Search

In this paper, we present an improved imperialist competitive algorithm for solving an inverse form of the Huxley equation, which is a nonlinear partial differential equation. To show the effectiveness of our proposed algorithm, we conduct a comparative analysis with the original imperialist competitive algorithm and a genetic algorithm. The improvement suggested in this study makes the original imperialist competitive algorithm a more powerful method for function approximation. The numerical results show that the improved imperialist competitive algorithm is an efficient algorithm for determining the unknown boundary conditions of the Huxley equation and solving the inverse form of nonlinear partial differential equations.

Abstract Background Prostate-specific membrane antigen (PSMA) PET/CT imaging is widely used for quantitative image analysis, especially in radioligand therapy (RLT) for metastatic castration-resistant prostate cancer (mCRPC). Unknown features influencing PSMA biodistribution can be explored by analyzing segmented organs at risk (OAR) and lesions. Manual segmentation is time-consuming and labor-intensive, so automated segmentation methods are desirable. Training deep-learning segmentation models is challenging due to the scarcity of high-quality annotated images. Addressing this, we developed shifted windows UNEt TRansformers (Swin UNETR) for fully automated segmentation. Within a self-supervised framework, the model’s encoder was pre-trained on unlabeled data. The entire model was fine-tuned, including its decoder, using labeled data. Methods In this work, 752 whole-body [68Ga]Ga-PSMA-11 PET/CT images were collected from two centers. For self-supervised model pre-training, 652 unlabeled images were employed. The remaining 100 images were manually labeled for supervised training. In the supervised training phase, 5-fold cross-validation was used with 64 images for model training and 16 for validation, from one center. For testing, 20 hold-out images, evenly distributed between two centers, were used. Image segmentation and quantification metrics were evaluated on the test set compared to the ground-truth segmentation conducted by a nuclear medicine physician. Results The model generates high-quality OARs and lesion segmentation in lesion-positive cases, including mCRPC. The results show that self-supervised pre-training significantly improved the average dice similarity coefficient (DSC) for all classes by about 3%. Compared to nnU-Net, a well-established model in medical image segmentation, our approach outperformed with a 5% higher DSC. This improvement was attributed to our model’s combined use of self-supervised pre-training and supervised fine-tuning, specifically when applied to PET/CT input. Our best model had the lowest DSC for lesions at 0.68 and the highest for liver at 0.95. Conclusions We developed a state-of-the-art neural network using self-supervised pre-training on whole-body [68Ga]Ga-PSMA-11 PET/CT images, followed by fine-tuning on a limited set of annotated images. The model generates high-quality OARs and lesion segmentation for PSMA image analysis. The generalizable model holds potential for various clinical applications, including enhanced RLT and patient-specific internal dosimetry.

We propose a new learning algorithm to train spiking neural networks (SNN) using conventional artificial neural networks (ANN) as proxy. We couple two SNN and ANN networks, respectively, made of integrate-and-fire (IF) and ReLU neurons with the same network architectures and shared synaptic weights. The forward passes of the two networks are totally independent. By assuming IF neuron with rate-coding as an approximation of ReLU, we backpropagate the error of the SNN in the proxy ANN to update the shared weights, simply by replacing the ANN final output with that of the SNN. We applied the proposed proxy learning to deep convolutional SNNs and evaluated it on two benchmarked datasets of Fashion-MNIST and Cifar10 with 94.56% and 93.11% classification accuracy, respectively. The proposed networks could outperform other deep SNNs trained with tandem learning, surrogate gradient learning, or converted from deep ANNs. Converted SNNs require long simulation times to reach reasonable accuracies while our proxy learning leads to efficient SNNs with much smaller simulation times. The source codes of the proposed method are publicly available at https://github.com/SRKH/ProxyLearning.

In this paper, we present an improved imperialist competitive algorithm for solving an inverse form of the Huxley equation, which is a nonlinear partial differential equation. To show the effectiveness of our proposed algorithm, we conduct a comparative analysis with the original imperialist competitive algorithm and a genetic algorithm. The improvement suggested in this study makes the original imperialist competitive algorithm a more powerful method for function approximation. The numerical results show that the improved imperialist competitive algorithm is an efficient algorithm for determining the unknown boundary conditions of the Huxley equation and solving the inverse form of nonlinear partial differential equations. [ABSTRACT FROM AUTHOR]

This paper aims to study the wrinkle formation of a prepreg with initial defect during steering in automated fiber placement (AFP). Wrinkle formation has a detrimental effect on the mechanical properties of the final product, limiting the AFP applications. A theoretical model for wrinkle formation has been developed in which a Pasternak foundation and a Koiter imperfection model are adapted to model viscoelastic characteristics of the prepreg tack and initial defect of the prepreg, respectively. The initial defect is defined as a slight deviation of the tow’s mid-plane from a horizontal shape. The initial defect is generated in the tow by moving the tow through the guidance system, pressure of the roller, and resin tackiness. Galerkin method, along with the finite difference method (FDM), are employed to solve the wrinkle problem equation. The proposed method is able to satisfy the different boundary conditions for the wrinkle problem completely. The numerical results show that increasing the initial defect leads to a decrease in critical load and an increase in critical steering radius. To validate the theoretical model, experimental results are presented and compared with model-predicted results. It is shown that the model is well able to capture the trends and values of wrinkle formation wavelengths obtained from the experiment.

Backpropagation is the foremost prevalent and common algorithm for training conventional neural networks with deep construction. Here we propose DS4NN, temporal backpropagation for deep spiking neural networks with one spike per neuron. We consider a convolutional spiking neural network consisting of simple non-leaky integrate-and-fire (IF) neurons, and a form of coding named time-tofirst-spike temporal coding in which, neurons are allowed to fire at most once in a specific time interval, which corresponds to simulation duration here. These features together improve the cost and the speed of network computation. We use a surrogate gradient at firing times to solve the non-differentiability of spike times concerning the membrane potential of spiking neurons, and to prevent the emergence of dead neurons in deep layers, we propose a relative encoding scheme for determining desired firing times. Evaluations on two classification tasks of MNIST and Fashion-MNIST datasets confirm the capability of DS4NN on the deep structure of SNNs. It achieves the accuracy of 99.3% (99.8%) and 91.6% (95.3%) on testing samples (training samples) of respectively MNIST and Fashion-MNIST datasets with the mean required number of 1126 and 1863 spikes in the whole network. This shows that the proposed approach can make fast decisions with low-cost computation and high accuracy. [ABSTRACT FROM AUTHOR]

Repeating spatiotemporal spike patterns exist and carry information. Here we investigated how a single spiking neuron can optimally respond to one given pattern (localist coding), or to either one of several patterns (distributed coding, i.e., the neuron's response is ambiguous but the identity of the pattern could be inferred from the response of multiple neurons), but not to random inputs. To do so, we extended a theory developed in a previous paper (Masquelier, 2017), which was limited to localist coding. More specifically, we computed analytically the signal-to-noise ratio (SNR) of a multi-pattern-detector neuron, using a threshold-free leaky integrate-and-fire (LIF) neuron model with non-plastic unitary synapses and homogeneous Poisson inputs. Surprisingly, when increasing the number of patterns, the SNR decreases slowly, and remains acceptable for several tens of independent patterns. In addition, we investigated whether spike-timing-dependent plasticity (STDP) could enable a neuron to reach the theoretical optimal SNR. To this aim, we simulated a LIF equipped with STDP, and repeatedly exposed it to multiple input spike patterns, embedded in equally dense Poisson spike trains. The LIF progressively became selective to every repeating pattern with no supervision, and stopped discharging during the Poisson spike trains. Furthermore, tuning certain STDP parameters, the resulting pattern detectors were optimal. Tens of independent patterns could be learned by a single neuron using a low adaptive threshold, in contrast with previous studies, in which higher thresholds led to localist coding only. Taken together these results suggest that coincidence detection and STDP are powerful mechanisms, fully compatible with distributed coding. Yet we acknowledge that our theory is limited to single neurons, and thus also applies to feed-forward networks, but not to recurrent ones.

Groundwater quality management is the most important issue in many arid and semi-arid countries, including Iran.Artificial neural network (ANN) has an extensive range of applications in water resources management. In this study,artificial neural network was developed using MATLAB R2013 software package, and Cl, EC, SO4 and NO3 qualitativeparameters were estimated and compared with the measured values, in order to evaluate the influence of key inputparameters. The number of neurons in the hidden layer was obtained by the trial-and-error method. For this purpose, datafrom 260 water samples of 13 wells in Bahabad plain were collected during 2003- 2013. The results show that theperformance of ANN model was more accurate for Cl (R=0.96), EC(R=0.98), and SO4(R=0.95), using back-propagationalgorithms according to the best chosen input parameters. It was observed that the use of ANN model for NO3 was notvery accurate, perhaps this was because of the different water sources or the impact of other parameters; thus, this result isin contrast with the study of Diamantopoulou et al. (2005). However, this study confirms that the number of neurons inthe hidden layer cannot be found using a specific formula (double the number of inputs plus one) for all parameters butcan be obtained using a trial-and-error method.

Nowadays, the importance of energy conservation and environmental sustainability issues has led to developments of environmentally friendly products like electric vehicles (EVs). The EVs are not already as widespread as internal combustion engine vehicles (ICEVs) and also the complicated inherent crash behaviour put their crashworthiness at the first development steps. However, besides the battery safety, other major issues such as corresponding vehicle crashworthiness as a consequence of engine and sub-related components elimination and also battery packaging should be addressed. Although before reaching a more specific crash limits for EVs safety evaluation test cases, one could use the ICEV's conventional crash scenarios for the structural crashworthiness evaluations. In this article, the side pole impact as the severe load case according to ENCAP was introduced and protection of the battery pack is improved by implementing a novel energy absorber element into the body sill side. The analyses have performed based on validated computational simulations. The finite element model of the structure is prepared by ANSA and the simulations carried out via ABAQUS and the non-linear explicit finite element software of PAM-CARSH. The various assessment criteria such as pole intrusion, internal energy, section loads before and after the conversion are analysed in order to find an optimum configuration of the structure for battery protection. Obtained results show a significant enhancement in vehicle crashworthiness performance. [ABSTRACT FROM AUTHOR]

The human visual system contains a hierarchical sequence of modules that take part in visual perception at different levels of abstraction, i.e., superordinate, basic, and subordinate levels. One important question is to identify the “entry” level at which the visual representation is commenced in the process of object recognition. For a long time, it was believed that the basic level had a temporal advantage over two others. This claim has been challenged recently. Here we used a series of psychophysics experiments, based on a rapid presentation paradigm, as well as two computational models, with bandpass filtered images of five object classes to study the processing order of the categorization levels. In these experiments, we investigated the type of visual information required for categorizing objects in each level by varying the spatial frequency bands of the input image. The results of our psychophysics experiments and computational models are consistent. They indicate that the different spatial frequency information had different effects on object categorization in each level. In the absence of high frequency information, subordinate and basic level categorization are performed less accurately, while the superordinate level is performed well. This means that low frequency information is sufficient for superordinate level, but not for the basic and subordinate levels. These finer levels rely more on high frequency information, which appears to take longer to be processed, leading to longer reaction times. Finally, to avoid the ceiling effect, we evaluated the robustness of the results by adding different amounts of noise to the input images and repeating the experiments. As expected, the categorization accuracy decreased and the reaction time increased significantly, but the trends were the same. This shows that our results are not due to a ceiling effect. The compatibility between our psychophysical and computational results suggests that the temporal advantage of the superordinate (resp. basic) level to basic (resp. subordinate) level is mainly due to the computational constraints (the visual system processes higher spatial frequencies more slowly, and categorization in finer levels depends more on these higher spatial frequencies).

The impact of air pollution and environmental issues on public health is one of the main topics studied in manycities around the world. Ozone is a greenhouse gas that contributes to global climate. This study was conducted topredict and model ozone of Yazd in the lower atmosphere by an adaptive neuro-fuzzy inference system (ANFIS). Allthe data were extracted from 721 samples collected daily over two successive years, from April 2012 to 29 March2014. The concentration of pollutants and meteorological variables including NOX, temperature, wind speed andwind direction were considered as input and ozone (O3) as the output of model. The results showed that among fivemembership functions used in the model, the Gaussian membership function with R2 equal to 0.949, RMSE equal to2.430 and correlation coefficient equal to 0.974 was obtained as the best model to predict the concentration of ozonein the lower atmosphere. This study showed that predicting and modelling ozone using an adaptive neuro-fuzzyinference system (ANFIS) is appropriate and, due to the expansion of the city of Yazd in the not too distant future, itis necessary to pay more attention to the permissible threshold values of pollutants such as ozone.

Groundwater quality management is one of the most important issues in many arid and semi-arid regions, including Iran.Nitrate (NO3-) is one of the most common anions contaminating groundwater. This study aimed to range nitrateconcentrations in water resources in Bahabad plain in Yazd province. To evaluate the nitrate data in this descriptive study,260 nitrate samples from 13 wells in Bahabad were assessed from 2003 to 2013. The two interpolation techniques ofkriging and inverse distance weighting (IDW) were used to obtain the spatial distribution of groundwater qualityparameters by means of Arcview GIS 10 software. The results of this study showed that the kriging method is moreaccurate than IDW for groundwater quality mapping, based on the lower root mean square error (RMSE) of kriging.Nitrate levels in samples from regional wells were lower than standard levels for Iran and the world. However, nitratecontamination tended to increase from 2003 to 2013. Furthermore, the greatest nitrate contamination was found in thesouthern part of Bahabad. In conclusion, kriging seems to be an appropriate method for estimating nitrate levels ingroundwater in Bahabad. We recommend action be taken in order to stop the increasing trend of groundwater nitratecontamination in this area.

The photocatalytic UV/TiO2 process has particular importance due to having high rate and efficiency in the removal of organic and inorganic contaminants from industrial wastewater. One of the problem of utilization of a catalyst with physical properties similar to TiO2 (anatase) is the separation of the catalyst from the wastewater effluent. In this study, synthesis of titanium oxide on the surface of alumina (particle size 150 to 200 µm) with the wet impregnation method was accomplished in order to create a catalyst with suitable physical properties to easy separation capability from industrial effluents. Hence, titanium isoprpylate compound was used and after the reaction of alumina on the surface, in order to Synthesis of titanium oxide anatase form, calcinations being done in the temperature of 500 ◦C in an electric furnace. The amount of anatase phase formation was measured by X-ray diffraction technique. Finally the removal of free cyanide in the presence of TiO2 and Al2O3/TiO2 was investigated in optimal conditions with the Change of parameters such as irradiation time of UV, the amount of catalyst and initial concentration of cyanide. Experiments were carried out by using a batch photoreactor and a high pressure Hg lamp (250 watt). The results indicated that a layer of anatase TiO2 formed on the surface of Al2O3particles which its value depends on the increasing frequency synthesis. The study of the kinetic of cyanide removal process in the presence of the synthetic catalyst Al2O3/ TiO2 showed that the curve of concentration versus time is logarithmic in this process which indicated the reaction is the first order The results also showed that the catalyst TiO2 has a greater Photocatalytic activity in removal of cyanid compared to Al2O3/ TiO2 due to its higher purity and tiny particle size. However, the physical properties of Al2O3 /TiO2 catalyst including easy separation and reuse from industrial effluent in removal process, could justify economical and practical of its application.

View-invariant object recognition is a challenging problem that has attracted much attention among the psychology, neuroscience, and computer vision communities. Humans are notoriously good at it, even if some variations are presumably more difficult to handle than others (e.g. 3D rotations). Humans are thought to solve the problem through hierarchical processing along the ventral stream, which progressively extracts more and more invariant visual features. This feed-forward architecture has inspired a new generation of bio-inspired computer vision systems called deep convolutional neural networks (DCNN), which are currently the best models for object recognition in natural images. Here, for the first time, we systematically compared human feed-forward vision and DCNNs at view-invariant object recognition task using the same set of images and controlling the kinds of transformation (position, scale, rotation in plane, and rotation in depth) as well as their magnitude, which we call variation level. We used four object categories: car, ship, motorcycle, and animal. In total, 89 human subjects participated in 10 experiments in which they had to discriminate between two or four categories after rapid presentation with backward masking. We also tested two recent DCNNs (proposed respectively by Hinton's group and Zisserman's group) on the same tasks. We found that humans and DCNNs largely agreed on the relative difficulties of each kind of variation: rotation in depth is by far the hardest transformation to handle, followed by scale, then rotation in plane, and finally position (much easier). This suggests that DCNNs would be reasonable models of human feed-forward vision. In addition, our results show that the variation levels in rotation in depth and scale strongly modulate both humans' and DCNNs' recognition performances. We thus argue that these variations should be controlled in the image datasets used in vision research.

Nowadays, brain signals are employed in various scientific and practical fields such as Medical Science, Cognitive Science, Neuroscience, and Brain Computer Interfaces. Hence, the need for robust signal analysis methods with adequate accuracy and generalizability is inevitable. The brain signal analysis is faced with complex challenges including small sample size, high dimensionality and noisy signals. Moreover, because of the non-stationarity of brain signals and the impacts of mental states on brain function, the brain signals are associated with an inherent uncertainty. In this paper, an evidence-based combining classifiers method is proposed for brain signal analysis. This method exploits the power of combining classifiers for solving complex problems and the ability of evidence theory to model as well as to reduce the existing uncertainty. The proposed method models the uncertainty in the labels of training samples in each feature space by assigning soft and crisp labels to them. Then, some classifiers are employed to approximate the belief function corresponding to each feature space. By combining the evidence raised from each classifier through the evidence theory, more confident decisions about testing samples can be made. The obtained results by the proposed method compared to some other evidence-based and fixed rule combining methods on artificial and real datasets exhibit the ability of the proposed method in dealing with complex and uncertain classification problems.

This paper aims to study the wrinkle formation of a prepreg with initial defect during steering in automated fiber placement (AFP). Wrinkle formation has a detrimental effect on the mechanical properties of the final product, limiting the AFP applications. A theoretical model for wrinkle formation has been developed in which a Pasternak foundation and a Koiter imperfection model are adapted to model viscoelastic characteristics of the prepreg tack and initial defect of the prepreg, respectively. The initial defect is defined as a slight deviation of the tow's mid-plane from a horizontal shape. The initial defect is generated in the tow by moving the tow through the guidance system, pressure of the roller, and resin tackiness. Galerkin method, along with the finite difference method (FDM), are employed to solve the wrinkle problem equation. The proposed method is able to satisfy the different boundary conditions for the wrinkle problem completely. The numerical results show that increasing the initial defect leads to a decrease in critical load and an increase in critical steering radius. To validate the theoretical model, experimental results are presented and compared with model-predicted results. It is shown that the model is well able to capture the trends and values of wrinkle formation wavelengths obtained from the experiment. [ABSTRACT FROM AUTHOR]

Frontal vehicle structure is of high importance through crash energy managements and crash boxes are the fundamental structural component for vehicle safety as well as after sales issues. Similar to many other vehicle components, the detail design of crash box is usually part of manufacture knowhow. However, some guide lines are always available. In this article a general procedure is introduced for designing of crash box with the aid of novel thin walled structures and according to conventional crash scenarios. The problem is followed through some basic steps. Firstly, the crash box idea is selected through a wide range of previous investigated elements and is packaged in a real bench vehicle. Then thanks to the protection provided by the new crash box on the other more expensive components (e.g. headlamp, cooling pack, etc.), the effectiveness of this element are acknowledged through the low speed offset crash. Further on the robustness of new proposed crash box is approved by high speed crash simulations. The quasi-static simulations implemented during the analyses are carried out by finite element explicit code (Abaqus) and the FE modeling and dynamic simulation through the next steps are also performed in ANSA and PAM CRASH respectively. Finally in addition to the general crash box design proposed procedure, the achieved results demonstrated that the corrugated conical thin walled tubes deforms in regular and rather stable shape under both axial and oblique loadings. They also produced a reasonable reaction force versus deformations which leads to stiff and crashworthy energy absorber in comparison to traditional rectangular and even some special models like as origami shapes, and so they could be a valuable selection for crash box implementations in passenger cars. [ABSTRACT FROM AUTHOR]

We propose a new supervised learning rule for multilayer spiking neural networks (SNNs) that use a form of temporal coding known as rank-order-coding. With this coding scheme, all neurons fire exactly one spike per stimulus, but the firing order carries information. In particular, in the readout layer, the first neuron to fire determines the class of the stimulus. We derive a new learning rule for this sort of network, named S4NN, akin to traditional error backpropagation, yet based on latencies. We show how approximated error gradients can be computed backward in a feedforward network with any number of layers. This approach reaches state-of-the-art performance with supervised multi-fully connected layer SNNs: test accuracy of 97.4% for the MNIST dataset, and 99.2% for the Caltech Face/Motorbike dataset. Yet, the neuron model that we use, nonleaky integrate-and-fire, is much simpler than the one used in all previous works. The source codes of the proposed S4NN are publicly available at https://github.com/SRKH/S4NN. [ABSTRACT FROM AUTHOR]

Repeating spatiotemporal spike patterns exist and carry information. Here we investigated how a single spiking neuron can optimally respond to one given pattern (localist coding), or to either one of several patterns (distributed coding, i.e., the neuron's response is ambiguous but the identity of the pattern could be inferred from the response of multiple neurons), but not to random inputs. To do so, we extended a theory developed in a previous paper (Masquelier, 2017), which was limited to localist coding. More specifically, we computed analytically the signal-to-noise ratio (SNR) of a multi-pattern-detector neuron, using a threshold-free leaky integrate-and-fire (LIF) neuron model with non-plastic unitary synapses and homogeneous Poisson inputs. Surprisingly, when increasing the number of patterns, the SNR decreases slowly, and remains acceptable for several tens of independent patterns. In addition, we investigated whether spike-timing-dependent plasticity (STDP) could enable a neuron to reach the theoretical optimal SNR. To this aim, we simulated a LIF equipped with STDP, and repeatedly exposed it to multiple input spike patterns, embedded in equally dense Poisson spike trains. The LIF progressively became selective to every repeating pattern with no supervision, and stopped discharging during the Poisson spike trains. Furthermore, tuning certain STDP parameters, the resulting pattern detectors were optimal. Tens of independent patterns could be learned by a single neuron using a low adaptive threshold, in contrast with previous studies, in which higher thresholds led to localist coding only. Taken together these results suggest that coincidence detection and STDP are powerful mechanisms, fully compatible with distributed coding. Yet we acknowledge that our theory is limited to single neurons, and thus also applies to feed-forward networks, but not to recurrent ones. [ABSTRACT FROM AUTHOR]

Background: Artificial neural networks (ANNs) and adaptive neural-fuzzy Inference system (ANFIS) are the best solutions to finding the correlation between some water parameters and human hormones. The correlation between thyroid stimulating hormone (TSH) and drinking water fluoride studied by ANNS and ANFIS models in Yazd city. Method: In this study, eighty people with thyroid gland disorder and 213 healthy people invited. Their thyroid hormones and fluoride drinking water analyzed. Results: The result of ANFIS showed R2 = 0.81 for test and R2 = 0.85 for train in all cases and controls data. This results were R2 = 0.73 and R2 = 0.81 for ANNs respectively. Conclusion: This models can be used as an alternative for show correlation between Drinking Water Fluoride and TSH Hormone and R2 = 0.85 gained from ANFIS was the best. [ABSTRACT FROM AUTHOR]

Retinal image of surrounding objects varies tremendously due to the changes in position, size, pose, illumination condition, background context, occlusion, noise, and non-rigid deformations. But despite these huge variations, our visual system is able to invariantly recognize any object in just a fraction of a second. To date, various computational models have been proposed to mimic the hierarchical processing of the ventral visual pathway, with limited success. Here, we show that the association of both biologically inspired network architecture and learning rule significantly improves the models׳ performance when facing challenging invariant object recognition problems. Our model is an asynchronous feedforward spiking neural network. When the network is presented with natural images, the neurons in the entry layers detect edges, and the most activated ones fire first, while neurons in higher layers are equipped with spike timing-dependent plasticity. These neurons progressively become selective to intermediate complexity visual features appropriate for object categorization. The model is evaluated on 3D-Object and ETH -80 datasets which are two benchmarks for invariant object recognition, and is shown to outperform state-of-the-art models, including DeepConvNet and HMAX. This demonstrates its ability to accurately recognize different instances of multiple object classes even under various appearance conditions (different views, scales, tilts, and backgrounds). Several statistical analysis techniques are used to show that our model extracts class specific and highly informative features. [ABSTRACT FROM AUTHOR]

The physical and chemical quality of the public drinking water supply (tap water), bottled water, and point of use (PoU) treated water was studied comparatively. The analyzed parameters were: turbidity, electrical conductivity, total dissolved solids, pH, hardness, sodium, potassium, chloride, and alkalinity. The samples were taken and analyzed based on standard methods references for the examination of the water and wastewater. The data analysis was conducted by SPSS 16 software. The results show that the concentration of the chemical and physical parameters in all waters is below limits as allowed by national and international drinking water guidelines and standards, although there is significant difference between three types of water. The quality of the tap water is consistent to mentioned guidelines and standards; therefore, the tap water is safe for drinking and it is no need to use other water resources instead of this water. The distrust to public water supply has caused a large number of citizens to use bottled and PoU treated waters for drinking aims. The consumption of these waters imposes huge costs to families because the costs of the 0.25, 1.5, and 20 L bottled waters are about 17,777, 1,776, and 500 times the price of the tap water, respectively. Increased public awareness and confidence about the quality of tap water is needed to prevent the loss of the income of the family. [ABSTRACT FROM PUBLISHER]