PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “model predictive control”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 163 records · Page 9Linked to original sources
PubMed2002

Bullet trajectory predicts the need for damage control: an artificial neural network model.

BACKGROUND: Effective use of damage control in trauma hinges on an early decision to use it. Bullet trajectory has never been studied as a marker for damage control. We hypothesize that this decision can be predicted by an artificial neural network (ANN) model based on the bullet trajectory and the patient's blood pressure. METHODS: A multilayer perceptron ANN predictive model was developed from a data set of 312 patients with single abdominal gunshot injuries. Input variables were the bullet path, trajectory patterns, and admission systolic pressure. The output variable was either a damage control laparotomy or intraoperative death. The best performing ANN was implemented on prospectively collected data from 34 patients. RESULTS: The model achieved a correct classification rate of 0.96 and area under the receiver operating characteristic curve of 0.94. External validation showed the model to have a sensitivity of 88% and specificity of 96%. Model implementation on the prospectively collected data had a correct classification rate of 0.91. Sensitivity analysis showed that systolic pressure, bullet path across the midline, and trajectory involving the right upper quadrant were the three most important input variables. CONCLUSION: Bullet trajectory is an important, hitherto unrecognized, factor that should be incorporated into the decision to use damage control.

Abdominal Injuries↗
PubMed2004

G-CSF and IL-8 for early diagnosis of sepsis in neonates and critically ill children - safety and cost effectiveness of a new laboratory prediction model: study protocol of a randomized controlled trial [ISRCTN91123847].

INTRODUCTION: Bacterial infection represents a serious risk in neonates and critically ill paediatric patients. Current clinical practice is characterized by frequent antibiotic treatment despite low incidence of true infection. However, some patients escape early diagnosis and progress to septic shock. Many new markers, including cytokines, have been suggested to improve decision making, but the clinical efficacy of these techniques remains uncertain. Therefore, we will test the clinical efficacy of a previously validated diagnostic strategy to reduce antibiotic usage and nosocomial infection related morbidity. METHODS: All patients admitted to the multidisciplinary neonatal and paediatric intensive care unit of a university children's hospital will be included. Patients will be allocated either to routine sepsis work up or to the intervention strategy with additional cytokine measurements. Physicians will be requested to estimate the pre-test probability of sepsis and pneumonia at initial suspicion. In the treatment arm, physicians will receive raw cytokine results, the likelihood ratio and the updated post-test probability. A high post-test probability will suggest that immediate initiation of antibiotic treatment is appropriate, whereas a low post-test probability will be supportive of watchful waiting or discontinuing prophylactic empirical therapy. Physicians may overrule the suggestions resulting from the post-test probability. CONCLUSION: This trial will ascertain the clinical efficacy of introducing new diagnostic strategies consisting of pre-test probability estimate, novel laboratory markers, and computer-generated post-test probability in infectious disease work up in critically ill newborns and children.

Anti-Bacterial Agents↗
PubMed2004

Prediction of neuropsychological impairment in multiple sclerosis: comparison of conventional magnetic resonance imaging measures of atrophy and lesion burden.

BACKGROUND: Cognition and magnetic resonance imaging correlations are well established in patients with multiple sclerosis (MS), but it is unclear whether lesion burden or atrophy accounts for most of the predictive variance. These indices have been directly compared in only a few studies. No such study included measurement of the third ventricle, which was strongly predictive of neuropsychological competence in the early literature. Furthermore, few studies accounted for the influence of age, premorbid intelligence, or depression. OBJECTIVE: To determine if conventional measures of lesion burden or atrophy predict cognitive dysfunction in MS while accounting for age, premorbid intelligence, and depression. METHODS: We studied 37 patients with MS and 27 controls matched according to demographic variables. Correlations between neuropsychological tests and the following magnetic resonance imaging indices were considered: T1 hypointense lesion volume, fluid-attenuated inversion recovery hyperintense lesion volume, third ventricle width, bicaudate ratio, and brain parenchymal fraction. Regression models predicting neuropsychological performance controlled for the effects of age, premorbid intelligence, and depression. We included only those tests discriminating patients with MS from controls. RESULTS: In each regression model, third ventricle width was the sole magnetic resonance imaging measure retained. When this variable was removed from consideration, brain parenchymal fraction was retained in all analyses. CONCLUSIONS: Brain atrophy accounts for more variance than lesion burden in predicting cognitive impairment in MS, and central atrophy in particular is strongly associated with neuropsychological morbidity. This finding may be explained in part by atrophy of the thalamus, a deep gray matter structure that mediates cognitive function via cortical and subcortical pathways. Enthusiasm for the clinical utility of third ventricle width is tempered by modest intraobserver and interobserver reliability.

Adult↗
PubMed1989

Dynamical neural network organization of the visual pursuit system.

The central nervous system is a parallel dynamical system which connects sensory input with motor output for the performance of visual tracking. This paper applies elementary control system tools to extend dynamical neural network models to the visual smooth pursuit system. Observed eye position responses to target motions and characteristics of the plant (eye muscles and orbital mechanics) place dynamical constraints on the interposed neural network controller. In the process of constructing a model for the controller, we show two previous pursuit system models, using efference copy and feedforward compensation, are equivalent from an input-output standpoint. We introduce a controller model possessing a potentially highly parallel implementation and offer an example with supporting neural firing rate data. Changes in time delays or other system dynamics are expected to lead to compensatory adaptive changes in the controller. A scheme to noninvasively simulate such changes in system dynamics was developed. Actual physiologic data of adaptive responses to increased time delay is presented as an example of the utility of this parallel controller. Compensatory changes in our parallel controller model are easily predicted. These results suggest a productive interaction between neural network modeling, neurophysiology, and control systems engineering.

Animals↗
PubMed2005

Acute ankle trauma: value of a short dedicated extremity MR imaging examination in prediction of need for treatment.

PURPOSE: To assess predictive value of a short magnetic resonance (MR) imaging examination with or instead of radiography performed in patients with acute ankle trauma to identify those who require additional treatment versus those who do not and can be discharged without further follow-up. MATERIALS AND METHODS: Informed consent was obtained from all participating patients, and the institutional review board approved the randomized controlled trial and use of data to create prediction models. In a prospective controlled trial, 197 patients with recent ankle trauma (92 women, 105 men) were randomized into two groups: those who underwent radiography and those who underwent a combination of radiography and MR imaging. Data about side of injury, trauma mechanism, and results of radiography and MR imaging were collected. Additional treatment was necessary in 109 of 197 patients after their initial hospital visit. With univariable and multivariable regression analysis, four models were created for prediction of treatment. RESULTS: In univariable analysis, age (odds ratio [OR], 1.02; 95% confidence interval: 1.00, 1.04), radiographic results (OR, 7.92; 95% confidence interval: 3.17, 19.8), and positive or uncertain results in patients who underwent MR imaging versus patients who did not (OR, 2.42; 95% confidence interval: 1.25, 4.70) were predictive of treatment. In the multivariable analysis, positive or uncertain MR imaging results (OR, 2.61; 95% confidence interval: 1.28, 5.30) contributed significantly to prediction of subsequent treatment. Negative MR imaging results did not contribute significantly (OR, 0.66; 95% confidence interval: 0.27, 1.61). CONCLUSION: A limited MR imaging examination in initial evaluation of acute ankle injury with radiography has additional predictive value in identification of patients who need treatment but does not add significant information in identification of those who can be discharged without further follow-up. A limited MR imaging examination cannot replace radiography for prediction of need for additional treatment.

Adult↗
PubMed2005

Temporal sequence learning, prediction, and control: a review of different models and their relation to biological mechanisms.

In this review, we compare methods for temporal sequence learning (TSL) across the disciplines machine-control, classical conditioning, neuronal models for TSL as well as spike-timing-dependent plasticity (STDP). This review introduces the most influential models and focuses on two questions: To what degree are reward-based (e.g., TD learning) and correlation-based (Hebbian) learning related? and How do the different models correspond to possibly underlying biological mechanisms of synaptic plasticity? We first compare the different models in an open-loop condition, where behavioral feedback does not alter the learning. Here we observe that reward-based and correlation-based learning are indeed very similar. Machine control is then used to introduce the problem of closed-loop control (e.g., actor-critic architectures). Here the problem of evaluative (rewards) versus nonevaluative (correlations) feedback from the environment will be discussed, showing that both learning approaches are fundamentally different in the closed-loop condition. In trying to answer the second question, we compare neuronal versions of the different learning architectures to the anatomy of the involved brain structures (basal-ganglia, thalamus, and cortex) and the molecular biophysics of glutamatergic and dopaminergic synapses. Finally, we discuss the different algorithms used to model STDP and compare them to reward-based learning rules. Certain similarities are found in spite of the strongly different timescales. Here we focus on the biophysics of the different calcium-release mechanisms known to be involved in STDP.

Forecasting↗
PubMed1998

The culture of caregiving and aggression in psychiatric settings.

Organizational culture was used as a general framework for a causal modeling study examining the relationships of work environment with staff beliefs about caregiving and aggression. Specifically, staff who perceived the organization as emphasizing control at the expense of growth and relationship issues would be more controlling in their approach to patients and perceive more aggression. A sample of 162 nursing staff participated from three hospitals. The results indicated that 10% of Aggression was predicted by two model variables indicating control. During exploratory model testing however, age, sex, socioeconomic status and satisfaction with the hospital were all predictive of a controlling belief about caretaking or aggression.

Adult↗
PubMed1998

Principal component ANN for modelling and control of baker's yeast production.

Modelling of baker's yeast production by the principal component based artificial neural networks (ANN) is presented. The models are derived for their application in adaptive control of fermentation by the internal model control (IMC) method. Modelling data are from industrial production in 40 m3 deep jet bioreactor and from computer simulations. The modelling effort is focused on selection of ANN structure and model verification. Principal component analysis of process variables results in projection of patterns to a space of low dimension, which enables determination of ANN structure, removes data colinearity and random components of measurement signals, and model degradation by over-training is eliminated. In view of IMC application, the models for prediction of the controlled variable (ethanol partial pressure) and the inverse model for manipulative variable (molasses feed rate) are determined. The models are tested for their predictability in the time horizon from 1 to 20 min. ANN models are derived with average relative errors for untrained patterns in the range from 1 to 10%.

Bioreactors↗
PubMed1983

A mathematical analysis of the myogenic hypothesis with special reference to autoregulation of renal blood flow.

To test the hypothesis that autoregulation of renal blood flow could result from myogenic regulation of arterial/arteriolar wall tension, we have explored a model based on the assumptions that (1) each preglomerular vessel segment reacts to a change in transmural pressure by altering its internal radius until the initial change in wall tension is reduced by a gain factor, (2) postglomerular structural resistance remains unchanged, (3) extravascular tissue pressure equals intrarenal venous pressure, and (4) the renal vascular system can be represented by one unbranched tube. General equations were obtained for flow and segmental radii and pressure as functions of aortic pressure. With a gain factor of 1 and a glomerular capillary pressure of 50% of aortic pressure under control conditions, the model predictions agree well with experimental data in dogs. Increasing aortic pressure from about 60% of control level causes only slight increase of blood flow. A rise in tissue pressure up to 40% of aortic pressure causes only moderate reduction. Changes in vessel radii begin in proximal vessel segments and spread distally toward glomerulus at increasing changes in aortic and tissue pressures from their control levels. Glomerular capillary pressure is autoregulated in proportion to blood flow. The degree of autoregulation is only moderately dependent on the gain factor: A moderate impairment caused by reducing the gain factor from 1 to 0.7 may be compensated by locating the myogenically responsive wall layer a distance 0.2 times the internal radius from the vessel lumen. "Superautoregulation," i.e., a rise in flow at reduced aortic pressure, is not possible. An upper limit of autoregulation is obtained only with the additional assumption of a fall in contractile force at extreme shortening of the muscle fibers. No definitive biological proof has yet been provided for a segmental wall tension-regulating mechanism in the preglomerular vessels, and obviously its existence cannot be proved by a mathematical model. However, if such a mechanism does exist, it can explain most of the renal resistance changes at varying arterial and intrarenal pressures, as well as the observed autoregulation of terminal interlobular arterial pressure.

Aorta↗
PubMed1997

Kinetics of Dissolution of beta-Tricalcium Phosphate

The rate of dissolution of beta-tricalcium phosphate (beta-Ca3 (PO4 )2 ; beta-TCP) has been measured in the solution system Ca(OH)2 -H3 PO4 -NaOH-HNO3 -H2 O. The effects of different parameters such as pH, temperature, time, and saturation have been investigated. At zero saturation, the logarithm of the dissolution rate is a linear function of the pH (log(j 0 ) = 2.02 - 0.82 pH; r 2 = 0.993; in mmol/m2 s), indicating diffusion-limited dissolution. A simple calculation of the theoretical rate of a diffusion-controlled process showed that our data are consistent with theory. Moreover, the activation energy for this process is low (E act = 3.9 kcal/mol) also suggesting that the beta-TCP dissolution is controlled by diffusion processes. At increased saturation, the initial beta-TCP dissolution rate decreases much faster than that predicted assuming a diffusion-controlled model. However, this latter model gives a good prediction of the results if it is assumed that beta-TCP dissolution is controlled by the dissolution of an interfacial layer of hydroxyapatite (Ca5 (PO4 )3 OH; HAp): log(j ) = -1.47 + 1.34 log(1 - S HAp ); r 2 = 0.959. The beta-TCP dissolution rate decreases very sharply with time. This effect increases at higher pH or saturation. Several explanations are proposed and discussed, even though none is conclusive.

Journal Article↗
PubMed2002

Pharmacokinetics of toxic chemicals in breast milk: use of PBPK models to predict infant exposure.

Factors controlling the transfer of potentially toxic chemicals in the breast milk of nursing mothers include both chemical characteristics, such as lipophilicity, and physiologic changes during lactation. Physiologically based pharmacokinetic (PBPK) models can aid in the prediction of infant exposure via breast milk. Benefits of these quantitative models include the ability to account for changing maternal physiology and transfer kinetics, as well as the chemical-specific characteristics, in order to produce more accurate estimates of neonatal risk. A recently developed PBPK model for perchlorate and iodide kinetics in the lactating and neonatal rat demonstrates the utility of PBPK modeling in predicting maternal and neonatal distribution of these two compounds. This model incorporates time-dependent changes in physiologic characteristics and includes interactions between iodide and perchlorate that alter the distribution and kinetics of iodide.

Adult↗
PubMed1999

The control of foot placement during compensatory stepping reactions: does speed of response take precedence over stability?

Rapid, reflex-like stepping movements are a prevalent and functional compensatory reaction to destabilization, however, little is known about the underlying control. In this paper, a model is developed to examine how speed and stability demands affect control of foot placement during forward and backward compensatory stepping reactions. The concept of the velocity stability margin (VSM) is introduced to characterize the degree to which the horizontal velocity of the falling body approaches biomechanical limits on the capacity to decelerate the center of mass; analogous limits on center-of-mass displacement are quantified in terms of the displacement stability margin (DSM). The model is used to predict, for any initial step characteristics, the variation in DSM and VSM that would occur as a function of changes in timing of foot placement. The VSM was found to prevail over the DSM in establishing limits of stability. Model simulations demonstrated that there typically exists a minimum swing duration that maximizes speed of response while meeting minimum requirements for stability (VSM > or = 0), as well as a slower speed of response (longer swing duration) at which stability (VSM) is maximized. Experimental data from platform-perturbation tests in 20 healthy young (22-28) and older (65-81) adults were used, in conjunction with the model, to investigate whether speed or stability takes precedence during natural behavior. Control of single-step reactions appeared to favor stability; although the model predicted that a minimally stable step (VSM = 0) could be attained by swing durations as short as 30 ms, the observed swing durations were, on average, 135 ms longer than this, and the average VSM was nearly as large (80%) as the optimally stable value predicted by the model. Control of the initial step of multiple-step reactions was distinctly different. The average swing duration was only 55 ms greater than the minimally stable value and the average VSM was 81% smaller than in the single-step reactions. This reduction in VSM is consistent with a need to execute additional steps and appears to support the validity of the model. This model may help to provide insight into the biomechanical factors that govern the neural control of compensatory stepping reactions.

Adult↗
PubMed2006

A sensorimotor map: modulating lateral interactions for anticipation and planning.

Experimental studies of reasoning and planned behavior have provided evidence that nervous systems use internal models to perform predictive motor control, imagery, inference, and planning. Classical (model-free) reinforcement learning approaches omit such a model; standard sensorimotor models account for forward and backward functions of sensorimotor dependencies but do not provide a proper neural representation on which to realize planning. We propose a sensorimotor map to represent such an internal model. The map learns a state representation similar to self-organizing maps but is inherently coupled to sensor and motor signals. Motor activations modulate the lateral connection strengths and thereby induce anticipatory shifts of the activity peak on the sensorimotor map. This mechanism encodes a model of the change of stimuli depending on the current motor activities. The activation dynamics on the map are derived from neural field models. An additional dynamic process on the sensorimotor map (derived from dynamic programming) realizes planning and emits corresponding goal-directed motor sequences, for instance, to navigate through a maze.

Action Potentials↗
PubMed2006

Frictional adhesion: A new angle on gecko attachment.

Directional arrays of branched microscopic setae constitute a dry adhesive on the toes of pad-bearing geckos, nature's supreme climbers. Geckos are easily and rapidly able to detach their toes as they climb. There are two known mechanisms of detachment: (1) on the microscale, the seta detaches when the shaft reaches a critical angle with the substrate, and (2) on the macroscale, geckos hyperextend their toes, apparently peeling like tape. This raises the question of how geckos prevent detachment while inverted on the ceiling, where body weight should cause toes to peel and setal angles to increase. Geckos use opposing feet and toes while inverted, possibly to maintain shear forces that prevent detachment of setae or peeling of toes. If detachment occurs by macroscale peeling of toes, the peel angle should monotonically decrease with applied force. In contrast, if adhesive force is limited by microscale detachment of setae at a critical angle, the toe detachment angle should be independent of applied force. We tested the hypothesis that adhesion is increased by shear force in isolated setal arrays and live gecko toes. We also tested the corollary hypotheses that (1) adhesion in toes and arrays is limited as on the microscale by a critical angle, or (2) on the macroscale by adhesive strength as predicted for adhesive tapes. We found that adhesion depended directly on shear force, and was independent of detachment angle. Therefore we reject the hypothesis that gecko toes peel like tape. The linear relation between adhesion and shear force is consistent with a critical angle of release in live gecko toes and isolated setal arrays, and also with our prior observations of single setae. We introduced a new model, frictional adhesion, for gecko pad attachment and compared it to existing models of adhesive contacts. In an analysis of clinging stability of a gecko on an inclined plane each adhesive model predicted a different force control strategy. The frictional adhesion model provides an explanation for the very low detachment forces observed in climbing geckos that does not depend on toe peeling.

Adhesiveness↗
PubMed2006

Dental attrition models predicting temporomandibular joint disease or masticatory muscle pain versus asymptomatic controls.

AIMS: To determine whether patients with temporomandibular joint disease or masticatory muscle pain can be usefully differentiated from asymptomatic controls using multifactorial classification tree models of attrition severity and/or rates. METHODS: Measures of attrition severity and rates in patients diagnosed with disc displacement (n = 52), osteoarthrosis (n = 74), or masticatory muscle pain only (n = 43) were compared against those in asymptomatic controls (n = 132). Cross-validated classification tree models were tested for fit with sensitivity, specificity, accuracy and log likelihood accountability. RESULTS: The model for identifying asymptomatic controls only required the three measures of attrition severity (anterior, mediotrusive and laterotrusive posterior) to be differentiated from the patients with a 74.2 +/- 3.8% cross-validation accuracy. This compared with cross-validation accuracies of 69.7 +/- 3.7% for differentiating disc displacement using anterior and laterotrusive attrition severity, 68.7 +/- 3.9% for differentiating disc displacement using anterior and laterotrusive attrition rates, 70.9 +/- 3.3% for differentiating osteoarthrosis using anterior attrition severity and rates, 94.6 +/- 2.1% for differentiating myofascial pain using mediotrusive and laterotrusive attrition severity, and 92.0 +/- 2.1% for differentiating myofascial pain using mediotrusive and anterior attrition rates. The myofascial pain models exceeded the > or =75% sensitivity and > or =90% specificity thresholds recommended for diagnostic tests, and the asymptomatic control model approached these thresholds. CONCLUSION: Multifactorial models using attrition severity and rates may differentiate masticatory muscle pain patients from asymptomatic controls, and have some predictive value for differentiating intracapsular temporomandibular disorder patients as well.

Adolescent↗
PubMed1993

Quantitative model of ColE1 plasmid copy number control.

Initiation of replication of the Escherichia coli plasmid ColE1 is inhibited by formation of a complex between a small plasmid RNA (RNA I) and the pre-primer for DNA synthesis (RNA II). Complex formation (and inhibition of replication) is enhanced by the plasmid-encoded Rom protein. The in vitro kinetics of complex formation were previously studied both experimentally and theoretically. The in vivo concentrations and half-lives of RNA I, RNA II and Rom protein have been measured recently. We present a dynamic model for the in vivo replication control mechanism that accounts for the measured concentration values. From the model we deduce a simple formula for the steady-state plasmid concentration. Our results agree with a previous simple steady-state analysis done by Brenner and Tomizawa, in that plasmid copy number is most strongly dependent on the per plasmid rate of RNA I synthesis. However, our model predicts other parameter dependencies that are not evident from or at variance with the previous analysis. Accordingly, we predict that plasmid copy number is greatly influenced by changes in the rate constant describing the formation of an initial unstable RNA I-RNA II complex, but is only slightly influenced by changes in the dissociation rate of this complex. Plasmid copy number per average cell volume is predicted to increase linearly with increases in the RNA II synthesis rate and with increases in the generation time of the host culture. Rom protein, which promotes conversion of the unstable RNA I-RNA II complex to a stable complex, serves to decrease copy number; however, its presence or absence does not seem to qualitatively alter the copy number control mechanism. Our model predicts the quantitative increase of plasmid copy number in rom- mutants. Several experiments are suggested to investigate the predictions of the model.

Bacterial Proteins↗
PubMed2006

Dopamine and cognitive control: the influence of spontaneous eyeblink rate, DRD4 exon III polymorphism and gender on flexibility in set-shifting.

Converging evidence suggests a modulatory role of dopamine in cognitive control. We investigated the influence of two correlates of dopaminergic activity, the spontaneous eyeblink rate and the DRD4 exon III polymorphism, and the potential impact of gender on flexibility in an attentional set-shifting paradigm. The objective of the study was to confirm previous findings of an association between high eyeblink rates and increased cognitive flexibility. These findings were replicated in 87 healthy volunteers this time using a continuous variable for eyeblink rates instead of a dichotomized variable. The interaction between eyeblink rate and DRD4 found in the previous study was lower and failed significance. Analysis of the collapsed sample of n=150 revealed a main effect of gender and an interaction of gender and eyeblink rate on cognitive control. The complete prediction model explained 26% of the total variance. These data suggest that (1) the eyeblink rate is a reliable predictor of dopamine-mediated flexibility of cognitive control and (2) it is useful to include gender as predictor in future studies of dopaminergic modulation of cognitive control.

Adolescent↗