Farid Yaghouby

Towards automatic detection of atrial fibrillation: A hybrid computational approach

In this study, new methods coupling genetic programming with orthogonal least squares (GP/OLS) and simulated annealing (GP/SA) were applied to the detection of atrial fibrillation (AF) episodes. Empirical equations were obtained to classify the samples of AF and Normal episodes based on the analysis of RR interval signals. Another important contribution of this paper was to identify the effective time domain features of heart rate variability (HRV) signals via an improved forward floating selection analysis. The models were developed using the MIT-BIH arrhythmia database. A radial basis function (RBF) neural networks-based model was further developed using the same features and data sets to benchmark the GP/OLS and GP/SA models. The diagnostic performance of the GP/OLS and GP/SA classifiers was evaluated using receiver operating characteristics analysis. The results indicate a high level of efficacy of the GP/OLS model with sensitivity, specificity, positive predictivity, and accuracy rates of 99.11%, 98.91%, 98.23%, and 99.02%, respectively. These rates are equal to 99.11%, 97.83%, 98.23%, and 98.534% for the GP/SA model. The proposed GP/OLS and GP/SA models have a significantly better performance than the RBF and several models found in the literature.

Quasi-supervised scoring of human sleep in polysomnograms using augmented input variables

The limitations of manual sleep scoring make computerized methods highly desirable. Scoring errors can arise from human rater uncertainty or inter-rater variability. Sleep scoring algorithms either come as supervised classifiers that need scored samples of each state to be trained, or as unsupervised classifiers that use heuristics or structural clues in unscored data to define states. We propose a quasi-supervised classifier that models observations in an unsupervised manner but mimics a human rater wherever training scores are available. EEG, EMG, and EOG features were extracted in 30s epochs from human-scored polysomnograms recorded from 42 healthy human subjects (18-79 years) and archived in an anonymized, publicly accessible database. Hypnograms were modified so that: 1. Some states are scored but not others; 2. Samples of all states are scored but not for transitional epochs; and 3. Two raters with 67% agreement are simulated. A framework for quasi-supervised classification was devised in which unsupervised statistical models-specifically Gaussian mixtures and hidden Markov models--are estimated from unlabeled training data, but the training samples are augmented with variables whose values depend on available scores. Classifiers were fitted to signal features incorporating partial scores, and used to predict scores for complete recordings. Performance was assessed using Cohens Κ statistic. The quasi-supervised classifier performed significantly better than an unsupervised model and sometimes as well as a completely supervised model despite receiving only partial scores. The quasi-supervised algorithm addresses the need for classifiers that mimic scoring patterns of human raters while compensating for their limitations.

Naive scoring of human sleep based on a hidden Markov model of the electroencephalogram.

Clinical sleep scoring involves tedious visual review of overnight polysomnograms by a human expert. Many attempts have been made to automate the process by training computer algorithms such as support vector machines and hidden Markov models (HMMs) to replicate human scoring. Such supervised classifiers are typically trained on scored data and then validated on scored out-of-sample data. Here we describe a methodology based on HMMs for scoring an overnight sleep recording without the benefit of a trained initial model. The number of states in the data is not known a priori and is optimized using a Bayes information criterion. When tested on a 22-subject database, this unsupervised classifier agreed well with human scores (mean of Cohens kappa > 0.7). The HMM also outperformed other unsupervised classifiers (Gaussian mixture models, k-means, and linkage trees), that are capable of naive classification but do not model dynamics, by a significant margin (p <; 0.05).

Robust genetic programming-based detection of atrial fibrillation using RR intervals

In this study, two variants of genetic programming, namely linear genetic programming (LGP) and multi-expression programming (MEP) are utilized to detect atrial fibrillation (AF) episodes. LGP- and MEP-based models are derived to classify samples of AF and Normal episodes based on the analysis of RR interval signals. A weighted least-squares (WLS) regression analysis is performed using the same features and data sets to benchmark the models. Another important contribution of this paper is identification of the effective time domain features of heart rate variability (HRV) signals upon an improved forward floating selection (IFFS) analysis. The models are developed using MIT-BIH arrhythmia database. The diagnostic performances of the LGP and MEP classifiers are evaluated through receiver operating characteristics (ROC) analysis. The results indicate that the LGP and MEP models are able to diagnose the AF arrhythmia with an acceptable high accuracy. The proposed models have significantly better diagnosis performances than the regression and several models found in the literature.

Validation of a closed-loop sensory stimulation technique for selective sleep restriction in mice

Experimental manipulation of sleep in rodents is an important tool for analyzing the mechanisms of sleep and related disorders in humans. Sleep restriction systems have relied in the past on manual sensory stimulation and recently on more sophisticated automated means of delivering the same. The ability to monitor and track behavior through the electroencephalogram (EEG) and other modalities provides the opportunity to implement more selective sleep restriction that is targeted at particular stages of sleep with flexible control over their amount, duration, and timing. In this paper we characterize the performance of a novel tactile stimulation system operating in closed-loop to interrupt rapid eye movement (REM) sleep in mice when it is detected in real time from the EEG. Acute experiments in four wild-type mice over six hours showed that a reduction of over 50% of REM sleep was feasible without affecting non-REM (NREM) sleep. The animals remained responsive to the stimulus over the six hour duration of the experiment.

Noninvasive dissection of mouse sleep using a piezoelectric motion sensor

BACKGROUND Changes in autonomic control cause regular breathing during NREM sleep to fluctuate during REM. Piezoelectric cage-floor sensors have been used to successfully discriminate sleep and wake states in mice based on signal features related to respiration and other movements. This study presents a classifier for noninvasively classifying REM and NREM using a piezoelectric sensor. NEW METHOD Vigilance state was scored manually in 4-s epochs for 24-h EEG/EMG recordings in 20 mice. An unsupervised classifier clustered piezoelectric signal features quantifying movement and respiration into three states: one active; and two inactive with regular and irregular breathing, respectively. These states were hypothesized to correspond to Wake, NREM, and REM, respectively. States predicted by the classifier were compared against manual EEG/EMG scores to test this hypothesis. RESULTS Using only piezoelectric signal features, an unsupervised classifier distinguished Wake with high (89% sensitivity, 96% specificity) and REM with moderate (73% sensitivity, 75% specificity) accuracy, but NREM with poor sensitivity (51%) and high specificity (96%). The classifier sometimes confused light NREM sleep - characterized by irregular breathing and moderate delta EEG power - with REM. A supervised classifier improved sensitivities to 90, 81, and 67% and all specificities to over 90% for Wake, NREM, and REM, respectively. COMPARISON WITH EXISTING METHODS Unlike most actigraphic techniques, which only differentiate sleep from wake, the proposed piezoelectric method further dissects sleep based on breathing regularity into states strongly correlated with REM and NREM. CONCLUSIONS This approach could facilitate large-sample screening for genes influencing different sleep traits, besides drug studies or other manipulations.

SegWay: A simple framework for unsupervised sleep segmentation in experimental EEG recordings

Sleep analysis in animal models typically involves recording an electroencephalogram (EEG) and electromyogram (EMG) and scoring vigilance state in brief epochs of data as Wake, REM (rapid eye movement sleep) or NREM (non-REM) either manually or using a computer algorithm. Computerized methods usually estimate features from each epoch like the spectral power associated with distinctive cortical rhythms and dissect the feature space into regions associated with different states by applying thresholds, or by using supervised/unsupervised statistical classifiers; but there are some factors to consider when using them:• Most classifiers require scored sample data, elaborate heuristics or computational steps not easily reproduced by the average sleep researcher, who is the targeted end user.• Even when prediction is reasonably accurate, small errors can lead to large discrepancies in estimates of important sleep metrics such as the number of bouts or their duration.• As we show here, besides partitioning the feature space by vigilance state, modeling transitions between the states can give more accurate scores and metrics. An unsupervised sleep segmentation framework, “SegWay”, is demonstrated by applying the algorithm step-by-step to unlabeled EEG recordings in mice. The accuracy of sleep scoring and estimation of sleep metrics is validated against manual scores.

Unsupervised Estimation of Mouse Sleep Scores and Dynamics Using a Graphical Model of Electrophysiological Measurements

The proportion, number of bouts, and mean bout duration of different vigilance states (Wake, NREM, REM) are useful indices of dynamics in experimental sleep research. These metrics are estimated by first scoring state, sometimes using an algorithm, based on electrophysiological measurements such as the electroencephalogram (EEG) and electromyogram (EMG), and computing their values from the score sequence. Isolated errors in the scores can lead to large discrepancies in the estimated sleep metrics. But most algorithms score sleep by classifying the state from EEG/EMG features independently in each time epoch without considering the dynamics across epochs, which could provide contextual information. The objective here is to improve estimation of sleep metrics by fitting a probabilistic dynamical model to mouse EEG/EMG data and then predicting the metrics from the model parameters. Hidden Markov models (HMMs) with multivariate Gaussian observations and Markov state transitions were fitted to unlabeled 24-h EEG/EMG feature time series from 20 mice to model transitions between the latent vigilance states; a similar model with unbiased transition probabilities served as a reference. Sleep metrics predicted from the HMM parameters did not deviate significantly from manual estimates except for rapid eye movement sleep (REM) ([Formula: see text]; Wilcoxon signed-rank test). Changes in value from Light to Dark conditions correlated well with manually estimated differences (Spearmans rho 0.43-0.84) except for REM. HMMs also scored vigilance state with over 90% accuracy. HMMs of EEG/EMG features can therefore characterize sleep dynamics from EEG/EMG measurements, a prerequisite for characterizing the effects of perturbation in sleep monitoring and control applications.

Experimental and computational analysis of mouse sleep-wake dynamics

Materials and methods In this study we explore the utility of a noninvasive method based on the signal from a piezoelectric sensor on the cage floor for scoring sleep-wake behavior in mice. It was previously demonstrated that the piezo signal can accurately discriminate sleep from wake activity; however, this was verified mostly by visual observation. Here we perform a more objective validation by correlating piezo measurements with EMG activity, which is dramatically suppressed during sleep. Furthermore, the piezo sensor is sensitive to respiration-related thoracic movements. Since breathing is relatively irregular in REM sleep compared to non-REM, we extract piezo features that reflect breathing regularity to try to distinguish between these sleep states.

A mathematical model of ultradian sleep-wake regulation in rodents

Rodent models are widely used for the experimental analysis of sleep. While this is motivated by similarities in brain circuitry and electrophysiological rhythms, unlike the circadian sleep-wake cycle in humans, rodent sleep is polyphasic, containing multiple bouts of sleep and wake minutes to hours in duration over the course of a day. Each sleep bout is punctuated by several brief arousals several seconds to minutes long. Physiologically motivated mathematical models replicate the shorter timescale of arousal within sleep, but not the longer one representing prolonged wakefulness. Here, we adapt a previously published “flip-flop” model of human sleep to capture the ultradian alternation of sleep and wakefulness in mice on the longer timescale. The resulting model reproduces both the mean durations of alternating sleep and wake bouts as well as the circadian trends in their bout durations documented in our experiments on mice.