Efficient accommodation of local minima in watershed model calibration

The Gauss-Marquardt-Levenberg (GML) method of computer-based parameter estimation, in common with other gradient-based approaches, suffers from the drawback that it may become trapped in local objective function minima, and thus report optimized parameter values that are not, in fact, optimized at all. This can seriously degrade its utility in the calibration of watershed models where local optima abound. Nevertheless, the method also has advantages, chief among these being its model-run efficiency, and its ability to report useful information on parameter sensitivities and covariances as a by-product of its use. It is also easily adapted to maintain this efficiency in the face of potential numerical problems (that adversely affect all parameter estimation methodologies) caused by parameter insensitivity and/or parameter correlation. The present paper presents two algorithmic enhancements to the GML method that retain its strengths, but which overcome its weaknesses in the face of local optima. Using the first of these methods an intelligent search for better parameter sets is conducted in parameter subspaces of decreasing dimensionality when progress of the parameter estimation process is slowed either by numerical instability incurred through problem ill-posedness, or when a local objective function minimum is encountered. The second methodology minimizes the chance of successive GML parameter estimation runs finding the same objective function minimum by starting successive runs at points that are maximally removed from previous parameter trajectories. As well as enhancing the ability of a GML-based method to find the global objective function minimum, the latter technique can also be used to find the locations of many non-global optima (should they exist) in parameter space. This can provide a useful means of inquiring into the well-posedness of a parameter estimation problem, and for detecting the presence of bimodal parameter and predictive probability distributions. The new methodologies are demonstrated by calibrating a Hydrological Simulation Program-FORTRAN (HSPF) model against a time series of daily flows. Comparison with the SCE-UA method in this calibration context demonstrates a high level of comparative model run efficiency for the new method. (c) 2006 Elsevier B.V. All rights reserved.

Clinical evaluation of the WOMAC 3.0 OA Index in numeric rating scale format using a computerized touch screen version

Background: The Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index is a previously described self-administered questionnaire covering three domains: pain, stiffness and function. It has been validated in patients with osteoarthritis (OA) of the hip or knee in a paper-based format. Aim: To validate the WOMAC 3.0 using a numerical rating scale in a computerized touch screen format allowing immediate evaluation of the questionnaire. In the computed version cartoons, written and audio instruments were included in order facilitate application. Methods: Fifty patients, demographically balanced, with radiographically proven primary hip or knee OA completed the classical paper and the new computerized WOMAC version. Subjects were randomized either to paper format or computerized format first to balance possible order effects, Results: The intra-class correlation coefficients for pain, stiffness and function values were 0.915, 0.745 and 0.940, respectively. The Spearman correlation coefficients for pain, stiffness and function were 0.88, 0.77 and 0.87, respectively. Conclusion: These data indicate that the computerized WOMAC OA index 3.0 is comparable to the paper WOMAC in all three dimensions. The computerized version would allow physicians to get an immediate result and if present a direct comparison with a previous exam. (C) 2002 OsteoArthritis Research Society International. Published by Elsevier Science Ltd. All rights reserved.

A genetic estimation algorithm for parameters of stochastic ordinary differential equations

A generic method for the estimation of parameters for Stochastic Ordinary Differential Equations (SODEs) is introduced and developed. This algorithm, called the GePERs method, utilises a genetic optimisation algorithm to minimise a stochastic objective function based on the Kolmogorov-Smirnov statistic. Numerical simulations are utilised to form the KS statistic. Further, the examination of some of the factors that improve the precision of the estimates is conducted. This method is used to estimate parameters of diffusion equations and jump-diffusion equations. It is also applied to the problem of model selection for the Queensland electricity market. (C) 2003 Elsevier B.V. All rights reserved.

Population-based continuous optimization, probabilistic modelling and mean shift

Evolutionary algorithms perform optimization using a population of sample solution points. An interesting development has been to view population-based optimization as the process of evolving an explicit, probabilistic model of the search space. This paper investigates a formal basis for continuous, population-based optimization in terms of a stochastic gradient descent on the Kullback-Leibler divergence between the model probability density and the objective function, represented as an unknown density of assumed form. This leads to an update rule that is related and compared with previous theoretical work, a continuous version of the population-based incremental learning algorithm, and the generalized mean shift clustering framework. Experimental results are presented that demonstrate the dynamics of the new algorithm on a set of simple test problems.

Objectives for multiple-species conservation planning

The first step in conservation planning is to identify objectives. Most stated objectives for conservation, such as to maximize biodiversity outcomes, are too vague to be useful within a decision-making framework. One way to clarify the issue is to define objectives in terms of the risk of extinction for multiple species. Although the assessment of extinction risk for single species is common, few researchers have formulated an objective function that combines the extinction risks of multiple species. We sought to translate the broad goal of maximizing the viability of species into explicit objectives for use in a decision-theoretic approach to conservation planning. We formulated several objective functions based on extinction risk across many species and illustrated the differences between these objectives with simple examples. Each objective function was the mathematical representation of an approach to conservation and emphasized different levels of threat Our objectives included minimizing the joint probability of one or more extinctions, minimizing the expected number of extinctions, and minimizing the increase in risk of extinction from the best-case scenario. With objective functions based on joint probabilities of extinction across species, any correlations in extinction probabilities bad to be known or the resultant decisions were potentially misleading. Additive objectives, such as the expected number of extinctions, did not produce the same anomalies. We demonstrated that the choice of objective function is central to the decision-making process because alternative objective functions can lead to a different ranking of management options. Therefore, decision makers need to think carefully in selecting and defining their conservation goals.

The cross-entropy method for continuous multi-extremal optimization

In recent years, the cross-entropy method has been successfully applied to a wide range of discrete optimization tasks. In this paper we consider the cross-entropy method in the context of continuous optimization. We demonstrate the effectiveness of the cross-entropy method for solving difficult continuous multi-extremal optimization problems, including those with non-linear constraints.

A fuzzy approach to texture segmentation

The texture segmentation techniques are diversified by the existence of several approaches. In this paper, we propose fuzzy features for the segmentation of texture image. For this purpose, a membership function is constructed to represent the effect of the neighboring pixels on the current pixel in a window. Using these membership function values, we find a feature by weighted average method for the current pixel. This is repeated for all pixels in the window treating each time one pixel as the current pixel. Using these fuzzy based features, we derive three descriptors such as maximum, entropy, and energy for each window. To segment the texture image, the modified mountain clustering that is unsupervised and fuzzy c-means clustering have been used. The performance of the proposed features is compared with that of fractal features.

Assessment of regional long-axis function during dobutamine echocardiography

Echocardiographic analysis of regional left ventricular function is based upon the assessment of radial motion. Long-axis motion is an important contributor to overall function. but has been difficult to evaluate clinically until the recent development of tissue Doppler techniques. We sought to compare the standard visual assessment of radial motion with quantitative tissue Doppler measurement of peak systolic velocity. timing and strain rate (SRI) in 104 patients with known or suspected coronary artery disease undergoing dobutamine stress echocardiography (DbE). A standard DbE protocol was used with colour tissue Doppler images acquired in digital cine-loop format. peak systolic velocity (PSV), time to peak velocity (TPV) and SRI were assessed off-line by an independent operator. Wall motion was assessed by an experienced reader. Mean PSV, TPV and SRI values were compared with wall motion and the presence of coronary artery disease by angiography. A further analysis included assessing the extent of jeopardized myocardium by comparing average values of PSV, TPV and SRI against the previously validated angiographic score. Segments identified as having normal and abnormal radial wall motion showed significant differences in mean PSV (7.9 +/- 3.8 and 5.9 +/- 3.3 cm/s respectively; P < 0.001), TPV (84 40 and 95 +/- 48 ms respectively; P = 0.005) and SRI (- 1.45 +/- 0.5 and - 1.1 +/- 0.9 s(-1) respectively; P < 0.001). The presence of a stenosed subtending coronary artery was also associated with significant differences from normally perfused segments for mean PSV (8.1 3.4 compared with 5.7 +/- 3.7 cm/s; P < 0.001), TPV (78 50 compared with 92 +/- 45 ms; P < 0.001) and SRI (- 1.35 0.5 compared with - 1.20 +/- 0.4 s(-1); P = 0.05). PSV, TPV and SRI also varied significantly according to the extent of jeopardized myocardium within a vascular territory. These results suggest that peak systolic velocity, timing of contraction and SRI reflect the underlying physiological characteristics of the regional myocardium during DbE, and may potentially allow objective analysis of wall motion.

Determination of reference values for glucose tolerance, insulin tolerance, and insulin sensitivity tests in clinically normal cats

Objective-To determine reference values and test variability for glucose tolerance tests (GTT), insulin tolerance tests (ITT), and insulin sensitivity tests (IST) in cats, Animals-32 clinically normal cats. Procedure-GTT, ITT, and IST were performed on consecutive days. Tolerance intervals tie, reference values) were calculated as means +/- 2.397 SD for plasma glucose and insulin concentrations, half-life of glucose (T-1/2glucose), rate constants for glucose disappearance (K-glucose and K-itt), and insulin sensitivity index (S-l). Tests were repeated after 6 weeks in 8 cats to determine test variability. Results-Reference values for T-1/2glucose, K-glucose, and fasting plasma glucose and insulin concentrations during GTT were 45 to 74 minutes, 0.93 to 1.54 %/min, 37 to 104 mg/dl, and 2.8 to 20.6 muU/ml, respectively. Mean values did not differ between the 2 tests. Coefficients of variation for T-1/2glucose, K-glucose, and fasting plasma glucose and insulin concentrations were 20, 20, 11, and 23%, respectively. Reference values for K-itt were 1.14 to 7.3%/min, and for S-l were 0.57 to 10.99 x 10(-4) min/muU/ml. Mean values did not differ between the 2 tests performed 6 weeks apart, Coefficients of variation for K-itt and S-l were 60 and 47%, respectively. Conclusions and Clinical Relevance-GTT, ITT, and IST can be performed in cats, using standard protocols. Knowledge of reference values and test variability will enable researchers to better interpret test results for assessment of glucose tolerance, pancreatic beta -cell function, and insulin sensitivity in cats.

EMA analysis of tongue function in children with dysarthria following traumatic brain injury

Primary objective : To investigate the speed and accuracy of tongue movements exhibited by a sample of children with dysarthria following severe traumatic brain injury (TBI) during speech using electromagnetic articulography (EMA). Methods and procedures : Four children, aged between 12.75-17.17 years with dysarthria following TBI, were assessed using the AG-100 electromagnetic articulography system (Carstens Medizinelektronik). The movement trajectories of receiver coils affixed to each child's tongue were examined during consonant productions, together with a range of quantitative kinematic parameters. The children's results were individually compared against the mean values obtained by a group of eight control children (mean age of 14.67 years, SD 1.60). Main outcomes and results : All four TBI children were perceived to exhibit reduced rates of speech and increased word durations. Objective EMA analysis revealed that two of the TBI children exhibited significantly longer consonant durations compared to the control group, resulting from different underlying mechanisms relating to speed generation capabilities and distances travelled. The other two TBI children did not exhibit increased initial consonant movement durations, suggesting that the vowels and/or final consonants may have been contributing to the increased word durations. Conclusions and clinical implications : The finding of different underlying articulatory kinematic profiles has important implications for the treatment of speech rate disturbances in children with dysarthria following TBI.

Perceptual and instrumental evaluation of voice and tongue function after carotid endarterectomy

Objective: Laryngeal and tongue function was assessed in 28 patients to evaluate the presence, nature, and resolution of superior recurrent laryngeal and hypoglossal nerve damage resulting from standard open primary carotid endarterectomy (CEA). Methods. The laryngeal and tongue function in 28 patients who underwent CEA were examined prospectively with various physiologic (Aerophone II, laryngograph, tongue transducer), acoustic (Multi-Dimensional Voice Program), and perceptual speech assessments. Measures were obtained from all participants preoperatively, and at 2 weeks and at 3 months postoperatively. Results. The perceptual speech assessment indicated that the vocal quality of roughness was significantly more apparent at the 2-week postoperative assessment than preoperatively. However, by the 3-month postoperative assessment these values had returned to near preoperative levels, with no significant difference detected between preoperative and 3-month postoperative levels or between 2-week and 3-month postoperative levels. Both the instrumental assessments of laryngeal function and the acoustic assessment of vocal quality failed to identify any significant difference on any measure across the three assessment periods. Similarly, no significant impairment in tongue strength, endurance, or rate of repetitive tongue movements was detected at instrumental assessment of tongue function. Conclusions: No permanent changes to vocal or tongue function occurred in this group of participants after primary CEA. The lack of any significant long-term laryngeal or tongue dysfunction in this group suggests that the standard open CEA procedure is not associated with high rates of superior recurrent and hypoglossal nerve dysfunction, as previously believed.

Kinematic analysis of jaw function in children following traumatic brain injury

Primary objective: To investigate jaw movements in children following traumatic brain injury (TBI) during speech using electromagnetic articulography (EMA). Methods and procedures: Jaw movements of two non-dysarthric children ( aged 12.75 and 13.08 years) who had sustained a TBI were recorded using the AG-100 EMA system (Carstens Medizineletronik) during word-initial consonant productions. Mean quantitative kinematic parameters and coefficient of variation ( variability) values were calculated and individually compared to the mean values obtained by a group of six control children ( mean age 12.57 years, SD 1.52). Main outcomes and results: The two children with TBI exhibited word-initial consonant jaw movement durations that were comparable to the control children, with sub-clinical reductions in speed being offset by reduced distances. Differences were observed between the two children in jaw kinematic variability, with one child exhibiting increased variability, while the other child demonstrated reduced or comparable variability compared to the control group. Conclusions: Possible sub-clinical impairments of jaw movement for speech were exhibited by two children who had sustained a TBI, providing insight into the consequences of TBI on speech motor control development.

Recombinant human activated protein C improves pulmonary function in ovine acute lung injury resulting from smoke inhalation and sepsis

Objective: To investigate the effects of recombinant human activated protein C (rhAPC) on pulmonary function in acute lung injury (ALI) resulting from smoke inhalation in association with a bacterial challenge. Design: Prospective, randomized, controlled, experimental animal study with repeated measurements. Setting: Investigational intensive care unit at a university hospital. Subjects: Eighteen sheep (37.2 +/- 1.0 kg) were operatively prepared and randomly allocated to either the sham, control, or rhAPC group (n = 6 each). After a tracheotomy had been performed, ALI was produced in the control and rhAPC group by insufflation of 4 sets of 12 breaths of cotton smoke. Then, a 30 mL suspension of live Pseudomonas aeruginosa bacteria (containing 2-5 x 10(11) colony forming units) was instilled into the lungs according to an established protocol. The sham group received only the vehicle, i.e., 4 sets of 12 breaths of room air and instillation of 30 mL normal saline. The sheep were studied in the awake state for 24 hrs and were ventilated with 100% oxygen. RhAPC (24 mu g/kg/hr) was intravenously administered. The infusion was initiated 1 hr post-injury and lasted until the end of the experiment. The animals were resuscitated with Ringer's lactate solution to maintain constant pulmonary artery occlusion pressure. Measurements and Main Results., In comparison with nontreatment in controls, the infusion of rhAPC significantly attenuated the fall in PaO2/FiO(2) ratio (control group values were 521 +/- 22 at baseline [BL], 72 +/- 5 at 12 hrs, and 74 +/- 7 at 24 hrs, vs. rhAPC group values of 541 +/- 12 at BL, 151 +/- 29 at 12 hours [p < .05 vs. control], and 118 +/- 20 at 24 hrs), and significantly reduced the increase in pulmonary microvascular shunt fraction (Qs/Qt; control group at BL, 0.14 +/- 0.02, and at 24 hrs, 0.65 +/- 0.08; rhAPC group at BL, 0.24 +/- 0.04, and at 24 hrs, 0.45 +/- 0.02 [p < .05 vs. control]) and the increase in peak airway pressure (mbar; control group at BL, 20 +/- 1, and at 24 hrs, 36 +/- 4; rhAPC group at BL, 21 +/- 1, and at 24 hrs, 28 +/- 2 [p < .05 vs. control]). In addition, rhAPC limited the increase in lung 3-nitrotyrosine (after 24 hrs [%]: sham, 7 +/- 2; control, 17 +/- 1; rhAPC, 12 +/- 1 [p < .05 vs. control]), a reliable indicator of tissue injury. However, rhAPC failed to prevent lung edema formation. RhAPC-treated sheep showed no difference in activated clotting time or platelet count but exhibited less fibrin degradation products (1/6 animals) than did controls (4/6 animals). Conclusions. Recombinant human activated protein C attenuated ALI after smoke inhalation and bacterial challenge in sheep, without bleeding complications.

Effects of early intensive postoperative physiotherapy on limb function after tibial plateau leveling osteotomy in dogs with deficiency of the cranial cruciate ligament

Objective-To determine effects of early intensive postoperative physiotherapy on limb function in dogs after tibial plateau leveling osteotomy (TPLO) for deficiency of the cranial cruciate ligament (CCL). Animals-8 adult dogs with CCL deficiency. Procedure-After TPLO, dogs underwent a physiotherapy program 3 times/wk (physiotherapy group; n = 4) or a walking program (home-exercise group; 4). All dogs were evaluated before surgery, 1 and 10 days after surgery, and 3 and 6 weeks after surgery. Thigh circumference (TC), stifle joint flexion and extension range of motion (ROM), lameness, and weight-bearing scores were recorded. Results-Before surgery, CCL-deficient limbs had significantly reduced TC and reduced flexion and extension ROMs, compared with values for the contralateral control limb. Six weeks after TPLO, the physiotherapy group had significantly larger TC than the home-exercise group, with the difference no longer evident between the affected and nonaffected limbs. Extension and flexion ROMs were significantly greater in the physiotherapy group, compared with values for the home-exercise group, 3 and 6 weeks after surgery. Six weeks after surgery, the difference in flexion and extension ROMs was no longer evident between the affected and nonaffected limbs in the physiotherapy group. Both groups had improvements for lameness and weight-bearing scores over time, but no difference was found between the 2 groups. Conclusions and Clinical Relevance-After TPLO in CCL-deficient dogs, early physiotherapy intervention should be considered as part of the postoperative management to prevent muscle atrophy, build muscle mass and strength, and increase stifle joint flexion and extension ROMs.