In this study, a semi-automatic, easy-to-use classification method for the identification and removal of fMRI noise is proposed and tested. The method relies on subject-level spatial independent component analysis (ICA) of fMRI data. Starting from a reference set of labeled independent components (ICs), novel ICs are classified as physiological/artefactual by combining a spatial correlation (SC) analysis with the reference ICs and relative power spectral (PS) analysis. Here, ICs from a task-based fMRI dataset were used as reference. SC and SP thresholds were set using a test dataset (5 subjects, same fMRI protocol) based on Receiving Operating Characteristic curves. The tool performance and versatility were measured on a resting-state fMRI dataset (5 subjects). Our results show that the method can automatically identify noise-related ICs with accuracy, specificity and sensitivity higher than 80% across different fMRI protocols. These findings also suggest that the reference set provided in the present study might be used to mark ICs coming from independent taskrelated or resting-state fMRI datasets.Clinical relevance- The new method will be included in a userfriendly, open-source tool for removal of noisy contributions from fMRI datasets to be used in clinical and research practices.

In emotional and cognitive research, the baseline is commonly used for standardization purposes in order to have a reference for the identification of the activation. Since no previous studies have investigated which moment of the experiment could be considered optimal for baseline evaluation, we designed an experimental protocol to analyze which time interval could be considered more effective in highlighting differences between the baseline state and the cognitive effort exhibited during tasks (specifically, reaction and working memory tasks). Several indexes were extracted from EEG signals during the visualization of the considered baseline stimuli and the execution of tasks. From our results, as regards to the considered Global Field Power (GFP) indexes (Attention and Memorization indexes), the last baseline stimulus seems to be the best one to highlight the difference in cognitive workload between the individual baseline condition and the two cognitive tasks. Instead, in terms of Engagement index (EI), the difference between Reaction Task (RT) and the individual baseline condition seems to be best highlighted by the relaxing video right after performed tasks. In conclusion, the best baseline position to maximize the differences in cognitive workload may vary among the considered indexes because of confounding effects and individual differences, but further analyses are required to validate this result.

Hemorrhage is a frequent complication in surgery patients; its identification and management have received increasing attention as a target for quality improvement in patient care in the Intensive Care Unit (ICU). The purposes of this work were 1) to find an early detection model for hemorrhage by exploring the range of data mining methods that are currently available, and 2) to compare prediction models utilizing continuously measured physiological data from bedside monitors to those using commonly obtained laboratory tests. We studied 3766 patients admitted to the University of Virginia Health System Surgical Trauma Burn ICU. Hemorrhage was defined as three or more units of red blood cells transfused within 24 h without red blood cell transfusion in the preceding 24 h. 222 patients (5.9%) experienced a hemorrhage, and multivariate models based on vital signs and their trends showed good results (AUC = 76.1%). The hematocrit, not surprisingly, had excellent performance (AUC = 87.7%). Models that included both continuous monitoring and laboratory tests had the best performance (AUC = 92.2%). The results point to a combined strategy of continuous monitoring and intermittent lab tests as a reasonable clinical approach to the early detection of hemorrhage in the surgical ICU.

In this study, a functional clustering approach is proposed and tested for the identification of brain functional networks emerging during sleep-related seizures. Stereo-EEG signals recorded in patients with Type II Focal Cortical Dysplasia (FCD type II), were analyzed. This novel approach is able to identify the network configuration changes in pre-ictal and early ictal periods, by grouping Stereo-EEG signals on the basis of the Cluster Index, after wavelet multiscale decomposition. Results showed that the proposed method is able to detect clusters of interacting leads, mainly overlapped on the Epileptogenic Zone (EZ) identified by a clinical expert, with distinctive configurations related to analyzed frequency ranges. This suggested the presence of coupling activities between the elements of the epileptic system at different frequency scales.

We surely agree with Wessel et al. that heart period fluctuations are linked to many physiological oscillations. As already explained in the 1996 heart rate variability (HRV) standards,1 these oscillations include those of respiration,2 blood pressure,3 etc. The shifts in the respiration-related HRV components into the nominal low-frequency band are also known4 as is the fact that the low-frequency HRV modulations cannot be taken as a single reliable expression of sympathetic activity.1 It is difficult to consider respiration as an overwhelming determinant of short-term HRV. Adding additional provocations to constant controlled breathing still leads …

Atrial fibrillation (AF) is usually detected by inspection of the electrocardiogram waveform, a task made difficult when the signal is distorted by noise. The RR interval time series is more frequently available and accurate, yet linear and nonlinear time series analyses that detect highly varying and irregular AF are vulnerable to the common finding of frequent ectopy. We hypothesized that different nonlinear measures might capture characteristic features of AF, normal sinus rhythm (NSR), and sinus rhythm (SR) with frequent ectopy in ways that linear measures might not. To test this, we studied 2722 patients with 24 h ECG recordings in the University of Virginia Holter database. We found dynamical phenotypes for the three rhythm classifications. As expected, AF records had the highest variability and entropy, and NSR the lowest. SR with ectopy could be distinguished from AF, which had higher entropy, and from NSR, which had different fractal scaling, measured as higher detrended fluctuation analysis slope. With these dynamical phenotypes, we developed successful classification strategies, and the nonlinear measures improved on the use of mean and variability alone, even after adjusting for age. Final models using all variables had excellent performance, with positive predictive values for AF, NSR and SR with ectopy as high as 97, 98 and 90%, respectively. Since these classifiers can reliably detect rhythm changes utilizing segments as short as 10 min, we envision their application in noisy settings and in personal monitoring devices where only RR interval time series may be available.

There is a growing interest in identifying neuroimaging-based biomarkers for Parkinson's disease (PD), a progressive neurodegenerative disorder in which the major pathologic substrate is the loss of pigmented dopaminergic neurons in the substantia nigra (SN). Recently, an MRI technique dubbed "neuromelanin-sensitive MRI" (NM-MRI), has been found to provide notable contrast between the SN and surrounding brain tissues with potential applications as biomarker of PD. The contrast in NM-MRI has been associated with magnetization transfer (MT) effects, and thus the goal of this study was to characterize the impact of MT on NM-MRI, and to demonstrate the feasibility of performing quantitative MT (qMT) imaging in human SN. The results of this study demonstrate that high-resolution rapid qMT imaging of the SN can be reliably obtained within reasonable scan times, thereby can be translatable into clinical practice.

The study exploits a Wiener-Granger causality (WGC) approach in the time domain to assess directionality of the dynamical interactions between QT interval and heart period (HP) during a graded head-up tilt protocol challenging the cardiovascular control as a function of the tilt table inclination. QT interval and HP are approximated from the surface ECG as the temporal distance between the R-wave apex and T-wave offset and between two consecutive R-wave peaks respectively. The adopted WGC approach accounts for the confounding effect of respiration (RESP) affecting both QT and HP. Causality ratios (CRs) from HP to QT given RESP (CRHP→QT|RESP) and from QT to HP given RESP (CRQT→HP|RESP) were computed and their significance was tested via F-test. We found that, regardless of the tilt table angle, CRHP→QT|RESP is significant, while CRQT→HP|RESP is negligible. CRHP→QT|RESP showed a trend towards a decrease with tilt table angle. These findings suggested that the causal direction from HP over QT is dominant compared to the reverse one and sympathetic activation induced a QT-HP uncoupling.

The analysis of non-linear dynamics of the coupling among interacting quantities can be very useful for understanding the cardiorespiratory and cardiovascular control mechanisms. In this chapter RP is used to detect and quantify the degree of non-linear coupling between respiration and spontaneous rhythms of both heart rate and blood pressure variability signals. RQA turned out to be suitable for a quantitative evaluation of the observed coupling patterns among rhythms, both in simulated and real data, providing different degrees of coupling. The results from the simulated data showed that the increased degree of coupling between the signals was marked by the increase of PR and PD, and by the decrease of ER. When the RQA was applied to experimental data, PD and ER turned out to be the most significant variables, compared to PR. A remarkable finding is the detection of transient 1:2 PL episodes between respiration and cardiovascular variability signals. This phenomenon can be associated to a sub-harmonic synchronization between the two main rhythms of HR and BP variability series.

This study assesses complexity of cardiovascular control in patients affected by type-1 variant of long QT (LQT1) syndrome. Complexity was assessed by refined multiscale entropy of heart period (HP) and QT interval variabilities. HP was taken as the time distance between two consecutive R peaks (RR) and QT interval was approximated as the time distance between the R-peak and T-wave apex (RTa) and between R-peak and T-wave end (RTe). RR, RTa and RTe intervals were automatically extracted from 24h Holter recordings and the daytime period was analyzed (from 02:00 to 06:00 PM). Non mutation carrier (NMC) individuals (n=11), utilized as a control group, were taken from the same family line of the mutation carrier (MC) subjects (n=26). We found that, while NMC and MC groups were indistinguishable based on time domain and complexity analyses of RR dynamics, complexity analysis of RTa and RTe variabilities clearly separates the two populations and suggests an impairment in the cardiac control mechanisms acting on the ventricles.

End stage renal disease condition increases the risk of cardiovascular disease. The mortality rates among hemodialysis patients are 20% higher than the general population, thus in recent years the preservation of the cardiovascular system has become a major point of focus for nephrology care in patients. Cardiovascular events jeopardize the life of a dialysis patient and must therefore be prevented. The aim of this study is to develop forecast models that can predict the cardiovascular outcome of incident hemodialysis (HD) patients. Data relating to the treatment methods and the physiological condition of patients was collected during the first 18months of renal replacement therapy and then used to predict the insurgence of cardiovascular events within a 6-month time window. Information regarding 4246 incident hemodialysis patients was collected. A Lasso logistic regression model and a random forest model were developed and used for predictive comparison. Every forecast model was tested on 20% of the data and a 5-fold cross validation approach was used to validate the random forest model. Random forest showed higher performance with AUC of the ROC curve and sensitivity higher than 70% in both the temporal windows models, proving that random forests are able to exploit non-linear patterns retrieved in the feature space. Out of bag estimates of variable importance and regression coefficients were used to gain insight into the models implemented. We found out that malnutrition and an inflammatory condition strongly influence cardiovascular outcome in incident HD patients. Indeed the most important variables in the model were blood test variables such as the total protein content, percentage value of albumin, total protein content, creatinine and C reactive protein. Age of patients and weight loss in the first six months of renal replacement therapy were also highly involved in the prediction. A greater understanding of the mechanisms involved in the insurgence of cardiovascular events in dialysis patients can ensure physicians to intervene in the appropriate manner when a high-risk cardiovascular condition is identified.

Sleep is a complex state characterized by important changes in the autonomic modulation of the cardiovascular activity. Heart rate variability (HRV) greatly changes during different sleep stages, showing a predominant parasympathetic drive to the heart during non-rapid eye movement (NREM) sleep and an increased sympathetic activity during rapid eye movement (REM) sleep. Respiration undergoes important modifications as well, becoming deeper and more regular with deep sleep and shallower and more frequent during REM sleep. The aim of the present study is to assess both autonomic cardiac regulation and cardiopulmonary coupling variations during different sleep stages in healthy subjects, using spectral and cross-spectral analysis of the HRV and respiration signals. Polysomnographic sleep recordings were performed in 11 healthy women and the HRV signal and the respiration signal were obtained. The spectral and cross-spectral parameters of the HRV signal and of the respiration signal were computed at low frequency and at breathing frequency (high frequency, HF) during different sleep stages. Results attested a sympatho-vagal balance shift toward parasympathetic modulation during NREM sleep and toward sympathetic modulation during REM sleep. Spectral analysis of the HRV signal and of the respiration signal indicated a higher respiration regularity during deep sleep, and a higher parasympathetic drive was also confirmed by an increase in the coherence between the HRV and the respiration signal in the HF band during NREM sleep. Our findings about sleep stage-dependent variations in the HRV signal and in the respiratory activity are in line with previous evidences and confirm spectral analysis of the HRV and the respiration signal to be a suitable tool for investigating cardiac autonomic modulation and cardio-respiratory coupling during sleep.

The main objective of this work is to develop and apply data mining methods for the prediction of patient outcome in nephrology care. Cardiovascular events have an incidence of 20% in the first year of hemodialysis (HD). Real data routinely collected during HD administration were extracted from the Fresenius Medical Care database EuCliD (39 independent variables) and used to develop a random forest predictive model for the forecast of cardiovascular events in the first year of HD treatment. Two feature selection methods were applied. Results of these models in an independent cohort of patients showed a significant predictive ability. Our better result was obtained with a random forest built on 6 variables only (AUC: 77.1% ± 2.9%; MCE: 31.6% ± 3.5%), identified by the variable importance out of bag (OOB) estimate.

In this paper, we propose the use of black box models for the system identification of the cardiopulmonary baroreflex control of arterial resistance and of ventricular contractility and of arterial baroreflex control of heart rate (HR) from invasive, continuous measurements of arterial blood pressure (ABP) and central venous pressure (CVP), and non invasive, continuous recordings of ECG and respiration. Two crucial phases of the abdominal aortic aneurism (AAA) repair were investigated: the clamping and declamping of aorta. The objective of the present work is to evaluate and to test the ability to monitor baroreflex responses to clamping and declamping maneuvers preceding and following aneurism removal.

To celebrate the 20th anniversary of the first Bioengineering Department in Italy at the Politecnico di Milano, a number of articles in this issue highlight the department's achievements. The department was eventually founded in 1990 after many years of research and didactic activity in the area that officially began in 1969 when the first teaching course, called biological electronics, was offered in the electronic engineering specialization.

Objective: This study aims to identify, starting from a single EEG trace, quantitative distinctive features characterizing the A phases of the Cyclic Alternating Pattern (CAP).Methods: The C3-A2 or C4-A1 EEG leads of the night recording of eight healthy adult subjects were used for this analysis. CAP was scored by an expert and the portions relative to NREM were selected. Nine descriptors were computed: band descriptors (low delta, high delta, theta, alpha, sigma and beta); Hjorth activity in the low delta and high delta bands; differential variance of the EEG signal. The information content of each descriptor in recognizing the A phases was evaluated through the computation of the ROC curves and the statistics sensitivity, specificity and accuracy.Results: The ROC curves show that all the descriptors have a certain significance in characterizing A phases. The average accuracy obtained by thresholding the descriptors ranges from 59.89 (sigma descriptor) to 72.44 (differential EEG variance).Conclusions: The results show that it is possible to attribute a significant quantitative value to the information content of the descriptors.Significance: This study gives a mathematical confirm to the features of CAP generally described qualitatively, and puts the bases for the creation of automatic detection methods. (C) 2011 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.

The study of neural correlates of motor execution is commonly performed by means of event-related processing of electroencephalographic (EEG) recordings, in which each event refers to a standardized, repeatable movement. Some authors have proposed a valuable single-parameter method, the Event-Related Synchronization and Desynchronization (ERS/ERD) approach, for the identification of motor-related power modulation in each EEG frequency band. Under evolving experimental conditions (such as learning or adaptation), though, the repetition of a motor scheme becomes time-variant, and the employment of single-parameter descriptors no longer represents the optimal choice. This occurrence is typically found in motor learning and adaptation studies. In this work we compared the performance of the ERS/ERD method with the multi-parametric Hilbert Huang Transform (HHT). Results confirmed the statistically significant equivalence of the two methods in providing indexes of neural synchronization and desynchronization. Moreover, HHT allowed the tracking of frequency shifts in the alpha and beta EEG bands. The two methods were tested on an EEG dataset recorded during a motor adaptation test.

This article presents a review on biomedical signal processing. Discussions on traditional approaches, nonstationary and nonlinear systems, signal fusion, physiological modeling, and the MMM (multivariate, multiorgan and multiscale) paradigm are included.