Longer Run Times Research Articles

Speeding up distributed pseudo-tree optimization procedures with cross edge consistency to solve DCOPs

The Distributed Pseudo-tree Optimization Procedure (DPOP) is a well-known message passing algorithm that provides optimal solutions to Distributed Constraint Optimization Problems (DCOPs) in cooperative multi-agent systems. However, the traditional DCOP formulation does not consider constraints that must be satisfied (hard constraints), rather it concentrates only on constraints that place no restriction on satisfaction (soft constraints). This is a serious shortcoming as many real-world applications involve both types of constraints. Traditional DPOP algorithms are not able to benefit from the existence of hard constraints, where an additional calculation is required to handle such constraints. This results in longer runtimes. Thus scalability remains an issue. Additionally, in the standard DPOP, the agents are arranged as a Depth First Search (DFS) pseudo-tree, but recent work has shown that the construction of pseudo-trees in this way often leads to chain-like communication structures that greatly impair the algorithm’s performance. To address these issues, we develop an algorithm that speeds up the DPOP algorithm by reducing the size of the messages exchanged and increases parallelism in the pseudo tree. For this purpose, initially, we improve the path for exchanging messages. Next, we introduce a new form of constraint propagation, which we call cross-edge consistency. Our theoretical evaluation shows that our proposed algorithm is complete and correct. In empirical evaluations, our algorithm achieves a significant reduction in the runtime, ranging from 4% to 96%, compared to the state-of-the-art.

Identification of Malignancies from Free-Text Histopathology Reports Using a Multi-Model Supervised Machine Learning Approach

We explored various Machine Learning (ML) models to evaluate how each model performs in the task of classifying histopathology reports. We trained, optimized, and performed classification with Stochastic Gradient Descent (SGD), Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbor (KNN), Adaptive Boosting (AB), Decision Trees (DT), Gaussian Naïve Bayes (GNB), Logistic Regression (LR), and Dummy classifier. We started with 60,083 histopathology reports, which reduced to 60,069 after pre-processing. The F1-scores for SVM, SGD KNN, RF, DT, LR, AB, and GNB were 97%, 96%, 96%, 96%, 92%, 96%, 84%, and 88%, respectively, while the misclassification rates were 3.31%, 5.25%, 4.39%, 1.75%, 3.5%, 4.26%, 23.9%, and 19.94%, respectively. The approximate run times were 2 h, 20 min, 40 min, 8 h, 40 min, 10 min, 50 min, and 4 min, respectively. RF had the longest run time but the lowest misclassification rate on the labeled data. Our study demonstrated the possibility of applying ML techniques in the processing of free-text pathology reports for cancer registries for cancer incidence reporting in a Sub-Saharan Africa setting. This is an important consideration for the resource-constrained environments to leverage ML techniques to reduce workloads and improve the timeliness of reporting of cancer statistics.

Evaluation of Korean-Language COVID-19-Related Medical Information on YouTube: Cross-Sectional Infodemiology Study.

BackgroundIn South Korea, the number of coronavirus disease (COVID-19) cases has declined rapidly and much sooner than in other countries. South Korea is one of the most digitalized countries in the world, and YouTube may have served as a rapid delivery mechanism for increasing public awareness of COVID-19. Thus, the platform may have helped the South Korean public fight the spread of the disease.ObjectiveThe aim of this study is to compare the reliability, overall quality, title–content consistency, and content coverage of Korean-language YouTube videos on COVID-19, which have been uploaded by different sources.MethodsA total of 200 of the most viewed YouTube videos from January 1, 2020, to April 30, 2020, were screened, searching in Korean for the terms “Coronavirus,” “COVID,” “Corona,” “Wuhan virus,” and “Wuhan pneumonia.” Non-Korean videos and videos that were duplicated, irrelevant, or livestreamed were excluded. Source and video metrics were collected. The videos were scored based on the following criteria: modified DISCERN index, Journal of the American Medical Association Score (JAMAS) benchmark criteria, global quality score (GQS), title–content consistency index (TCCI), and medical information and content index (MICI).ResultsOf the 105 total videos, 37.14% (39/105) contained misleading information; independent user–generated videos showed the highest proportion of misleading information at 68.09% (32/47), while all of the government-generated videos were useful. Government agency–generated videos achieved the highest median score of DISCERN (5.0, IQR 5.0-5.0), JAMAS (4.0, IQR 4.0-4.0), GQS (4.0, IQR 3.0-4.5), and TCCI (5.0, IQR 5.0-5.0), while independent user–generated videos achieved the lowest median score of DISCERN (2.0, IQR 1.0-3.0), JAMAS (2.0, IQR 1.5-2.0), GQS (2.0, IQR 1.5-2.0), and TCCI (3.0, IQR 3.0-4.0). However, the total MICI was not significantly different among sources. “Transmission and precautionary measures” were the most commonly covered content by government agencies, news agencies, and independent users. In contrast, the most mentioned content by news agencies was “prevalence,” followed by “transmission and precautionary measures.”ConclusionsMisleading videos had more likes, fewer comments, and longer running times than useful videos. Korean-language YouTube videos on COVID-19 uploaded by different sources varied significantly in terms of reliability, overall quality, and title–content consistency, but the content coverage was not significantly different. Government-generated videos had higher reliability, overall quality, and title–content consistency than independent user–generated videos.

Design optimization of variable stiffness composites by using multi-fidelity surrogate models

Variable stiffness (VS) composites made by fiber steering have received intensive attention due to the tailorability of the stiffness and strength properties. However, a significantly larger number of elements are required to converge for VS composites, which results in much longer runtime and makes the design and optimization more complicated. In this paper, an efficient design optimization method assisted by multi-fidelity surrogate models is presented for the buckling design of VS composites. To reduce the computational burden, a multi-fidelity surrogate model called hierarchical Kriging is constructed through a few expensive high-fidelity samples and many cheap low-fidelity samples. Fine and coarse finite element (FE) analysis is performed respectively to calculate the structural responses for corresponding datasets. The efficient global optimization based on a modified expected improvement criterion is employed and used to adaptively add new samples of variable-fidelity. Two case studies, a composite plate subjected to uniform uniaxial compression and a composite cylinder under pure bending, are investigated. The effects of different number of design variables and coarse FE model mesh density on the optimum configuration are studied to demonstrate the effectiveness and robustness of the method. The results indicate that the present method can remarkably reduce the number of high-fidelity FE evaluations and improve the optimization efficiency when compared with available methods in the literature. Additionally, the investigation in the mechanism of loading carrying capacity improvement shows that the increase is mainly due to the load redistribution.

QoS-Aware Cloudlet Load Balancing in Wireless Metropolitan Area Networks

With advances in wireless communication technology, more and more people depend heavily on portable mobile devices for business, entertainments and social interactions. This poses a great challenge of building a seamless application experience across different computing platforms. A key issue is the resource limitations of mobile devices due to their portable size, however this can be overcome by offloading computation-intensive tasks from the mobile devices to clusters of nearby computers called cloudlets through wireless access points. As increasing numbers of people access the Internet via mobile devices, it is reasonable to envision in the near future that cloudlet services will be available for the public through easily accessible public wireless metropolitan area networks (WMANs). However, the outdated notion of treating cloudlets as isolated data-centers-in-boxes must be discarded as there are clear benefits to connecting multiple cloudlets together to form a network. In this paper we investigate how to balance the workload among cloudlets in an WMAN to optimize mobile application performance. We first introduce a novel system model to capture the response time delays of offloaded tasks and formulate an optimization problem with the aim to minimize the maximum response time of all offloaded tasks. We then propose two algorithms for the problem: one is a fast heuristic, and another is a distributed genetic algorithm that is capable of delivering a more accurate solution compared with the first algorithm, but at the expense of a much longer running time. We finally evaluate the performance of the proposed algorithms in realistic simulation environments. The experimental results demonstrate the significant potential of the proposed algorithms in reducing the user task response time, maximizing user experience.

Physics potential of the ESSnu SB

The ESSnu SB project proposes to base a neutrino “Super Beam” of unprecedented luminosity at the European Spallation Source. The original proposal identified the second peak of the oscillation probability as the optimal to maximize the discovery potential to leptonic CP violation. However this choice reduces the statistics at the detector and penalizes other complementary searches such as the determination of the atmospheric oscillation parameters, particularly the octant of theta _{23} as well as the neutrino mass ordering. We explore how these shortcomings can be alleviated by the combination of the beam data with the atmospheric neutrino sample that would also be collected at the detector. We find that the combination not only improves very significantly these drawbacks, but also enhances both the CP violation discovery potential and the precision in the measurement of the CP violating phase, for which the facility was originally optimized, by lifting parametric degeneracies. We then reassess the optimization of the ESSnu SB setup when the atmospheric neutrino sample is considered, with an emphasis in performing a measurement of the CP violating phase as precise as possible. We find that for the presently preferred value of delta sim -pi /2, shorter baselines and longer running time in neutrino mode would be optimal. In these conditions, a measurement better than 14^circ would be achievable for any value of the theta _{23} octant and the mass ordering. Conversely, if present and next generation facilities were not able to discover CP violation, longer baselines and more even splitting between neutrino and neutrino modes would be preferable. These choices would allow a 5 sigma discovery of CP violation for around a 60% of the possible values of delta and to determine its value with a precision around 6^circ if it is close to 0 or pi .

Kinematic Analysis of Gait in a Patient with Nonspecific Low Back Pain

Low back pain is one of the most commonly encountered musculoskeletal diseases in modern society, and it is estimated that about 70 to 85% of the population has experienced low back pain. Individuals with low back pain usually modify the pattern of movement during gait to protect themselves, that is, they use different strategies of trunk movement and lower limbs protectively to avoid painful movements. However, in the long term, these repeated and prolonged changes and tensions during gait can cause mechanical dysfunctions, overloading the lumbar vertebral column and consequently causing damage. The objective of this study was to analyze gait kinematics in patients classified in the low back pain subgroups and to identify the possible biomechanical changes of the movement in these individuals. It is a cross-sectional observational study composed of a convenience sample of 17 patients, 24.6 ± 3.0 years old, from the Federal University of Santa Catarina located in the city of Araranguá-SC. Individuals classified in the manipulation subgroup obtained better results in the length of the gait, hip angle in the initial phase of gait and gait speed. The directional preference group obtained a smaller step length and passed. The traction group had the longest running time compared to the other subclassification groups of low back pain. Through the present study, it is possible to affirm that patients with nonspecific chronic low back pain present changes in gait. The individuals classified in the directional preference subgroup were the ones that had the lowest performance in the analyzed variables when compared to the subgroups of traction, manipulation and stabilization.

Memory-efficient self-cross-product for large matrices using R and Python

In many quantitative studies, calculations of matrix self-cross-products (B’B) are needed, where B is any matrix of interest. For matrix B with many number of rows, there might be memory limitations in storing B’B. Also, calculating B’B has a computational complexity of m 2n, for B with n and m number of rows and columns, respectively. Because B’B is symmetric, almost half of the calculations and the memory usage are redundant. The half-matrix multiplication algorithm (HMMA) was introduced, which creates B’B upper-triangular. Matrix multiplication functions %*% and crossprod in R, numpy.dot in Python, and user-defined HMMA functions hmma r and hmma py in R and Python were compared, for matrix B with 40,000 real numbers, and various dimensions. Runtime of B’B calculation was less than a second when B had more than 4 rows. The longest runtime was for B with 1 row and crossprod (21.3 sec), and then numpy.dot (9.7 sec). Considering B with 4 or less number of rows, hmma_py, %*%, and hmma r ranked 1 to 3 for the shortest runtime. The memory usage of a (40,000 × 40,000) B’B was 12.8 Gb, and the main advantage of HMMA was reducing it to the half.

A novel parallel Markov clustering method in biological interaction network analysis under multi-GPU computing environment

The various clustering methods are widely applied in analyzing biological interaction networks, such as the protein–protein interaction and the genetic interaction networks. With the rapid growth of these biological datasets in scale, much longer runtime is required to make cluster analyses on them. In this paper, we propose a novel parallel Markov clustering (MCL) method based on the ELLPACK-R sparse matrix format that can run on multiple graphic processing units (GPUs) equipped standalone computers. The method is implemented using the Compute Unified Device Architecture (CUDA) programming framework, and fine-grained warp-level optimization is introduced for improving the performance. The BioGRID, a large-scale and freely accessible database of protein and genetic interactions, is adopted as the dataset in the experiment. The method has been assessed on a desktop computer equipped with two NVIDIA GTX 1070 GPUs. The results show that the proposed multi-GPU method can conduct MCL clustering on the full-size BioGRID database with about 6.5 min, that is much faster than the CPU serial MCL implementation which needs almost an hour and a half execution time.

The simultaneous quantification of phytosterols and tocopherols in liposomal formulations using validated atmospheric pressure chemical ionization- liquid chromatography –tandem mass spectrometry

A novel liquid chromatography tandem mass spectrometry (LC–MS/MS) method was developed and validated to simultaneously quantify phytosterols (brassicasterol, campesterol, stigmasterol and β-sitosterol) and tocopherols (alpha, beta, gamma and delta) entrapped in the lipid bilayer of a liposomal formulation. Apart from liposomes (a pharmaceutical product), the developed method was able to quantify target analytes in agricultural products, thus showing wide applications. Atmospheric pressure chemical ionization (APCI) was employed due to the enhanced ionization of phytosterols and tocopherols in comparison to electrospray ionization. Unlike published work, the chromatographic conditions were modified to simplify the analytical approach. For the first time, a simple isocratic elution (acetonitrile:methanol 99:1 v/v) was utilized for the separation of four phytosterols and four tocopherols in a single run. A substantially better baseline separation of phytosterols were obtained in comparison to reported methods by using poroshell C18 column. The method has a total run time of 7 min, which is the shortest run time among all reported quantitative methods for the simultaneous determination of four phytosterols and four tocopherols. Calibration curves for all phytosterols were linear in the range of 0.05−10 μg/mL. In the case of tocopherols, alpha tocopherol showed linear response in the range of 0.25−10 μg/mL. However, gamma and delta tocopherols exhibited quadratic relationship in the same concentration range (0.25−10 μg/mL). Validation parameters met the International Conference on Harmonization (ICH) guidelines in terms of selectivity, accuracy, precision, repeatability, sensitivity, matrix effects, dilution integrity and stability. The method was, for the first time, successfully applied for the quantifying phytosterols and tocopherols entrapped inside liposomes. An interesting chromatographic phenomenon was observed during sample analysis. Alpha tocopherol (entrapped in the liposomal lipid bilayer) was found to elute at two retention times, 2.53 min and 3.60 min. Such dual separation was not observed in calibration standards and quality controls. It was concluded that the chiral recognition ability of liposomes made up of phosphatidylcholine separated the enantiomers of alpha tocopherol, giving rise to two peaks at two different retention time. To sum, the reported novel LC–MS/MS method addresses three major analytical shortcomings, namely i)longer run time, ii)complex gradient elution and iii)poor baseline separation of phytosterols and tocopherols.

Design and Implementation of an Infrared-Based Sensor for Finger Movement Detection

With the increasing interest in smart devices and convenient remote control, the need for accurate wireless means of control has become imperative. This gives rise to research in the field of gesture and finger movement detection. This design focuses on exploring techniques involved in hand and finger movement detection, using the depth-sensing infrared cameras embedded on Xbox Kinect Module. The generated 3-D images are first filtered along the z-axis, then two distinct techniques; Haar-Like Features, and Deep Learning using a Convolution Neural Network, are performed on the images to detect hands. Useful metrics like, Precision, Recall, F1-Score and Accuracy are then used to evaluate the efficiency of these techniques. The results show that while the deep learning model is the most accurate with a weighted accuracy of 1.0 (due to the absence of noise in the images) in contrast with 0.97 observed for the Haar-Like features, the Haar-like features technique runs faster due to its static nature. These findings point to the conclusion that the deep learning model is a better technique for detecting hands despite its longer running time.

A sensitivity study of Surface Wind simulations over Coastal Ghana to selected Time Control and Nudging options in the Weather Research and Forecasting Model

Over the years, the Weather Research and Forecasting Model (WRF) has been gaining popularity as a low-cost alternative source of data for wind resource assessments. This paper investigates the impact of selected time control, and nudging options on wind simulations in WRF. We conducted 15 numerical experiments, combining 5 simulation run-times and 3 options for disabling nudging in the Planetary Boundary Layer (PBL) in WRF. Hourly wind speed and direction predictions were compared with actual measurements at 40 m, 50 m and 60 m a.g.l. From our results, we recommend that, for optimum performance, the method of disabling nudging in the PBL should be chosen with simulation run times in mind. For wind simulations in our study area, up to 2 days run-times with nudging disabled below 1600 m in model configurations gives the best wind speed predictions. However, disabling nudging below the model-calculated PBL height offers more consistent results and produces relatively less prediction error with longer run times.

Research on management strategy of coordination behavior of task conflicts in in-service thermal power unit operation based on big data modeling

With the development of the internet and information technology, the in-service thermal power unit is facing more challenges, and the innovation of the operation and management mode of the in-service thermal power unit is urgent and necessary. From the perspective of work conflict, this paper constructs a multi-objective genetic algorithm, which introduces big data modelling technology into the management innovation of in-service thermal power units. The algorithm solves the relationship between various operating entities in active thermal power units through functions. In order to get the optimal solution for vehicle distribution. Firstly, the contingency theory is introduced into the innovative design scheme of the in-service thermal pow?er unit information system to optimize the management decision-making distribution path in the big data environment, design the multi-objective genetic algorithm steps, construct the non-dominated set, and combine the target cross-variation operations. The genetic sub-categories are jointly derived, and then the relationship between the parties in the management and decision-making innovation management activities of the in-service thermal power units is solved. The experimental results show that the shortest running time of the algorithm during the experimental operation is 0.56 seconds, and the longest running time is 2.48 seconds. The average running time in the whole process is less than 1 second, which meets the actual demand. The genetic algorithm can help the in-service thermal power unit. Reasonable arrangements for managing the delivery route of the decision-making fleet. The research in this paper has implications for the management innovation of in-service thermal power units in the information environment, and further expands the application field of big data modelling, which has practical significance.

Regularity-Aware Routability-Driven Macro Placement Methodology for Mixed-Size Circuits With Obstacles

This paper introduces a routability-driven macro placement algorithm for mixed-size circuits and pays special attention to the effect of regular placement of macros. Once macros are placed with regularity, powerplanning will become easier and better routability can be obtained. Our methodology consists of two stages. First, placement prototyping stage distributes cells and macros over a placement region while keeping those macros and cells with a strong connection and in similar hierarchies tied together. This is facilitated by clustering macros and cells properly in advance. In the second stage, a deterministic macro legalization algorithm is proposed to arrange macros without deteriorating the result obtained in the previous stage. Although several macro legalization algorithms have been proposed in recent years, most of these works adopt the simulated annealing algorithm which usually takes longer runtime. Moreover, they either cannot handle preplaced macros or require preplaced macros to be abutted to chip boundaries. Unlike these approaches, our method iteratively places a macro by extracting available space for it. Experimental results demonstrate the efficiency and effectiveness of our method according to real industry circuits.

MoMo: discovery of statistically significant post-translational modification motifs.

Post-translational modifications (PTMs) of proteins are associated with many significant biological functions and can be identified in high throughput using tandem mass spectrometry. Many PTMs are associated with short sequence patterns called 'motifs' that help localize the modifying enzyme. Accordingly, many algorithms have been designed to identify these motifs from mass spectrometry data. Accurate statistical confidence estimates for discovered motifs are critically important for proper interpretation and in the design of downstream experimental validation. We describe a method for assigning statistical confidence estimates to PTM motifs, and we demonstrate that this method provides accurate P-values on both simulated and real data. Our methods are implemented in MoMo, a software tool for discovering motifs among sets of PTMs that we make available as a web server and as downloadable source code. MoMo re-implements the two most widely used PTM motif discovery algorithms-motif-x and MoDL-while offering many enhancements. Relative to motif-x, MoMo offers improved statistical confidence estimates and more accurate calculation of motif scores. The MoMo web server offers more proteome databases, more input formats, larger inputs and longer running times than the motif-x web server. Finally, our study demonstrates that the confidence estimates produced by motif-x are inaccurate. This inaccuracy stems in part from the common practice of drawing 'background' peptides from an unshuffled proteome database. Our results thus suggest that many of the papers that use motif-x to find motifs may be reporting results that lack statistical support. The MoMo web server and source code are provided at http://meme-suite.org. Supplementary data are available at Bioinformatics online.

Plug Power simplifies hydrogen services for backup power market

In the US, Plug Power is now providing the capability to provide tiered hydrogen services in-house for its customers, allowing it to quickly supply even the most remote sites, with cost-effective solutions that allow longer runtimes with a larger capacity fuel storage.

An optimized Monte Carlo ray tracing optical simulation model and its applications to line-focus concentrating solar collectors

The Monte Carlo Ray Tracing (MCRT) method has been confirmed flexible and efficient in the optical simulation of Concentrating Solar Collectors (CSCs), but it usually needs higher computing cost and longer runtime or its results fluctuate in multiple runs. The parameters of the way of random number generation, the number of rays, running times, grid numbers, and random number generation times all exerted effects on the simulation results. It was found that running the MCRT model with less number of rays for several more times could mitigate the fluctuation of results and decrease the total runtime simultaneously. Taken the Line-focus CSC with a metal-glass receiver and a parabolic reflector as an example, the maximum (emax) and average (eavg) relative errors of the MCRT method with 1 × 108 rays running for once, 2 × 107 rays running for once and 3 × 106 rays running for five times were all lower than the threshold values (Emax = 5% and Eavg = 0.5%), but the total runtime was about 410 s, 82 s and 63 s respectively. On these bases, an optimized MCRT model was proposed by combining the MCRT method with the iteration method, where the minimum running times (tmin) and the maximum running times (tmax) were introduced, and they could be changed conveniently to meet the requirements of different optical simulations. By applying the proposed model to the Line-focus CSC with a more complex cavity receiver or compound parabolic reflector, the total runtime varied in the range of 268–413 s and 26–102 min respectively, indicating that the runtime reduction was significant when the limit of relative errors were acceptable. The proposed model is beneficial to mitigate the fluctuation, improve the accuracy and reduce the runtime of the MCRT method. It can also be further used to the optical simulation of various kinds of CSCs.

Encoding Scheme to Reduce Energy Consumption of Delivering Data in Radio Frequency Powered Battery-Free Wireless Sensor Networks

Wireless sensor networks (WSNs) have been widely deployed in intelligent transportation systems. Battery-free WSN (BF-WSN) scavenges energy from the environment. In the BF-WSN, whose nodes harvest radio frequency (RF) energy, delivering data with low energy consumption is important for the BF-WSN nodes to have a longer runtime. Some existing baseband encoding or carrier-modulation schemes exhibit the energy consumption disparity (ECD) between transmitting bit 0 and bit 1. An example is FM0 (bi-phase space) baseband encoding, which is widely applied in the backscattering communications in radio frequency identification systems and wireless identification sensing platform. With FM0, transmitting bit 0 consumes much greater energy than bit 1. To achieve energy-saving in delivering data with the ECD, we, in this paper, formulate the optimization problem (OP) that maximizes percentage of energy saving. The solution of the OP leads to the optimal energy-efficient codebook. Based on the derived codebook, we present the energy-efficient data delivery scheme (EEDDS), with which the sender transmits the codewords in the codebook and the receiver recovers the corresponding original data blocks using the codebook. Simulation results show that the proposed EEDDS outperforms the traditional data delivery scheme in terms of energy consumption in the case when the applied baseband encoding or carrier-modulation schemes occupy the ECD.

A fast algorithm for calculating isothermal phase behavior using machine learning

Compositional models are frequently used to describe fluids in petroleum reservoir simulation, particularly for simulations of enhanced oil recovery. While compositional models are more accurate than black oil models, they incur a larger computational cost, in part, due to more complex phase-equilibrium calculations and can result in longer run times. Here, we develop an algorithm to reduce the cost of phase-equilibrium calculations for compositional models by applying two machine learning techniques: relevance vector machines and artificial neural networks. We test the algorithm on three fluid data sets and find a speedup of over 20% with an error of 0.01%, and a speedup of over 90% with a maximum error of 5%. These results suggest that the algorithm can be used to reduce the overall run time of compositional reservoir simulations with a small impact on accuracy.

Evaluation of performance of distributed delay model for chemotherapy-induced myelosuppression

The distributed delay model has been introduced that replaces the transit compartments in the classic model of chemotherapy-induced myelosuppression with a convolution integral. The maturation of granulocyte precursors in the bone marrow is described by the gamma probability density function with the shape parameter (ν). If ν is a positive integer, the distributed delay model coincides with the classic model with ν transit compartments. The purpose of this work was to evaluate performance of the distributed delay model with particular focus on model deterministic identifiability in the presence of the shape parameter. The classic model served as a reference for comparison. Previously published white blood cell (WBC) count data in rats receiving bolus doses of 5-fluorouracil were fitted by both models. The negative two log-likelihood objective function (-2LL) and running times were used as major markers of performance. Local sensitivity analysis was done to evaluate the impact of ν on the pharmacodynamics response WBC. The ν estimate was 1.46 with 16.1% CV% compared to ν=3 for the classic model. The difference of 6.78 in -2LL between classic model and the distributed delay model implied that the latter performed significantly better than former according to the log-likelihood ratio test (P=0.009), although the overall performance was modestly better. The running times were 1s and 66.2min, respectively. The long running time of the distributed delay model was attributed to computationally intensive evaluation of the convolution integral. The sensitivity analysis revealed that ν strongly influences the WBC response by controlling cell proliferation and elimination of WBCs from the circulation. In conclusion, the distributed delay model was deterministically identifiable from typical cytotoxic data. Its performance was modestly better than the classic model with significantly longer running time.