Search

Many in-hospital mortality risk prediction scores dichotomize predictive variables to simplify the score calculation. However, hard thresholding in these additive stepwise scores of the form "add x points if variable v is above/below threshold t" may lead to critical failures. In this paper, we seek to develop risk prediction scores that preserve clinical knowledge embedded in features and structure of the existing additive stepwise scores while addressing limitations caused by variable dichotomization. To this end, we propose a novel score structure that relies on a transformation of predictive variables by means of nonlinear logistic functions facilitating smooth differentiation between critical and normal values of the variables. We develop an optimization framework for inferring parameters of the logistic functions for a given patient population via cyclic block coordinate descent. The parameters may readily be updated as the patient population and standards of care evolve. We tested the proposed methodology on two populations: (1) brain trauma patients admitted to the intensive care unit of the Dell Children's Medical Center of Central Texas between 2007 and 2012, and (2) adult ICU patient data from the MIMIC II database. The results are compared with those obtained by the widely used PRISM III and SOFA scores. The prediction power of a score is evaluated using area under ROC curve, Youden's index, and precision-recall balance in a cross-validation study. The results demonstrate that the new framework enables significant performance improvements over PRISM III and SOFA in terms of all three criteria., (Copyright © 2015 Elsevier Inc. All rights reserved.)

Soft-constraint semi-supervised affinity propagation (SCSSAP) adds supervision to the affinity propagation (AP) clustering algorithm without strictly enforcing instance-level constraints. Constraint violations lead to an adjustment of the AP similarity matrix at every iteration of the proposed algorithm and to addition of a penalty to the objective function. This formulation is particularly advantageous in the presence of noisy labels or noisy constraints since the penalty parameter of SCSSAP can be tuned to express our confidence in instance-level constraints. When the constraints are noiseless, SCSSAP outperforms unsupervised AP and performs at least as well as the previously proposed semi-supervised AP and constrained expectation maximization. In the presence of label and constraint noise, SCSSAP results in a more accurate clustering than either of the aforementioned established algorithms. Finally, we present an extension of SCSSAP which incorporates metric learning in the optimization objective and can further improve the performance of clustering.

Spaceflight-induced orthostatic intolerance has been studied for decades. Although ∼22% of the astronaut corps are women, most mechanistic studies use mostly male subjects, despite known sex differences in autonomic control and postflight orthostatic intolerance. We studied adrenergic, baroreflex, and autonomic indexes during continuous infusions of vasoactive drugs in men and women during a 60-day head-down bed rest. Volunteers were tested before bed rest (20 men and 10 women) and around day 30 (20 men and 10 women) and day 60 (16 men and 8 women) of bed rest. Three increasing doses of phenylephrine (PE) and sodium nitroprusside were infused for 10 min after an infusion of normal saline. A 20-min rest period separated the phenylephrine and sodium nitroprusside infusions. Autonomic activity was approximated by spectral indexes of heart rate and blood pressure variability, and baroreflex sensitivity was measured by the spontaneous baroreflex slope. Parasympathetic modulation and baroreflex sensitivity decreased with bed rest, with women experiencing a larger decrease in baroreflex sensitivity by day 30 than men. The sympathetic activation of men and parasympathetic responsiveness of women in blood pressure control during physiological stress were preserved throughout bed rest. During PE infusions, women experienced saturation of the R-R interval at high frequency, whereas men did not, revealing a sex difference in the parabolic relationship between high-frequency R-R interval, a measurement of respiratory sinus arrhythmia, and R-R interval. These sex differences in blood pressure control during simulated microgravity reveal the need to study sex differences in long-duration spaceflight to ensure the health and safety of the entire astronaut corps.

This study examined the reliability and validity of using customized templates to acquire panoramic ultrasound (US) images for determining cross-sectional area (CSA) and volume in the vastus lateralis (VL), rectus femoris (RF), medial gastrocnemius (MG) and lateral gastrocnemius (LG). Panoramic US and magnetic resonance imaging (MRI) images were analyzed by two trained investigators. The inter-experimenter reliability (coefficient of variation [CV]) of panoramic US ranged from 2.4% to 4.1% and the intraclass correlation (ICC) ranged from 0.963 to 0.991, whereas the inter-experimenter CV of MRI ranged from 2.8% to 3.8% and the ICC from 0.946 to 0.986. Bland-Altman plots demonstrated high agreement between US and MRI; however, values obtained from MRI were systematically larger than those obtained from US. The present results indicate that using a customized US template provides reliable measures of leg muscle CSA and, thus, could be used to characterize muscle CSA and volume., (Copyright © 2012 World Federation for Ultrasound in Medicine & Biology. All rights reserved.)

A paradox regarding the classic power spectral analysis of heart rate variability (HRV) is whether the characteristic high- (HF) and low-frequency (LF) spectral peaks represent stochastic or chaotic phenomena. Resolution of this fundamental issue is key to unraveling the mechanisms of HRV, which is critical to its proper use as a noninvasive marker for cardiac mortality risk assessment and stratification in congestive heart failure (CHF) and other cardiac dysfunctions. However, conventional techniques of nonlinear time series analysis generally lack sufficient sensitivity, specificity and robustness to discriminate chaos from random noise, much less quantify the chaos level. Here, we apply a 'litmus test' for heartbeat chaos based on a novel noise titration assay which affords a robust, specific, time-resolved and quantitative measure of the relative chaos level. Noise titration of running short-segment Holter tachograms from healthy subjects revealed circadian-dependent (or sleep/wake-dependent) heartbeat chaos that was linked to the HF component (respiratory sinus arrhythmia). The relative 'HF chaos' levels were similar in young and elderly subjects despite proportional age-related decreases in HF and LF power. In contrast, the near-regular heartbeat in CHF patients was primarily nonchaotic except punctuated by undetected ectopic beats and other abnormal beats, causing transient chaos. Such profound circadian-, age- and CHF-dependent changes in the chaotic and spectral characteristics of HRV were accompanied by little changes in approximate entropy, a measure of signal irregularity. The salient chaotic signatures of HRV in these subject groups reveal distinct autonomic, cardiac, respiratory and circadian/sleep-wake mechanisms that distinguish health and aging from CHF.

Accurate QRS detection is an important first step for the analysis of heart rate variability. Algorithms based on the differentiated ECG are computationally efficient and hence ideal for real-time analysis of large datasets. Here, we analyze traditional first-derivative based squaring function (Hamilton-Tompkins) and Hilbert transform-based methods for QRS detection and their modifications with improved detection thresholds. On a standard ECG dataset, the Hamilton-Tompkins algorithm had the highest detection accuracy (99.68% sensitivity, 99.63% positive predictivity) but also the largest time error. The modified Hamilton-Tompkins algorithm as well as the Hilbert transform-based algorithms had comparable, though slightly lower, accuracy; yet these automated algorithms present an advantage for real-time applications by avoiding human intervention in threshold determination. The high accuracy of the Hilbert transform-based method compared to detection with the second derivative of the ECG is ascribable to its inherently uniform magnitude spectrum. For all algorithms, detection errors occurred mainly in beats with decreased signal slope, such as wide arrhythmic beats or attenuated beats. For best performance, a combination of the squaring function and Hilbert transform-based algorithms can be applied such that differences in detection will point to abnormalities in the signal that can be further analyzed.

Linear and nonlinear indices of heart rate variability (HRV) have been shown to predict mortality in congestive heart failure (CHF). However, most nonlinear indices describe only the fractality or complexity of HRV but not the intrinsic chaotic properties. In the present study, we performed linear (time- and frequency-domain), complexity (sample entropy), fractal (detrended fluctuation analysis) and chaos (numerical titration) analyses on the HRV of 50 CHF patients from the United Kingdom heart failure evaluation and assessment of risk trial database. Receiver operating characteristic and survival analysis yielded the chaos level to be the best predictor of mortality (followed by low/high frequency power ratio, LF/HF), such that these indices were significant in both univariate and multivariate models. These results indicate the power of heart rate chaos analysis as a potential prognostic tool for CHF.

Obstructive sleep apnea syndrome (OSAS) is observed in approximately 2% of children. Heart rate variability (HRV) is a potentially simple, non-invasive diagnostic screening tool for OSAS. In this study, we investigated the diagnostic potential of HRV using power spectral analysis, numerical titration, sample entropy, and detrended fluctuation analysis. Effects of sleep stages (REM and NREM sleep) are evaluated. The results show that the heart rate chaos intensity, as measured by the noise limit in numerical titration, is significantly higher during REM sleep than NREM sleep in all patient groups. By using the receiver-operating characteristic analysis, the detection of OSAS yielded a specificity of 72.2% and sensitivity of 81.3% using the numerical-titration technique. The findings suggest that sleep state and disordered breathing are important determinants of cardiac autonomic control. Nonlinear techniques such as numerical titration, when used in conjunction with spectral analysis of HRV could be an effective screening tool for pediatric OSAS.

Accurate processing of electrocardiogram (ECG) signals requires a sensitive and robust QRS detection method. In this study, three methods are quantitatively compared using a similar algorithm structure but applying different transforms to the differentiated ECG. The three transforms used are the Hilbert transformer, the squaring function, and a second discrete derivative stage. The first two have been widely used in ECG and heart rate variability analysis while the second derivative method aims to explain the success of the Hilbert transform. The algorithms were compared in terms of the number of false positive and false negative detections produced for records of the MIT/BIH Arrhythmia Database. The Hilbert transformer and the squaring function both produced a sensitivity and positive predictivity of over 99%, though the squaring function had a lower overall detection error rate. The second derivative resulted in the highest overall detection error rate. Different algorithms performed better for diverse ECG characteristics; suggesting that an algorithm can be specified for different recordings, the algorithms can be combined based on each one's characteristics to determine a new more accurate method, or an additional detection stage can be added to reduce the number of false negatives.