SEARCH · PubMed Health

Results for “Optimization”

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 1,027 records · Page 57Linked to original sources

PubMed2007

Optimization for high-density cultivation of heterotrophic Chlorella based on a hybrid neural network model.

AIMS: The purpose of this study was to develop a reliable hybrid neural network (HNN) model for heterotrophic growth of Chlorella, based on which optimization for fed-batch (FB) cultivation of Chlorella may be successfully realized. METHODS AND RESULTS: Deterministic kinetic model was preliminarily developed for the optimization of FB cultivation of Chlorella. The highest biomass concentration and the maximum productivity were obtained as: 104.9 g l(-1) dry cell weight and 0.613 g l(-1) h(-1), respectively. After several cultivations had been performed, an HNN model was developed. The efficiency of biomass production was further increased by the optimization using this model. The highest biomass concentration and the maximum productivity attained was: 116.2 g l(-1) dry cell weight and 1.020 g l(-1) h(-1), respectively. CONCLUSION: The HNN model agreed well with experimental results in different cultivations. Comparison between the HNN model and the deterministic model showed that the former had better generalization ability, which made it a reliable tool in modelling and optimization. SIGNIFICANCE AND IMPACT OF THE STUDY: The high cell density and productivity of biomass obtained in this study is of significance for the commercial cultivation of Chlorella. The simple and efficient optimization strategy proposed in this paper may be employed in heterotrophic mass culture of Chlorella as well as other similar organisms.

Biomass↗

PubMed1997

The optimal pacing rate: an unpredictable parameter.

A three phase relation has been demonstrated between increasing heart rate and cardiac output at rest. Phase I with cardiac output increasing with increasing heart rate, phase II a plateau, and phase III decreasing cardiac output with any further increase in heart rate. The "optimal rate" can be defined as the heart rate at the onset of phase II. Twenty patients were studied, 13 male, mean age 60 years (range 31-71 years). All had chronic complete heart block and established DDD pacing. A maximal exercise test was performed to determine peak sinus rate. Exercise hemodynamics were measured using an ambulatory monitor (Capintec Vest), which permits measurement of relative cardiac output and relative ejection fraction. The patients were programmed to VVI pacing at a rate of 60 beats/min and performed three exercise tests at different workloads. The order of workloads was randomized and selected from a range (0, 25, 50, or 75 W) depending on fitness. After 3-minute stabilization, the VVI pacing rate was increased at 1-minute intervals until higher than peak sinus rate giving a total exercise time of 12 minutes. The "optimal rate band" was determined at each workload. The mean of this "optimal rate band" for each workload varied in a nonlinear manner. There was no correlation between "mean optimal rate" and age or the peak rate predicted by the Astrand formula. Current definitions of chronotropic incompetence are inaccurate. Are some of these people at their "optimal rate" already? The arbitrary selection of rate response curves on age related criteria may lead to an impaired hemodynamic response.

Adult↗

PubMed1998

An integrated dual sensor system automatically optimized by target rate histogram.

The use of combined sensors and advanced algorithms using different principles can improve rate performance over a single sensor system. Combinations of sensors and more sophisticated algorithms, however, invariably increase the complexity of pacemaker programming. An automatically optimized combined minute ventilation and activity DDDR pacemaker was developed to minimize repeated sensor adjustment. The device used subthreshold (below cardiac stimulation threshold) lead impedance to detect lead configuration at implantation automatically, followed by "implant management," including setting of lead polarity and initiation of DDDR pacing. Automatic sensor adaptation was achieved by programming a "target rate histogram" based on the patient's activity level and frequency of exertion, and the rate profile optimization process matched the recorded integrated sensor response to the target rate histogram profile. In nine patients implanted with the DX2 pacemakers, the implant management gave 100% accuracy in the detection of lead polarity. Rate profile optimization automatically increased the pacing rate during exercise between discharge and 3-month follow-up (hall walk: 78 +/- 3 vs 98 +/- 3 beats/min, and maximal treadmill exercise: 89 +/- 6 vs 115 +/- 5 beats/min, P < 0.001) with a significant increase in exercise duration during maximal exercise (7.18 +/- 1 min vs 9.56 +/- 2 min, P = 0.05). The accuracy of rate profile optimization versus manual programming was assessed at 1 month, and there was no significant difference between pacing rate kinetics and maximal pacing rate between the two methods of programming. In conclusion, pacemaker automaticity can be initiated at implantation and the self-optimized rate adaptive response appeared to be comparable to that derived from a manual programming procedure, which may reduce the need to perform time consuming sensor programming.

Aged↗

PubMed1999

Prediction of optimal atrioventricular delay in patients with implanted DDD pacemakers.

In patients with an implanted DDD pacemaker (PM), the atrial contribution may be interrupted by too short an atrioventricular (AV) delay, and filling time may be shortened by too long an AV delay. The AV delay at which the end of the A wave on transmitral flow coincides with complete closure of the mitral valve may be optimal. The subjects were 15 patients [70.3+/-12.3 (SD) years old] with an implanted DDD PM. Cardiac output (CO) and pulmonary capillary wedge pressure (PCWP) were measured by Swan-Ganz catheter. Transmitral flow was recorded by pulsed Doppler echocardiography. AV delay was prolonged stepwise by 25 msc. When the AV delay was set at 155+/-26 ms, the end of the A wave coincided with complete closure of the mitral valve. When the AV delay was prolonged 25, 50, 75, and 100 ms from this AV delay, the interval between the end of the A wave and complete closure of mitral the valve was prolonged 16+/-5, 39+/-6, 65+/-4 and 88+/-5 ms, respectively (r = 0.97, P<0.0001) and diastolic mitral regurgitation was observed during this period. Thus, the optimal AV delay may be predicted as follows: the slightly prolonged AV delay minus the interval between the end of the A wave and complete closure of the mitral valve. When the AV delay was set at 215 ms, there was a significant positive correlation between the predicted optimal AV delay (166+/-23 ms) and the optimal AV delay (CO: 161+/-26 msec, r = 0.93, P<0.0001, PCWP: 161+/-28 msec, r = 0.95, P<0.0001). In conclusion, optimal AV delay can be predicted by this simple formula: slightly prolonged AV delay minus the interval between end of A wave and complete closure of mitral valve at the AV delay setting.

Aged↗

PubMed2000

Preserving normal ventricular activation versus atrioventricular delay optimization during pacing: the role of intrinsic atrioventricular conduction and pacing rate.

The purpose of the study was to compare the effects of DDD pacing with optimal AV delay and AAI pacing on the systolic and diastolic performance at rest in patients with prolonged intrinsic AV conduction (first-degree AV block). We studied 17 patients (8 men, aged 69 +/- 9 years) with dual chamber pacemakers implanted for sick sinus syndrome in 15 patients and paroxysmal high degree AV block in 2 patients. Aortic flow and mitral flow were evaluated using Doppler echocardiography. Study protocol included the determination of the optimal AV delay in the DDD mode and comparison between AAI and DDD with optimal AV delay for pacing rate 70/min and 90/min. Stimulus-R interval during AAI (ARI) was 282 +/- 68 ms for rate 70/min and 330 +/- 98 ms for rate 90/min (P < 0.01). The optimal AV delay was 159 +/- 22 ms. AV delay optimization resulted in an increase of an aortic flow time velocity integral (AFTVI) of 16% +/- 9%. At rate 70/min the patients with ARI < or = 270 ms had higher AFTVI in AAI than in DDD (0.214 +/- 0.05 m vs 0.196 +/- 0.05 m, P < 0.01), while the patients with ARI > 270 ms demonstrated greater AFTVI under DDD compared to AAI (0.192 +/- 0.03 m vs 0.166 +/- 0.02 m, P < 0.01). At rate 90/min AFTVI was higher during DDD than AAI (0.183 +/- 0.03 m vs 0.162 +/- 0.03 m, P < 0.01). Mitral flow time velocity integral (MFTVI) at rate 70/min was higher in DDD than in AAI (0.189 +/- 0.05 m vs 0.173 +/- 0.05 m, P < 0.01), while at rate 90/min the difference was not significant in favor of DDD (0.149 +/- 0.05 m vs 0.158 +/- 0.04 m). The results suggest that in patients with first-degree AV block the relative impact of DDD and AAI pacing modes on the systolic performance depends on the intrinsic AV conduction time and on pacing rate.

Aged↗

PubMed2004

A fast and simple echocardiographic method of determination of the optimal atrioventricular delay in patients after biventricular stimulation.

The optimization of atrioventricular (AV) delay is known to significantly contribute to maximum cardiac performance. The aim of this study was to validate a new, fast, and simple echocardiographic method of identifying the AV delay that provides the maximum cardiac output (CO). Right heart catheterization and Doppler echocardiography of transmitral filling were performed simultaneously in 18 patients with heart failure and at least minimum functional mitral regurgitation treated with atrial synchronized biventricular pacing. CO derived from catheterization and Doppler filling parameters were measured at the predicted optimal AV delay (oAVD), the short AV delay (oAVD - 50 ms), and the long AV delay (oAVD + 28 ms on average/range, +10 ms to +50 ms) during a constant heart rate. The AV delay was regarded as optimal if the end of atrial contraction (represented by the end of A wave of transmitral filling) coincided with the beginning of ventricular contraction (heralded by the onset of the systolic component of mitral regurgitation). Prediction of the optimal AV delay included the following steps: (1) The maximum AV delay at which full ventricular capture is still preserved was found under electrocardiographic control. (2) This value, decreased by 5 to 10 ms, was designated as "the testing long AV delay," and the time interval from the end of the A wave to the onset of the systolic component of mitral regurgitation (time t1) was measured at this setting. (3) oAVD was simply calculated as "the testing long AV delay"- time t1. The CO measured at the oAVD (4.5 +/- 0.7 1. min-1) significantly exceeded those at the short AV delay (4.3 +/- 0.7 1. min-1, P < 0.01) and the long AV delay (4.4 +/- 0.8 1. min-1, P < 0.01), respectively. The method correctly determined the maximum CO in 78% of the patients. In conclusion, Doppler echocardiography enables very rapid and accurate optimization of AV synchrony in patients after the implantation of a biventricular pacemaker.

Aged↗

PubMedDate not supplied

Linking obstetric and midwifery practice with optimal outcomes.

OBJECTIVE: To compare midwifery and medical care practices and measure optimal perinatal outcomes using a new clinimetric instrument. DESIGN: Prospective descriptive cohort design. SETTING: A large, inner city obstetric service with medical and midwifery services. PARTICIPANTS: Three hundred seventy-five of 400 consecutively enrolled patients were participated (25 excluded due to extreme risk status or missing data); 92% were of minority race/ethnicity and 54% had less than a high school education. Of the 375 patients, 179 received physician care and 196 received nurse-midwife care. MAIN OUTCOME MEASURES: The Optimality Index-US was measured. Health record data were extracted and scored using the Optimality Index-US to summarize the optimality of processes and outcomes of care as well as the woman's preexisting health status. RESULTS: Midwifery patients had more optimal care processes (less use of technology and intervention) with no difference in neonatal outcomes, even when preexisting risk was taken into account. CONCLUSION: Even among moderate-risk patients, the midwifery model of care with its limited use of interventions can produce outcomes equivalent to or better than those of the biomedical model.

Adolescent↗

PubMedDate not supplied

MAROS: a decision support system for optimizing monitoring plans.

The Monitoring and Remediation Optimization System (MAROS), a decision-support software, was developed to assist in formulating cost-effective ground water long-term monitoring plans. MAROS optimizes an existing ground water monitoring program using both temporal and spatial data analyses to determine the general monitoring system category and the locations and frequency of sampling for future compliance monitoring at the site. The objective of the MAROS optimization is to minimize monitoring locations in the sampling network and reduce sampling frequency without significant loss of information, ensuring adequate future characterization of the contaminant plume. The interpretive trend analysis approach recommends the general monitoring system category for a site based on plume stability and site-specific hydrogeologic information. Plume stability is characterized using primary lines of evidence (i.e., Mann-Kendall analysis and linear regression analysis) based on concentration trends, and secondary lines of evidence based on modeling results and empirical data. The sampling optimization approach, consisting of a two-dimensional spatial sampling reduction method (Delaunay method) and a temporal sampling analysis method (Modified CES method), provides detailed sampling location and frequency results. The Delaunay method is designed to identify and eliminate redundant sampling locations without causing significant information loss in characterizing the plume. The Modified CES method determines the optimal sampling frequency for a sampling location based on the direction, magnitude, and uncertainty in its concentration trend. MAROS addresses a variety of ground water contaminants (fuels, solvents, and metals), allows import of various data formats, and is designed for continual modification of long-term monitoring plans as the plume or site conditions change over time.

Cost-Benefit Analysis↗

PubMed2005

A simplified procedure to determine the optimal rate of freezing biological systems.

The effect of several cell-level parameters on the predicted optimal cooling rate B(opt) of an arbitrary biological system has been studied using a well-defined water transport model. An extensive investigation of the water transport model revealed three key cell level parameters: reference permeability of the membrane to water L(pg), apparent activation energy E(Lp), and the ratio of the available surface area for water transport to the initial volume of intracellular water (SA/WV). We defined B(opt) as the "highest" cooling rate at which a predefined percent of the initial water volume is trapped inside the cell (values ranging from 5% to 80%) at a predefined end temperature (values ranging from -5 degrees C to -40 degrees C). Irrespective of the choice of the percent of initial water volume trapped and the end temperature, an exact and linear relationship exists between L(pg), SA/WV, and B(opt0. However, a nonlinear and inverse relationship is found between E(Lp) and B(opt). Remarkably, for a variety of biological systems a comparison of the published experimentally determined values of B(opt) agreed quite closely with numerically predicted B(opt) values when the model assumed 5% of initial water is trapped inside the cell at a temperature of -15 degrees C. This close agreement between the experimental and model predicted optimal cooling rates is used to develop a generic optimal cooling rate chart and a generic optimal cooling rate equation that greatly simplifies the prediction of the optimal rate of freezing of biological systems.

Animals↗

PubMed2005

Evaluation of a particle swarm algorithm for biomechanical optimization.

Optimization is frequently employed in biomechanics research to solve system identification problems, predict human movement, or estimate muscle or other internal forces that cannot be measured directly. Unfortunately, biomechanical optimization problems often possess multiple local minima, making it difficult to find the best solution. Furthermore, convergence in gradient-based algorithms can be affected by scaling to account for design variables with different length scales or units. In this study we evaluate a recently-developed version of the particle swarm optimization (PSO) algorithm to address these problems. The algorithm's global search capabilities were investigated using a suite of difficult analytical test problems, while its scale-independent nature was proven mathematically and verified using a biomechanical test problem. For comparison, all test problems were also solved with three off-the-shelf optimization algorithms--a global genetic algorithm (GA) and multistart gradient-based sequential quadratic programming (SQP) and quasi-Newton (BFGS) algorithms. For the analytical test problems, only the PSO algorithm was successful on the majority of the problems. When compared to previously published results for the same problems, PSO was more robust than a global simulated annealing algorithm but less robust than a different, more complex genetic algorithm. For the biomechanical test problem, only the PSO algorithm was insensitive to design variable scaling, with the GA algorithm being mildly sensitive and the SQP and BFGS algorithms being highly sensitive. The proposed PSO algorithm provides a new off-the-shelf global optimization option for difficult biomechanical problems, especially those utilizing design variables with different length scales or units.

Algorithms↗

PubMed2006

In vitro assessment of a motion-based optimization method for locating the talocrural and subtalar joint axes.

The locations of the joint axes of the ankle complex vary considerably between subjects, yet no noninvasive method with demonstrated accuracy exists for locating these axes. The moments of muscle and ground reaction forces about the joint axes are dependent on axis locations, making knowledge of these locations critical to accurate musculoskeletal modeling of the foot and ankle. The accuracy of a computational optimization method that fits a two-revolute model to measured motion was assessed using computer-generated data, a two-revolute mechanical linkage, and three lower-leg cadaver specimens. Motions were applied to cadaver specimens under axial load while bone-mounted markers attached to the tibia, talus, and calcaneus were tracked using a video-based motion analysis system. Estimates of the talocrural and subtalar axis locations were computed from motions of the calcaneus relative to the tibia using the optimization method. These axes were compared to mean helical axes computed directly from tibia, talus, and calcaneus motions. The optimization method performed well when the motions were computer-generated or measured in the mechanical linkage, with angular differences between optimization and mean helical axes ranging from 1 deg to 5 deg. In the cadaver specimens, however, these differences exceeded 20 deg. Optimization methods that locate the anatomical joint axes of the ankle complex by fitting two revolute joints to measured tibia-calcaneus motions may be limited because of problems arising from non-revolute behavior.

Algorithms↗

PubMed1999

Robust optimization of total joint replacements incorporating environmental variables.

Direct search techniques for the optimal design of biomechanical devices are computationally intensive requiring many iterations before converging to a global solution. This, along with the incorporation of environmental variables such as multiple loading conditions and bone properties, makes direct search techniques infeasible. In this study, we introduced new methods that are based on the statistical design and analysis of computer experiments to account efficiently for environmental variables. Using data collected at a relatively small set of training sites, the method employs a computationally inexpensive predictor of the structural response that is statistically motivated. By using this predictor in place of the simulator (e.g., finite element model), a sufficient number of iterations can be performed to facilitate the optimization of the complex system. The applicability of these methods was demonstrated through the design of a femoral component for total hip arthroplasty incorporating variations in joint force orientation and cancellous bone properties. Beams on elastic foundation (BOEF) finite element models were developed to simulate the structural response. These simple models were chosen for their short computation time. This allowed us to represent the actual structural response surface by an exhaustive enumeration of the design and environmental variable space, and provided a means by which to validate the statistical predictor. We were able to predict the structural response and the optimal design accurately using only 16 runs of the computer code. The general trends predicted by the BOEF models were in agreement with previous three-dimensional finite element computer simulations, and experimental and clinical results, which demonstrated that the important features of intramedullary fixation systems were captured. These results indicate that the statistically based optimization methods are appropriate for optimization studies using computationally demanding models.

Algorithms↗

PubMed2000

Optimization of x-ray imaging geometry (with specific application to flat-panel cone-beam computed tomography).

A theoretical method is presented that allows identification of optimal x-ray imaging geometry, considering the effects of x-ray source distribution, imaging task, x-ray scatter, and imager detective quantum efficiency (DQE). Each of these factors is incorporated into the ICRU-recommended figure of merit for image quality, the detectability index, which is maximized to determine the optimal system configuration. Cascaded systems analysis of flat-panel imagers (FPIs) is extended to incorporate the effects of x-ray scatter directly in the DQE, showing that x-ray scatter degrades DQE as an additive noise source. Optimal magnification is computed for FPI configurations appropriate to (but not limited to) cone-beam computed tomography (CBCT). The sensitivity of the results is examined as a function of focal spot size, imaging task (e.g., ideal observer detection or discrimination tasks), x-ray scatter fraction, detector resolution, and additive noise. Nominal conditions for FPI-CBCT result in optimal magnification of approximately 1.4-1.6, depending primarily on the magnitude of the x-ray scatter fraction. The methodology is sufficiently general that examination of optimal geometry for other FPI applications (e.g., chest radiography, fluoroscopy, and mammography) is possible. The degree to which increased exposure can be used to compensate for x-ray scatter degradation is quantified.

Algorithms↗

PubMed2002

Optimization of a 90Sr/90Y radiation source train stepping for intravascular brachytherapy.

A steepest-descent gradient algorithm is developed to optimize the stepping of a 90Sr/90Y radiation source train (RST) for intravascular brachytherapy (IVB). The objective function is to deliver a uniform dose in a coronary target vessel and minimize the dose in adjacent normal vessel tissue at the proximal and distal edges of the coronary target vessel. Based on the target length and number of dwell points (number of steps), the algorithm modulates the dwell times and corresponding dwell positions that optimize the weighted addition of staggered EGS4 Monte Carlo (MC) calculated dose distribution from a single RST. Stepping treatment plans are generated for target vessel lengths of 3.0, 3.3, and 3.8 cm. For both the unoptimized and optimized plans, the dose heterogeneity in the target vessel wall, and length of nontarget vessel receiving 3 Gy, is assessed to compare plans. Optimization results show a 14% dose uniformity within the target is achievable for all vessel lengths. Further, the dose in the adjacent normal tissue is lower in the optimized plans than the unoptimized plans. The work presented in this paper provides a model to address the finite length of RST in IVB treatments. While the results presented are specific to the 90Sr/90Y RST, the methods should apply to other finite length RSTs.

Algorithms↗

PubMed2003

The use of linear programming in optimization of HDR implant dose distributions.

The introduction of high dose rate brachytherapy enabled optimization of dose distributions to be used on a routine basis. The objective of optimization is to homogenize the dose distribution within the implant while simultaneously satisfying dose constraints on certain points. This is accomplished by varying the time the source dwells at different locations. As the dose at any point is a linear function of the dwell times, a linear programming approach seems to be a natural choice. The dose constraints are inherently linear inequalities. Homogeneity requirements are linearized by minimizing the maximum deviation of the doses at points inside the implant from a prescribed dose. The revised simplex method was applied for the solution of this linear programming problem. In the homogenization process the possible source locations were chosen as optimization points. To avoid the problem of the singular value of the dose at a source location from the source itself we define the "self-contribution" as the dose at a small distance from the source. The effect of varying this distance is discussed. Test cases were optimized for planar, biplanar and cylindrical implants. A semi-irregular, fan-like implant with diverging needles was also investigated. Mean central dose calculation based on 3D Delaunay-triangulation of the source locations was used to evaluate the dose distributions. The optimization method resulted in homogeneous distributions (for brachytherapy). Additional dose constraints--when applied--were satisfied. The method is flexible enough to include other linear constraints such as the inclusion of the centroids of the Delaunay-triangulation for homogenization, or limiting the maximum allowable dwell time.

Brachytherapy↗

PubMed2003

Comparison of IMRT optimization based on a pencil beam and a superposition algorithm.

To investigate the role of sophisticated dose calculation methods for treatment planning, we compared conventional pencil beam optimized 6 and 15 MV intensity-modulated treatment plans with optimizations based on the superposition technique. Five lung and five head and neck IMRT cases with spatial resolutions of bixels and dose voxels usually employed in clinical practice were considered for tumor volumes between 15 and 500 cm3. We investigated the systematic error of the pencil beam algorithm and the pencil beam induced error to the optimal solution of bixel weights. For the lung cases, the pencil beam overestimated the mean dose deposited inside the planning target volume (PTV) by about 8%, for small lung tumors even up to 20.6%. In the head and neck cases only a slight overestimation in mean PTV dose of 1.5% was observed. The optimization with the superposition method substantially improved the dose coverage of the considered radiation targets. Additionally, for the head and neck cases, the brainstem was significantly spared by about 4% mean PTV dose through the use of the superposition technique. Our studies showed that, in target regions with intricate tissue inhomogeneities, superposition or Monte Carlo techniques have to be used for the optimization and the final dose calculation of intensity-modulated treatment plans.

Algorithms↗

PubMed2004

Simultaneous optimization of beam orientations, wedge filters and field weights for inverse planning with anatomy-based MLC fields.

As an alternative between manual planning and beamlet-based IMRT, we have developed an optimization system for inverse planning with anatomy-based MLC fields. In this system, named Ballista, the orientation (table and gantry), the wedge filter and the field weights are simultaneously optimized for every beam. An interesting feature is that the system is coupled to Pinnacle3 by means of the PinnComm interface, and uses its convolution dose calculation engine. A fully automatic MLC segmentation algorithm is also included. The plan evaluation is based on a quasi-random sampling and on a quadratic objective function with penalty-like constraints. For efficiency, optimal wedge angles and wedge orientations are determined using the concept of the super-omni wedge. A bound-constrained quasi-Newton algorithm performs field weight optimization, while a fast simulated annealing algorithm selects the optimal beam orientations. Moreover, in order to generate directly deliverable plans, the following practical considerations have been incorporated in the system: collision between the gantry and the table as well as avoidance of the radio-opaque elements of a table top. We illustrate the performance of the new system on two patients. In a rhabdomyosarcoma case, the system generated plans improving both the target coverage and the sparing of the parotide, as compared to a manually designed plan. In the second case presented, the system successfully produced an adequate plan for the treatment of the prostate while avoiding both hip prostheses. For the many cases where full IMRT may not be necessary, the system efficiently generates satisfactory plans meeting the clinical objectives, while keeping the treatment verification much simpler.

Algorithms↗

PubMed2005

Patient specific optimization of the relation between CT-hounsfield units and proton stopping power with proton radiography.

The purpose of this work is to show the feasibility of using in vivo proton radiography of a radiotherapy patient for the patient individual optimization of the calibration from CT-Hounsfield units to relative proton stopping power. Water equivalent tissue (WET) calibrated proton radiographs of a dog patient treated for a nasal tumor were used as baseline in comparison with integrated proton stopping power through the calibrated CT of the dog. In an optimization procedure starting with a stoichiometric calibration curve, the calibration was modified randomly. The result of this iteration is an optimized calibration curve which was used to recalculate the dose distribution of the patient. One result of this experiment was that the mean value of the deviations between WET calculations based on the stoichiometric calibration curve and the measurements was shifted systematically away from zero. The calibration produced by the optimization procedure reduced this shift to around 0.4 mm. Another result was that the precision of the calibration, reflected as the standard deviation of the normally distributed deviations between WET calculation and measurement, could be reduced from 7.9 to 6.7 mm with the optimized calibration. The dose distributions based on the two calibration curves showed major deviations at the distal end of the target volume.

Animals↗