Stéphane Bibian

Quantifying cortical activity during general anesthesia using wavelet analysis

This paper reports on a novel method for quantifying the cortical activity of a patient during general anesthesia as a surrogate measure of the patients level of consciousness. The proposed technique is based on the analysis of a single-channel (frontal) electroencephalogram (EEG) signal using stationary wavelet transform (SWT). The wavelet coefficients calculated from the EEG are pooled into a statistical representation, which is then compared to two well-defined states: the awake state with normal EEG activity, and the isoelectric state with maximal cortical depression. The resulting index, referred to as the wavelet-based anesthetic value for central nervous system monitoring (WAV/sub CNS/), quantifies the depth of consciousness between these two extremes. To validate the proposed technique, we present a clinical study which explores the advantages of the WAV/sub CNS/ in comparison with the BIS monitor (Aspect Medical Systems, MA), currently a reference in consciousness monitoring. Results show that the WAV/sub CNS/ and BIS are well correlated (r=0.969) during periods of steady-state despite fundamental algorithmic differences. However, in terms of dynamic behavior, the WAV/sub CNS/ offers faster tracking of transitory changes at induction and emergence, with an average lead of 15-30 s. Furthermore, and conversely to the BIS, the WAV/sub CNS/ regains its preinduction baseline value when patients are responding to verbal command after emergence from anesthesia. We conclude that the proposed analysis technique is an attractive alternative to BIS monitoring. In addition, we show that the WAV/sub CNS/ dynamics can be modeled as a linear time invariant transfer function. This index is, therefore, well suited for use as a feedback sensor in advisory systems, closed-loop control schemes, and for the identification of the pharmacodynamic models of anesthetic drugs.

Time delay compensation of digital control for DC switchmode power supplies using prediction techniques

One of the major disadvantages of digital control is the limited control loop bandwidth due to the inherent time delay introduced by the zero-order-hold effect and the computational time delay. To alleviate this problem, two practical and straightforward predictive schemes based on linear extrapolation are proposed. With the proposed schemes, the computational time delay of the control loop is compensated and the control loop bandwidth is increased. It is shown that, using the proposed techniques, the control loop bandwidth can be increased up to two times that of the conventional digital control loop. Also, the computational overhead needed to implement these techniques is kept to the minimum. A lab prototype system of a 1 kV full bridge DC power supply was set up for the proof of concept. The prototype system operated at 10 kHz and was controlled by a TI TMS320F240 DSP (20-MHz 16-bit fixed-point). Simulation and experimental results prove the validity of the proposed techniques.

Digital control with improved performance for boost power factor correction circuits

In this paper, an approach for the design of a digital controller for a PFC pre-regulator is proposed. The controller is modified to account for large control periods and computational delays, and can therefore be implemented on processors with few available computational resources. Results show that, even with a very low control rate, system specifications can be met using the proposed technique.

A wavelet based de-noising technique for ocular artifact correction of the electroencephalogram

This paper investigates a wavelet based denoising of the electroencephalogram (EEG) signal to correct for the presence of the ocular artifact (OA). The. proposed technique is based on an over-complete wavelet expansion of the EEG as follows: i) a stationary wavelet transform (SWT) is applied to the corrupted EEG; ii) the thresholding of the coefficients in the lower frequency bands is performed; iii) the de-noised signal is reconstructed. This paper demonstrates the potential of the proposed technique for successful OA correction. The advantage over conventional methods is that there is no need for the recording of the electrooculogram (EOG) signal itself. The approach works both for eye blinks and eye movements. Hence, there is no need to discriminate between different artifacts. To allow for a proper comparison, the contaminated EEG signals as well as the corrected signals are presented together with their corresponding power spectra.

Introduction to Automated Drug Delivery in Clinical Anesthesia

Control technology has been applied to a wide variety of industrial and domestic applications to improve performance, safety and efficiency. Anesthesia, a critical aspect of clinical and emergency medicine, has not yet benefited from such technological advances. The lack of dedicated feedback sensors, and the large inter- and intra-patient variability in terms of patients’ response to drug administration, have seriously limited the effectiveness and reliability of closed-loop controllers in clinical settings. However, recent advances in sensing devices, along with robust nonlinear control theories, have generated new hopes that the gap between manual and automated control of anesthesia can finally be bridged. This paper addresses the pharmacological principles of clinical anesthesia in a context appropriate for control engineers. Concepts and terminology, monitoring issues, as well as drug dose vs. response relationships, are covered.

Dynamic behavior of BIS, M-entropy and neuroSENSE brain function monitors

ObjectivesThe objective of this paper is to assess the suitability of brain function monitors for use in closed-loop anesthesia or sedation delivery. In such systems, monitors used as feedback sensors should preferably be Linear and Time Invariant (LTI) in order to limit sensor-induced uncertainty which can cause degraded performance. In this paper, we evaluate the suitability of the BIS A2000 (Aspect Medical Systems, MA), the M-Entropy Monitor (GE HealthCare), and the NeuroSENSE Monitor (NeuroWave Systems Inc, OH), by verifying whether their dynamic behavior conforms to the LTI hypothesis.MethodsWe subjected each monitor to two different composite EEG signals containing step-wise changes in cortical activity. The first signal was used to identify Linear Time-Invariant (LTI) models that mathematically capture the dynamic behavior of each monitor. The identification of the model parameters was carried out using standard Recursive Least Squares (RLS) estimation. The second signal was used to assess the performance of the model, by comparing the output of the monitor to the simulated output predicted by the model.ResultsWhile a LTI model was successfully derived for each monitor using the first signal, only the model derived for NeuroSENSE was capable to reliably predict the monitor output for the second input signals. This indicates that some algorithmic processes within the BIS A2000 and M-Entropy are non-linear and/or time variant.ConclusionWhile both BIS and M-Entropy monitors have been successfully used in closed-loop systems, we were unable to obtain a unique LTI model that could capture their dynamic behavior during step-wise changes in cortical activity. The uncertainty in their output during rapid changes in cortical activity impose limitations in the ability of the controller to compensate for rapid changes in patients’ cortical state, and pose additional difficulties in being able to provide mathematically proof for the stability of the overall closed-loop system. Conversely, the NeuroSENSE dynamic behavior can be fully captured by a linear and time invariant transfer function, which makes it better suited for closed-loop applications.

High performance predictive dead-beat digital controller for DC power supplies

In this paper, a digital control technique based on a predictive dead-beat control concept is developed for switch-mode DC/DC power supply applications. As compared to a conventional digital controller, the proposed technique offers much improved control performance. In addition, the design procedure is simplified and the computational resources required for the calculation of the control algorithm are significantly reduced.

A simple prediction technique for the compensation of digital control time delay in DC switchmode power supplies

Two predictive schemes based on a linear extrapolation technique are developed to compensate the sampling time delay present in digital control, thus increasing the bandwidth of the control loop. Characterized by a low computational effort, these schemes are perfectly suited for fast systems such as high-performance DC switchmode power supplies. Results with and without the proposed prediction schemes are provided for comparison.

PATIENT VARIABILITY AND UNCERTAINTY QUANTIFICATION IN ANESTHESIA: PART II – PKPD UNCERTAINTY

Abstract The outcome of any surgery is particularly dependent on the adequate delivery of anesthetic drugs. Not surprisingly clinical researchers have been trying to automatize their delivery in order to provide anesthesiologists with titration tools that can target the exact needs of each individual patient. As compared to todays population-normed drug delivery strategy close loop drug delivery systems would provide patients with customized pharmacological action, thereby improving surgery outcome. While some anesthesia close loop designs have already shown promising results within controlled clinical protocols, the pharmacological variability that exists between patients needs to be addressed within a mathematical framework to prove the stability of the control laws, and gain faster and wider acceptance of these systems by the clinical community and regulatory committees. This paper is the second of a series of 2 papers addressing the issue of pharmacological variability and PKPD uncertainty. In the first paper, we presented our own drug modeling approach, which we applied towards the identification of 44 adult patient models for propofol, a central nervous system depressant drug. The individual patient models have shown a large inter-patient variability. In this paper, we further expand on our previous result in order to derive an uncertainty metrics that can be used in the control design to ensure stability and assess performances.

Quantifying uncertainty bounds in anesthetic PKPD models

A major challenge faced when designing controllers to automate anesthetic drug delivery is the large variability that exists between and within patients. This intra- and inter-patient variability have been reported to lead to instability. Hence, defining and quantifying uncertainty bounds provides a mean to validate the control design, ensure its stability and assess performance. In this work, the intra- and inter-patient variability measured from thiopental induction data is used to define uncertainty bounds. It is shown that these bounds can be reduced by up to 40% when using a patient-specific model as compared to a population-normed model. It is also shown that identifying only the overall static gain of the patient system already decreases significantly this uncertainty.