Biomedical subjects

P S Crooke

Publications and source records attributed to P S Crooke.

13 recordsLinked to original sources

PubMed2006

Hybrid grammar-based approach to nonlinear dynamical system identification from biological time series.

We introduce a grammar-based hybrid approach to reverse engineering nonlinear ordinary differential equation models from observed time series. This hybrid approach combines a genetic algorithm to search the space of model architectures with a Kalman filter to estimate the model parameters. Domain-specific knowledge is used in a context-free grammar to restrict the search space for the functional form of the target model. We find that the hybrid approach outperforms a pure evolutionary algorithm method, and we observe features in the evolution of the dynamical models that correspond with the emergence of favorable model components. We apply the hybrid method to both artificially generated time series and experimentally observed protein levels from subjects who received the smallpox vaccine. From the observed data, we infer a cytokine protein interaction network for an individual's response to the smallpox vaccine.

Algorithms↗

PubMed2006

Optimal phasic tracheal gas insufflation timing: an experimental and mathematical analysis.

OBJECTIVE: To investigate the modulation of CO2 clearance by changes in the duration of tracheal gas flow application during tracheal gas insufflation (TGI). DESIGN: Combination of bench studies using a commercial test lung and a commercially available intensive care ventilator and mathematical analysis using a clearance model derived from first principles. SETTING: University pulmonary research laboratory. PATIENTS: None. INTERVENTIONS: Experiments using TGI were performed on a test lung at two combinations of tidal volume and frequency. TGI was limited to part of the expiratory phase (the terminal 10-100% of expiration), and two different TGI catheter flow rates were studied. Permutations over a range of compliances, dead-space volumes, catheter flows, and TGI durations were collected. A mathematical model incorporating key ventilatory and TGI-related variables was developed to provide a first-principles theoretical foundation for interpreting the experimental results. MEASUREMENTS AND MAIN RESULTS: In the physical model, alveolar Pco2 attained a minimum value with TGI flow applied during the terminal 40-60% of the expiratory phase, a finding that was consistent over an almost eight-fold range of expiratory time constants. The mathematical model shows the same qualitative pattern as the experimental model, indicating that the observed behaviors are not an experimental artifact. CONCLUSION: The optimal duration of expiratory TGI flow application is stable over a wide range of impedance characteristics. Such stability suggests that near maximal effect of expiratory TGI could be obtained by applying TGI flow solely within the final 50% of the expiratory phase. Such uniform restriction of the application profile might both simplify technique implementation and decrease adverse consequences.

Carbon Dioxide↗

PubMed2005

Mathematical models for pressure controlled ventilation of oleic acid-injured pigs.

One-compartment, mathematical models for pressure controlled ventilation, incorporating volume dependent compliances, linear and nonlinear resistances, are constructed and compared with data obtained from healthy and (oleic acid) lung-injured pigs. Experimental data are used to find parameters in the mathematical models and were collected in two forms. Firstly, the P(e)-V curves for healthy and lung injured pigs were constructed; these data are used to compute compliance functions for each animal. Secondly, dynamic data from pressure controlled ventilation for a variety of applied pressures are used to estimate resistance parameters in the models. The models were then compared against the collected dynamic data. The best mathematical models are ones with compliance functions of the form C(V) = a + bV where a and b are constants obtained from the P(e)-V curves and the resistive pressures during inspiration change from a linear relation P(r) = RQ to a nonlinear relation P(r) = RQ(epsilon) where Q is the flow into the one-compartment lung and epsilon is a positive number. The form of the resistance terms in the mathematical models indicate the possible presence of gas-liquid foams in the experimental data.

Animals↗

PubMed2001

Dynamic behavior during noninvasive ventilation: chaotic support?

Acute noninvasive ventilation is generally applied via face mask, with modified pressure support used as the initial mode to assist ventilation. Although an adequate seal can usually be obtained, leaks frequently develop between the mask and the patient's face. This leakage presents a theoretical problem, since the inspiratory phase of pressure support terminates when flow falls to a predetermined fraction of peak inspiratory flow. To explore the issue of mask leakage and machine performance, we used a mathematical model to investigate the dynamic behavior of pressure-supported noninvasive ventilation, and confirmed the predicted behavior through use of a test lung. Our mathematical and laboratory analyses indicate that even when subject effort is unvarying, pressure-support ventilation applied in the presence of an inspiratory leak proximal to the airway opening can be accompanied by marked variations in duration of the inspiratory phase and in autoPEEP. The unstable behavior was observed in the simplest plausible mathematical models, and occurred at impedance values and ventilator settings that are clinically realistic.

Critical Care↗

PubMed1998

Patient-ventilator interaction: a general model for nonpassive mechanical ventilation.

A general mathematical model for the dynamic behaviour of a single-compartment respiratory system in response to an arbitrary applied inspiratory airway pressure and arbitrary respiratory muscle activity is investigated. The model is used to compute explicit expressions for ventilation and pressure variables of clinical interest for clinician-selected and impedance-determined inputs. The outcome variables include tidal volume, end-expiratory pressure, minute ventilation, mean alveolar pressure, average pleural pressure, as well as the work performed by the ventilator and the respiratory muscles. It is also demonstrated that under suitable conditions, there is a flow reversal that can occur during inspiration.

Computer Simulation↗

PubMed1996

Theoretical interactions between ventilator settings and proximal deadspace ventilation during tracheal gas insufflation.

OBJECTIVE: To investigate the theoretical interactions between ventilator settings, tracheal gas insufflation (TGI), and alveolar ventilation. DESIGN: We derived differential equations governing compartmental volume changes in a one-compartment model of TGI-assisted ventilation and equations governing gas dilution in the airway proximal to the TGI catheter and the additional CO2 clearing ventilation arising from this dilution. This additional ventilation was called proximal ventilation. Validation was conducted in a mechanical lung analog. Model predictions for proximal ventilation were then generated over wide ranges of frequency, duty cycle, and tidal volume. RESULTS: Significant interactions were identified between ventilator settings and proximal ventilation. The persistence of end-expiratory flow from the lung decreased proximal dilution by fresh gas and thereby reduced TGI-aided proximal ventilation. Changes in end-expiratory lung flow resulting from alterations in ventilator settings were correlated inversely with proximal ventilation. CONCLUSIONS: During TGI with constant catheter flow, ventilator settings that promote end-expiratory flow of gas from the lung diminish proximal ventilation. When frequency increases, the decrease in dilution efficiency of the individual breath is partially offset by the increase in cycle number, an effect which is magnified by any concomitant decrease in inspired tidal volume. Prolongation of the duty cycle tends to decrease proximal ventilation. Increases in expiratory resistance, including those arising from the external ventilator circuit or the endotracheal tube, also impair proximal ventilation.

Carbon Dioxide↗

PubMed1994

Implications of a biphasic two-compartment model of constant flow ventilation for the clinical setting.

PURPOSE: To investigate the theoretical effects of changing frequency (f), duty cycle (D), or end-inspiratory pause length on the distribution of ventilation and compartmental pressure in a heterogeneous, two compartment pulmonary model inflated by constant flow. METHODS: Differential equations governing compartmental volume changes were derived and solved. Validation was conducted in a mechanical lung analogue with two mechanically independent compartments. Model predictions were then generated over wide ranges of f, D, or end-inspiratory pause. RESULTS: Disparity of compartmental end-expiratory pressure was identified as the primary mechanism by which changes in f, D, or pause alter the distribution of ventilation. Distribution of peak pressures was less sensitive to such changes. Compartmental ventilation was much less uniform than compartmental peak pressure. Ventilation could not be made entirely uniform by changes of f, D, or pause within the usual clinical range. CONCLUSIONS: In a linear, two compartment model of the respiratory system, disparity of compartmental end-expiratory pressures is the primary mechanism by which changes of f, D, or pause alter the distribution of ventilation during inflation with constant flow. Ventilation is less evenly distributed than peak alveolar pressure, and there are limits to the beneficial effects on the distribution of ventilation to be gained from manipulations of machine settings.

Evaluation Studies as Topic↗

PubMed1993

Comparison of mathematical and mechanical models of pressure-controlled ventilation.

Recent evidence that volume-cycled mechanical ventilation may itself produce lung injury has focused clinical attention on the pressure waveform applied to the respiratory system. There has been an increasing use of pressure-controlled ventilation (PCV), because it limits peak cycling pressure and provides a decelerating flow profile that may improve gas exchange. In this mode, however, the relationships are of machine adjustments to ventilation and alveolar pressure are not straightforward. Consequently, setting selection remains largely an empirical process. In previous work, we developed a biexponential model of PCV that provides a conceptual framework for understanding these interactions (J. Appl. Physiol. 67: 1081-1092, 1989). We tested the validity of this mathematical model in a single-compartment analogue of the respiratory system across wide ranges of clinician-set variables (frequency, duty cycle, applied pressure) and impedance conditions (inspiratory and expiratory resistance and system compliance). Our data confirm the quantitative validity of the proposed model when approximately rectilinear waves of pressure are applied and appropriate values for impedance are utilized. Despite a fixed-circuit configuration, however, resistance proved to be a function of each clinician-set variable, requiring remeasurement of system impedance as adjustments in these variables were made. With further modification, this model may provide a practical as well as a conceptual basis for understanding minute ventilation and alveolar pressure fluctuations during PCV in the clinical setting.

Air Pressure↗

PubMed1993

A general mathematical model for respiratory dynamics relevant to the clinical setting.

We have developed and validated a general mathematical model for the dynamic behavior of the single-compartment respiratory system in response to an arbitrary waveform of applied inspiratory pressure. Our general model for ventilation applies to all integrable functions of applied pressure, and it enables computation of most ventilation and pressure variables of clinical interest from clinician-selected and impedance-determined inputs readily measured or estimated at the bedside. Interactions between both phases of the ventilatory cycle are considered by assuming that deflation occurs passively from a unicompartment lung. Because this flexible model appears both capable of accurate prediction and robust to major violations of its underlying linear assumptions, it may prove to be of value in a variety of scientific, educational, and clinical settings.

Humans↗

PubMed1992

Lung mechanics and gas exchange during pressure-control ventilation in dogs. Augmentation of CO2 elimination by an intratracheal catheter.

Increased awareness of pressure-related injury to the alveolar-capillary interface has renewed interest in modes of ventilation that limit alveolar distention such as pressure-controlled ventilation (PCV). We examined respiratory system mechanics and gas exchange during PCV in six dogs. Our data conformed to the predictions of our single-compartment mathematical model of respiratory dynamics during PCV (J Appl Physiol 1989; 67:1081-92). For a fixed pressure (Pset) and inspiratory time fraction (Tl/Ttot) (15 cm H2O and 0.3, respectively), minute ventilation (VE) reached a well-defined plateau as frequency (f) increased from 10 to 50 breaths/min and tidal volume (VT) fell progressively. Concomitantly, the physiologic dead-space fraction (VD/VT) increased from 0.50 +/- 0.04 to 0.85 +/- 0.04, and arterial PCO2 (PaCO2) rose from 39 +/- 4 to 76 +/- 12 mm Hg. At a fixed combination of frequency, applied pressure, and Tl/Ttot (40 breaths/min, 15 cm H2O, and 0.3), VE did not change when we introduced fresh gas continuously from an intratracheal catheter. However, PaCO2 and VD/VT fell progressively as catheter flow increased from zero to 14 L/min (60 +/- 12 to 40 +/- 12 mm Hg and 0.83 +/- 0.03 to 0.25 +/- 0.14 mm Hg, respectively). We conclude that during PCV at a fixed Pset and Tl/Ttot increasing frequency caused VT to fall and VE to reach a plateau. Declining VT was associated with a rise in PaCO2 because of a subsequent fall in alveolar ventilation. Insufflating fresh gas by an intratracheal catheter increased alveolar ventilation and improved CO2 elimination by washing out the anatomic dead space.(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗

PubMed1989

Determinants and limits of pressure-preset ventilation: a mathematical model of pressure control.

In recent years, four square-wave modes of pressure-preset mechanical ventilation (PPV)--pressure control, pressure support, inverse ratio, and airway pressure release ventilation--have been introduced to clinical practice. Conceptually, they share important features. Yet, because there remains widespread uncertainty regarding their ventilatory characteristics, efficacy, and appropriate use, the potential range of application is only now being investigated. To construct a unifying mathematical model of PPV, we developed a system of equations for prediction of the major "outcome" variables of PPV--tidal volume, minute ventilation, auto-positive end-expiratory pressure, mean alveolar pressure, and mechanical work--from the primary clinical "inputs" from patient (resistance, compliance) and clinician (applied pressure, frequency, inspiratory time fraction). Our analysis revealed distinct bounding limits for the outcome variables of ventilation and pressure and important implications for their clinical determinants. Although simplifying assumptions were required to enable construction of this mathematical analogue of respiratory system behavior, this model provides a firm conceptual framework for understanding the physiological interactions between PPV and the patients they are intended to help.

Humans↗

PubMed1988

Semisolid state fermentation of baker's yeast in an air-fluidized bed fermentor.

In an attempt to grow microorganisms other than fungi using a solid-state fermentation process, a model system of Baker's yeast (Saccharomyces cerevisiae) was cultured in an air-fluidized bed fermentor. A semisolid potato mixture (pretreated with alpha-amylase) was used for the substrate in this highly aerated system. The growth of Baker's yeast in this air-fluidized bed process was easily controllable and very reproducible. Once feasible moisture levels and air flow rates were determined, the independent variables studied were the amount of the enzyme used for digesting the potato starch, the size of the yeast inoculum, and the concentration of the added defined medium.

Culture Media↗

PubMed1986

Relating damped oscillations to sustained limit cycles describing real and ideal batch fermentation processes.

A batch fermentation model is presented in which the specific growth rate and yield functions are chosen such that sustained oscillations in both the cell and substrate concentration occur. This phenomenon is shown to be a Hopf bifurcation in the underlying system of non-linear ordinary differential equations which comprises the model. It is shown that for oscillations in the substrate concentration to occur it is necessary for the yield term to depend on both the cell and substrate levels.

Cell Count↗