82%. The results highlight the feasibility of deploying a multivariate statistical approach of SVR classification in the clinical setting, simply using a non-invasive and easy-to-measure PPG waveform signal.

Fingertip photoplethysmographic waveform variability and systemic vascular resistance in intensive care unit patients

Low frequency variability in the fingertip photoplethysmogram (PPG) waveform has been utilized for inferring sympathetic vascular control, but its relationship with a quantitative measure of vascular tone has not been established. In this study, we examined the association between fingertip PPG waveform variability (PPGV) and systemic vascular resistance (SVR) obtained from thermodilution cardiac output (CO) and intra-arterial pressure measurements in 48 post cardiac surgery intensive care unit patients. Among the hemodynamic measurements, both CO (P < 0.05) and SVR (P < 0.0001) had statistically significant relationships with the normalized low frequency power (LF(nu)) of PPGV. The LF(nu) of baseline PPGV had moderate but significant positive correlation with SVR (r = 0.54, P < 0.0001), and a value below 52.5 nu was able to identify SVR < 900 dyn s cm⁻⁵ with sensitivity of 59% and specificity of 95%. The results have provided quantitative evidence to confirm the link between fingertip PPGV and sympathetic vascular control. Suppression of LF vasomotor waves leading to dominance of respiration-related HF fluctuations in the fingertip circulation was a specific (though not sensitive) marker of systemic vasodilatation, which could be potentially utilized for the assessment of critical care patients.

Pulse transit time variability analysis in an animal model of endotoxic shock

The use of non-invasively measured pulse transit time (PTT) to monitor the cardiovascular systems in critically ill patients, like sepsis, can be of significant clinical value. In this study, the potential of PTT and its variability in cardiovascular system monitoring in a mechanically ventilated and anesthetized rabbit model of endotoxic shock was assessed. Eight adult New Zealand white rabbits, which were treated with endotoxin bolus infusion, were studied. Measurements of PTT, pre-ejection period (PEP), and vascular transit time (VTT) were obtained in pre- and post-intervention stages (before and 90 minutes after the administration of endotoxin). The decrease in mean PTT (p < 0.05) and PEP (p < 0.01) in the post-intervention stage reflected sympathetic activation, whilst the increase in respiratory variation in PTT (p < 0.01), PEP (p 〈 0.01), and VTT (p < 0.01) could be attributed to an enhancement of respiratory variation in stroke volume associated with hypovolemia in endotoxic shock. The relationship between beat-to-beat variability in PTT and all other cardiovascular time series were further investigated through linear regression analysis, which revealed that PTT was most strongly correlated with VTT (R(2) ≥ 0.84 with positive slope). Computation of coherence and phase shift in the ventilating frequency band (HF: 0.50 - 0.75 Hz) showed that the respiratory variation in PTT was synchronized with both PEP and VTT (coherence > 0.84 with phase shift less than one cardiac beat). These results highlighted the potential value of PTT and its respiratory variation in characterizing the pathophysioloigcal hemodynamic change in endotoxic shock.