The work to be described forms one part of a research project under the umbrella term of 'Artificial Intelligence in Medicine.' More specifically the domain of interest is Critical Care Medicine, where the are...
详细信息
The work to be described forms one part of a research project under the umbrella term of 'Artificial Intelligence in Medicine.' More specifically the domain of interest is Critical Care Medicine, where the area of concern is the care of patients who require mechanical ventilatory support. The Artificial Intelligent Respirator System (AIRS) is being developed as an aid to clinicians for the management of such patients. The system architecture and a methodology for the design, implementation and evaluation of AIRS is presented.
The use of nonparametric modeling methods for the analysis of nonlinear neurophysiological systems is discussed. The obtained models are in the form of Volterra-Wiener functionals or equivalent block-structured models...
详细信息
The use of nonparametric modeling methods for the analysis of nonlinear neurophysiological systems is discussed. The obtained models are in the form of Volterra-Wiener functionals or equivalent block-structured models (whenever appropriate). The latter are more compact representations and lend themselves to easier physiological interpretation. A particular class of block-structured models, containing a threshold-trigger operator, is presented and shown to be suitable for studies of spike generating neural systems. This class of models may offer a general framework for the study of information processing in the nervous system. Illustrative examples from actual applications are presented, and the role of nonlinear models in enhancing our understanding of some aspects of nonlinear neurophysiological function (viz., rectification, amplitude compression, nonlinear inhibition and spike encoding) is discussed.
Following measurements of plasma calcium (Ca) and 45Ca concentrations, the circadian behaviour of Ca metabolism may be viewed as the temporal expression of a self-organized system. We present a self-oscillatory model ...
详细信息
Following measurements of plasma calcium (Ca) and 45Ca concentrations, the circadian behaviour of Ca metabolism may be viewed as the temporal expression of a self-organized system. We present a self-oscillatory model M for rat Ca metabolism, based on compartmental formalism and including a second-order autocatalytic process. The main aspects of the physiological interpretation are given with special reference to the bone liquid/solid interface. Mechanism for the mineral reactions which generate the circadian interfacial dynamics is proposed. Conceptual consequences of the connection of Ca metabolism to a self-regulated system towards hormonal control are discussed.
Several interrelated methodological issues of this symposium are addressed, with the focus of formulating and quantifying kinetic models whose purpose is to answer particular quantitative questions about living system...
详细信息
Several interrelated methodological issues of this symposium are addressed, with the focus of formulating and quantifying kinetic models whose purpose is to answer particular quantitative questions about living systems. The unifying concept is kinetic experiment design, that is, decisions about what to do in the experimental laboratory to achieve this goal, and also how to do it 'best.' The overriding factor in this endeavor is the paucity of 'good-enough' data available from many types of biosystem studies. Optimal sampling schedule (OSS) design, based in information theory and asymptotic theory of mathematical statistics, represents one approach to optimizing the precision of model parameter estimates from small data sets. Some practical results of kinetic experiments performed with optimal sampling schedule designs are discussed, based on experiences in real endocrine and metabolic system studies in several animal species. Interval identifiability and parameter interval analysis methods and results are reviewed.
Qualitative simulation faces an intrinsic problem of scale: the number of limit hypotheses grows exponentially with the number of parameters approaching limits. We present a method called Time-Scale Abstraction for st...
详细信息
Qualitative simulation faces an intrinsic problem of scale: the number of limit hypotheses grows exponentially with the number of parameters approaching limits. We present a method called Time-Scale Abstraction for structuring a complex system as a hierarchy of smaller, interacting equilibrium mechanisms. Within this hierarchy, a given mechanism views a slower one as being constant, and a faster one as being instantaneous. A perturbation to a fast mechanism may be seen by a slower mechanism as a displacement of a monotonic function constraint. We demonstrate the time-scale abstraction hierachy using the interaction between the water and sodium balance mechanisms in medical physiology, an example drawn from a larger, fully implemented, program. Where the structure of a large system permits decomposition by time-scale, this abstraction method permits qualitative simulation of otherwise intractibly complex systems.
Many complex processes are extremely difficult to control automatically using conventional control algorithms, this difficulty being mainly due to nonlinearities and time varying behaviour. Fuzzy logic control theory ...
详细信息
Many complex processes are extremely difficult to control automatically using conventional control algorithms, this difficulty being mainly due to nonlinearities and time varying behaviour. Fuzzy logic control theory has however proved capable of successfully handling these problems. This is the main scope of this paper which is concerned with the design of a controller based upon this theory for a biological system comprising muscle relaxant drug administration. In deriving the necessary linguistic rules, use is made of observed responses from a model of the above mentioned system. The performance of the controlled system is further investigated and modifications are introduced in order to improve the above derived rules. Further simulation studies of the 'well' controlled process include sensitivity to parameter changes and noise-corrupted measurements.
In vivo evaluation in humans of transport and metabolism of substrates in skeletal muscle of the forearm is important but difficult. Available approaches are briefly reviewed. In this paper we first outline a general ...
详细信息
In vivo evaluation in humans of transport and metabolism of substrates in skeletal muscle of the forearm is important but difficult. Available approaches are briefly reviewed. In this paper we first outline a general modeling strategy to address this issue in the context of multiple tracer experiment design. A partly-flow compartmental model to assess the kinetics of glucose transport from a dual-tracer study is then presented. The effect of insulin is examined.
An important feature of TCA cycle models in current use for estimating isotope dilution is the assumption that no carbon enters the TCA cycle except the gluconeogenic flux from pyruvate to oxaloacetate and the flux of...
详细信息
An important feature of TCA cycle models in current use for estimating isotope dilution is the assumption that no carbon enters the TCA cycle except the gluconeogenic flux from pyruvate to oxaloacetate and the flux of acetyl CoA to citrate that resupplies two carbon units to the cycle. We have evaluated the impact of additional fluxes exchanging carbon with TCA cycle pools in the pathway between citrate and oxaloacetate (example, exchange of glutamate with 2-oxyglutarate). The findings are discussed. This work illustrates the value of algebraic models of the TCA cycle and related pathways as an aid in developing improved practical tracer methods for estimating gluconeogenesis.
In certain biomedical modeling parameter estimation applications, the collected data sets are characterized by relatively few observations. This paper reviews different estimation approaches which are applicable to th...
详细信息
In certain biomedical modeling parameter estimation applications, the collected data sets are characterized by relatively few observations. This paper reviews different estimation approaches which are applicable to the multiple output, sparse data system, nonlinear parameter estimation problem. Several of these estimation procedures are illustrated using data obtained from a pharmacokinetic study of the renal imaging agent Tc-99m DMSA.
By first using some simple linear curve fitting method, (such as spline fitting), on data following a compartmental model, direct application of linear regression can then be made to the system of differential equatio...
详细信息
By first using some simple linear curve fitting method, (such as spline fitting), on data following a compartmental model, direct application of linear regression can then be made to the system of differential equations describing this data. This allows insight about the system and information about the parameters governing the system to be obtained from tracer and tracee measurements. It simplifies both the process of determining which parameters are estimable from various measurements, as well as the estimation process itself. Since it does not rely on knowledge of a simple closed form of the solution it has the potential to make full use of data measured over very long time periods under non-equilibrium conditions. Essentially it allows introduction of sequential estimation methods of statistics. These results can then be used to predict future substrate concentration from known substrate production, or determine substrate production from concentration measurements.
暂无评论