Search

COVID-19 is a novel virus that attacks the upper respiratory tract and the lungs. Its person-to-person transmissibility is considerably rapid and this has caused serious problems in approximately every facet of individuals' lives. While some infected individuals may remain completely asymptomatic, others have been frequently witnessed to have mild to severe symptoms. In addition to this, thousands of death cases around the globe indicated that detecting COVID-19 is an urgent demand in the communities. Practically, this is prominently done with the help of screening medical images such as Computed Tomography (CT) and X-ray images. However, the cumbersome clinical procedures and a large number of daily cases have imposed great challenges on medical practitioners. Deep Learning-based approaches have demonstrated a profound potential in a wide range of medical tasks. As a result, we introduce a transformer-based method for automatically detecting COVID-19 from X-ray images using Compact Convolutional Transformers (CCT). Our extensive experiments prove the efficacy of the proposed method with an accuracy of 99.22% which outperforms the previous works.

Imbalanced Data (ID) is a problem that deters Machine Learning (ML) models from achieving satisfactory results. ID is the occurrence of a situation where the quantity of the samples belonging to one class outnumbers that of the other by a wide margin, making such models’ learning process biased towards the majority class. In recent years, to address this issue, several solutions have been put forward, which opt for either synthetically generating new data for the minority class or reducing the number of majority classes to balance the data. Hence, in this paper, we investigate the effectiveness of methods based on Deep Neural Networks (DNNs) and Convolutional Neural Networks (CNNs) mixed with a variety of well-known imbalanced data solutions meaning oversampling and undersampling. Then, we propose a CNN-based model in combination with SMOTE to effectively handle imbalanced data. To evaluate our methods, we have used KEEL, breast cancer, and Z-Alizadeh Sani datasets. In order to achieve reliable results, we conducted our experiments 100 times with randomly shuffled data distributions. The classification results demonstrate that the mixed Synthetic Minority Oversampling Technique (SMOTE)-Normalization-CNN outperforms different methodologies achieving 99.08% accuracy on the 24 imbalanced datasets. Therefore, the proposed mixed model can be applied to imbalanced binary classification problems on other real datasets. Validerad;2023;Nivå 2;2023-03-22 (joosat);Licens fulltext: CC BY License

Predicting treatment outcomes in traumatic brain injury (TBI) patients is challenging worldwide. The present study aimed to achieve the most accurate machine learning algorithms to predict the outcomes of TBI treatment by evaluating demographic features, laboratory data, imaging indices, and clinical features. We used data from 3347 patients admitted to a tertiary trauma centre in Iran from 2016 to 2021. After the exclusion of incomplete data, 1653 patients remained. We used machine learning algorithms such as Random Forest (RF) and Decision Tree (DT) with ten-fold cross-validation to develop the best prediction model. Our findings reveal that among different variables included in this study, the motor component of the Glasgow Coma Scale, condition of pupils, and condition of cisterns were the most reliable features for predicting in-hospital mortality, while the patients’ age takes the place of cisterns condition when considering the long-term survival of TBI patients. Also, we found that the RF algorithm is the best model to predict the short-term mortality of TBI patients. However, the generalized linear model (GLM) algorithm had the best performance (with an accuracy rate of 82.03 ± 2.34) in predicting the long-term survival of patients. Our results showed that using appropriate markers, and machine learning algorithms can provide a reliable prediction of TBI patients’ survival in the short- and long-term with reliable and easily accessible features of patients.

This study aimed to investigate the association between regional body fat distribution and the prevalence of Diabetes mellitus (DM) in adult populations using machine learning. We applied machine learning methods to data from a cohort study to analyze the relationship between fat in different body areas and diabetes. All measurement was done by "Tanita Segmental Body Composition Analyzer BC-418 MA Tanita Corp, Japan". The correlation between the used parameters and DM was measured using some machine learning algorithms i.e. SVM, SGD, KNN, MLP, Adaboost and EDINet. A total of 4661 participants were included. The top features that reported higher importance in classification models were age, fat mass, and percentages in legs, arms, and trunk area. Fat-free mass in the legs, arm, and trunk area was reversely associated with diabetes. Our proposed method significantly outperformed the others. It has the best performances in Accuracy, Precision, Recall-0, Recall-1, and F1-score, which were 93.57, 93.67, 96.11, 74.55 and 93.62, respectively. Our machine learning models showed that regional body fat could have specific impacts on diabetes based on the location of the fat accumulation. The most predictor values of diabetes were age, fat mass, and percentages in arms, legs, and trunk area. Further studies on different ages, gender, ethnic groups, and races are recommended.

A circular zero-sum flow for a graph G is a function $$f:E(G) \rightarrow {\mathbb {R}}{\setminus }\{0\}$$ such that for every vertex v, $$\sum _{e\in E_v}f(e)=0$$, where $$E_v$$ is the set of all edges incident with v. If for each edge e, $$1\le |f(e)| \le r-1$$, where $$r\ge 2$$ is a real number, then f is called a circular zero-sum r-flow. Also, if r is a positive integer and for each edge e, f(e) is an integer, then f is called a zero-sum r-flow. If G has a circular zero-sum flow, then the minimum $$r\ge 2$$ for which G has a circular zero-sum r-flow is called the circular zero-sum flow number of G and is denoted by $$\Phi _c(G)$$. Also, the minimum integer $$r\ge 2$$ for which G has a zero-sum r-flow is called the flow number for G and is denoted by $$\Phi (G)$$. In this paper, we investigate circular zero-sum r-flows of regular graphs. In particular, we show that if G is k-regular with m edges, then $$\Phi _c(G)=2$$ for even k and even m, $$\Phi _c(G)=1+\frac{k+2}{k-2}$$ for even k and odd m, and $$\Phi _c(G)\le 1+(\frac{k+1}{k-1})^2$$ for odd k. It was proved that for every k-regular graph G with $$k\ge 3$$, $$\Phi (G)\le 5$$. Here, using circular zero-sum flows, we present a new proof of this result when $$k \ne 5$$. Finally, we prove that a graph G has a circular zero-sum flow f such that for every edge e, $$l(e) \le f(e) \le u(e)$$, if and only if for every partition of V(G) into three subsets A, B, C, $$\begin{aligned} l(A,C)+2l(A) \le u(B,C)+2u(B), \end{aligned}$$where l(A, C) is the sum of values of l on the edges between A, C, and l(A) is the sum of values of l on the edges with both ends in A (the definitions of u(B, C) and u(B) are analogous).

A mixed graph is obtained from a graph by orienting some of its edges. The Hermitian adjacency matrix of a mixed graph with the vertex set { v 1 , … , v n } , is the matrix H = [ h i j ] n × n , where h i j = − h j i = i if there is a directed edge from v i to v j , h i j = 1 if there exists an undirected edge between v i and v j , and h i j = 0 otherwise. The Hermitian spectrum of a mixed graph is defined to be the spectrum of its Hermitian adjacency matrix. In this paper we study mixed graphs which are determined by their Hermitian spectrum (DHS). First, we show that each mixed cycle is switching equivalent to either a mixed cycle with no directed edges ( C n ), a mixed cycle with exactly one directed edge ( C n 1 ), or a mixed cycle with exactly two consecutive directed edges with the same direction ( C n 2 ) and we determine the spectrum of these three types of cycles. Next, we characterize all DHS mixed paths and mixed cycles. We show that all mixed paths of even order, except P 8 and P 14 , are DHS. It is also shown that mixed paths of odd order, except P 3 , are not DHS. Also, all cospectral mates of P 8 , P 14 and P 4 k + 1 and two families of cospectral mates of P 4 k + 3 , where k ≥ 1 , are introduced. Finally, we show that the mixed cycles C 2 k and C 2 k 2 , where k ≥ 3 , are not DHS, but the mixed cycles C 4 , C 4 2 , C 2 k + 1 , C 2 k + 1 2 , C 2 k + 1 1 and C 2 j 1 except C 7 1 , C 9 1 , C 12 1 and C 15 1 , are DHS, where k ≥ 1 and j ≥ 2 .

BackgroundCoronary artery disease (CAD) is one of the crucial reasons for cardiovascular mortality in middle-aged people worldwide. The most typical tool is angiography for diagnosing CAD. The challenges of CAD diagnosis using angiography are costly and have side effects. One of the alternative solutions is the use of machine learning-based patterns for CAD diagnosis.MethodsHence, this paper provides a new hybrid machine learning model called genetic support vector machine and analysis of variance (GSVMA). The analysis of variance (ANOVA) is known as the kernel function for the SVM algorithm. The proposed model is performed based on the Z-Alizadeh Sani dataset so that a genetic optimization algorithm is used to select crucial features. In addition, SVM with ANOVA, linear SVM (LSVM), and library for support vector machine (LIBSVM) with radial basis function (RBF) methods were applied to classify the dataset.ResultsAs a result, the GSVMA hybrid method performs better than other methods. This proposed method has the highest accuracy of 89.45% through a 10-fold crossvalidation technique with 31 selected features on the Z-Alizadeh Sani dataset.ConclusionWe demonstrated that SVM combined with genetic optimization algorithm could be lead to more accuracy. Therefore, our study confirms that the GSVMA method outperforms other methods so that it can facilitate CAD diagnosis.

Several methods have been used to accelerate previtellogenesis and vitellogenesis stages in fish, including hormonal induction, sustained-release delivery systems, and oral delivery of gonadotropin-releasing hormone (GnRH). In this study, we proposed the oral administration of GnRH analog + nanoparticles of chitosan to accelerate oogenesis in goldfish as a model fish in reproductive biology and aquaculture. In this regard, adult female goldfish were fed with six experimental groups: chitosan, 50 μg GnRHa/kg b.w., 100 μg GnRHa/kg b.w., chitosan + 50 μg GnRHa/kg b.w., and chitosan + 100 μg GnRHa/kg b.w., and diet without any additive as the control for 40 days in triplicate. Every 10 days, ovarian samples were collected, and gonadosomatic index (GSI), oocyte diameter (OD), zona radiata thickness (Zr), and diameter of the follicular layer (Fl) were measured to assess ovarian developmental stage for each treatment. Additionally, blood sampling was done to measure serum 17β-estradiol concentration at the end of the experiment. All parameters remained unchanged during the experiment in the chitosan-fed group. In the group fed with 100 μg GnRH or chitosan nanoparticle + 100 μg GnRHa, these parameters in general were increased. However, the effects in 50 μg GnRHa or chitosan nanoparticle + 50 μg GnRHa treatments were uncertain; they affected serum E2 levels as a trend toward a significant increase was observed in goldfish treated with chitosan nanoparticle + 100 μg GnRHa. Finally, the results indicated the oral administration of chitosan + 100 μg GnRHa/kg b.w. significantly accelerated the oocyte development and growth of ovary.

In this paper, a new formulation for analyzing free vibration of thin rectangular nanoplates under different thermal conditions is obtained based on the higher-order nonlocal strain gradient theory. Governing equations and non-classical boundary conditions of the nanoplate are derived by using the variational approach. The exact solution is obtained as a function of higher-order and lower-order nonlocal parameters, strain gradient length scale and temperature difference using Navier solution procedure. The influences of small-scale parameters on the vibrational behavior of the nanoplate are investigated for various thermal conditions. High and low temperature conditions are considered to study the effects of changes in temperature and small-scale parameters. It has been shown that increasing the nonlocal parameters decrease the natural frequency of the nanoplate, while increasing the strain gradient length scale will increase it. Also, the natural frequency of the nanoplate will increase by increasing the temperature difference in low temperature conditions, but it will decrease by increasing the temperature difference in high temperature conditions. Non-uniform behaviors are reported for some cases and softening effect and hardening effect are studied. To validate the solutions, the results are compared with previous researches.

Feed deprivation has deleterious effects on fish behavior and stress physiology which may susceptible them to disease outbreak. Functional ingredients in diets may substantially impact the physiology and stress responses of host organisms. Here, we hypothesized that the administration of a dietary prebiotic might attenuate the negative influences of feed deprivation on the behavioral profile of anxiety and physiological responses to stress in zebrafish (Danio rerio). Fish were fed with either basal or mannan-oligosaccharide supplemented (0.4% MOS/kg diet) diets, once per day (normal-control: CN, and normal-prebiotic: PN) or once every other day (starved-control: CS, and starved-prebiotic: PS) for 8weeks. Afterwards, fish were subjected to a novel tank test to measure anxiety. Fish from the CS treatment exhibited more pronounced bottom-dwelling behavior than the other treatments. The number of transitions from the bottom to the top third of the novel tank was significantly higher in PN fish than the CS specimens. No significant differences were found between the CN and PS treatments in all of the anxiety behaviors. CS fish showed higher baseline cortisol levels than the other treatments, which was in line with higher expression of CRH gene in fish subjected to this treatment. Cortisol levels and CRH gene expression of the subjects were also measured after induction of two routine aquaculture stressors. CN and PS fish exhibited similar patterns of cortisol responses at most of the sampling times after stress, and PN specimens showed a significantly lower concentration of cortisol than the other treatments in most cases. Expression of the CRH gene was higher in feed deprived fish immediately after stress induction. Overall, the results show that feed deprivation in some cases influenced anxiety-like behaviors and elevated stress response in zebrafish juveniles; however, the addition of MOS to the diet helped deprived fish exhibit behaviors more typical of normally fed animals.

In this paper different digital audio watermarking techniques have been proposed. Currently, more attention is given to combination of Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD) techniques for watermarking purpose. Available DWT–SVD audio watermarking techniques cannot be applied to speech signals efficiently. However, Linear Predictive Analysis (LPA) technique can model digital speech signals (20–30 ms) in more flexible and efficient ways than DWT. In this paper, a novel digital speech watermarking technique is proposed by applying both LPA and SVD. Quantization Index Modulation (QIM) is further applied to embed the watermark bits. The experimental results show that not only time and memory were reduced significantly as compared to different DWT–SVD audio watermarking techniques, but also the proposed technique was more robust and imperceptible for speech watermarking than other DWT–SVD audio watermarking techniques.

As we previously mentioned about the complexity of the contemporary urban systems, our intention is to determine a framework for extracting and then recognizing each subsystem as constituent layers of a city and see whether each and every layer of a city could play an important role in the measurement of the urban livability.

Due to the complexity of the urban environment, urban planners and designers need to simplify this complex system for the better understanding of the urban changes. The aim of this book is centralized on the use of novel methods in recognizing urban phenomenon and the way to deal with them. Nowadays thanks to the appearance of the AI techniques this aim becomes feasible, up-to-date, and immediate. Therefore we try to introduce AI techniques and their application on the process of urban planning and design by the aim of improving the quality of life within cities.

Recently, computer vision is becoming popular and effective tools for analyzing urban from urban planners and designers perspectives. On the other words, computer vision shed light on the urban modifications and changes in order to study and investigate urban changes rapidly. The application of computer vision not only can be applied as a macro view to pursuit any changes among inter-cities along aerial images but, it also can be utilized to provide micro view on intra-city activities along Closed Circuit Televisions (CCTVs). In macro perspective, computer vision can extract crucial information such as green per capita, footprint, fringe open land, urban/suburban built area, and water surface from Real Aperture Radar (RAR) images.

Temperature has profound effects on physiology of ectothermic animals. However, the effects on temperature variation on behavioral traits are poorly studied in contrast to physiological endpoints. This may be important as even small differences in temperatures have large effects on physiological rates including overall metabolism, and behavior is known to be linked to metabolism at least in part. The primary aim of this study was to determine the effects of ambient temperature on boldness responses of a species of fish commonly used in behavioral experiments, the Siamese fighting fish (Betta splendens). At 26°C, subjects were first examined for baseline behaviors over three days, using three different (but complementary) 'open field' type assays tested in a fixed order. Those same fish were next exposed to either the same temperature (26°C) or a higher temperature (30°C) for 10days, and then the same behavioral assays were repeated. Those individuals exposed to increased temperatures reduced their latency to leave the release area (area I), spent more time in area III (farthest from release area), and were more active overall; together we infer these behaviors to reflect an increase in general 'boldness' with increased temperature. Our results add to a limited number of studies of temperature effects on behavioral tendencies in ectotherms that are evident even after some considerable acclimation. From a methodological perspective, our results indicate careful temperature control is needed when studying behavior in this and other species of fish.

The Siamese fighting fish (Betta splendens) is well known as an aggressive fish with unique spawning and parental care behavior. During reproduction, male fish construct a bubble nest, court females, protect the brood, and defend the territory through aggressive displays. Aggression in male Siamese fighting fish has long been the subject of investigation; however, the kinematics of aggression during contests have been largely overlooked. Here we investigated how nest-holding, male Siamese fighting fish use two different types of displays, gill flaring and fin spreading, towards intruders during various reproductive phases; before (BB) and after bubble nest building, and after spawning (AS), and hatching (AH). Males were more aggressive towards male than female intruders and the level of aggression changed significantly between reproductive phases. Gill flaring, the more energetically costly display, was the dominant initial display towards male and female intruders in BB, AS, AH phases. However, defending males switched to fin spreading after prolonged exposure to intruders. The results suggest that Siamese fighting fish use gill flaring as an acute response in order to defend their territory; this response may be replaced by fin spreading as a chronic response, probably to reduce the energetic costs during the contest.

Numerical abilities have been found to be adaptive in different contexts, including mating, foraging, fighting assessment and antipredator strategies. In species with parental care, another potential advantage is the possibility to adjust parental behavior in relation to the numerosity of the progeny. The finding that many fish vary their parental investment in relation to brood size advocates the existence of a mechanism for appraising offspring number, an aspect that has never been directly investigated. Here we tested the ability of parental female convict cichlid (Amatitlania nigrofasciata) to discriminate between two fry groups differing in number by measuring time spent attempting to recover groups of fry experimentally displaced from the next. Females spent more time trying to recover the fry from larger groups when tested with contrasts 6 versus 12 (1:2) and 6 versus 9 fry (2:3); however, they showed no preference in the 6 versus 8 (3:4) contrast, suggesting that this task exceeds their discrimination capacity.

In this paper, a Multi-Factor Authentication (MFA) method is developed by a combination of Personal Identification Number (PIN), One Time Password (OTP), and speaker biometric through the speech watermarks. For this reason, a multipurpose digital speech watermarking applied to embed semi-fragile and robust watermarks simultaneously in the speech signal, respectively to provide tamper detection and proof of ownership. Similarly, the blind semi-fragile speech watermarking technique, Discrete Wavelet Packet Transform (DWPT) and Quantization Index Modulation (QIM) are used to embed the watermark in an angle of the wavelet's sub-bands where more speaker specific information is available. For copyright protection of the speech, a blind and robust speech watermarking are used by applying DWPT and multiplication. Where less speaker specific information is available the robust watermark is embedded through manipulating the amplitude of the wavelet's sub-bands. Experimental results on TIMIT, MIT, and MOBIO demonstrate that there is a trade-off among recognition performance of speaker recognition systems, robustness, and capacity which are presented by various triangles. Furthermore, threat model and attack analysis are used to evaluate the feasibility of the developed MFA model. Accordingly, the developed MFA model is able to enhance the security of the systems against spoofing and communication attacks while improving the recognition performance via solving problems and overcoming limitations.

Discriminative speaker features vary from one frame to another. Many researcher studied most of the discrimination features for speaker recognition by ignoring phonemic and linguistic information. However, other studies have proposed digital speech watermarking to improve the security of speaker recognition through channel transmission. Although digital speech watermark may be used with speaker information, it can affect speaker information which then can degrade the performance. This paper uses state-of-the-art in frame selection to propose a new way to preserve most of the discriminative features of speaker and to secure speech signals by applying digital speech watermarking technique. In this paper, linear predictive analysis was applied for each frame to extract formants, gain and residual errors. A frequency weighted function was used to quantify formants, and high order correlation as well as error gain is used for weighting the residual errors. The experimental results showed an overall (12 %) effectiveness in terms of performance, time and memory of frame selection technique for speaker recognition and digital speech watermarking.

The effects of pharmacological waste on aquatic ecosystems are increasingly highlighted in ecotoxicology research. Many of these products are designed for human physiology but owing to the conservative nature of vertebrate evolution they also tend to have effects on aquatic organisms and fishes in particular when they find their way into aquatic systems via wastewater effluent. One area of research has focused on reproductive control and the associated hormone treatments. Many of these hormones affect the reproductive physiology of fishes and may cause feminization of male reproductive traits. Alternative medicines have also been widely used particularly in traditional cultures but few of these alternative treatments have been assessed with respect to their potential impact on aquatic ecosystems. Rue (Ruta graveolens) has been used as a male contraceptive in traditional medicines but its effects on fish behavior and reproductive anatomy have yet to be established. Here we show that treating Siamese fighting fish, Betta splendens, with extract of rue has a significant effect on key aggressive/reproductive behaviors and the propensity to explore novel objects (boldness). In all cases the respective behaviors were reduced relative to controls and sham injected fish. Histological analysis of the testes revealed that rue exposure reduced the number of spermatozoa but increased the number of spermatocytes relative to controls.

The present study aimed to detect the pepsinogen gene expression and pepsin enzyme activity, followed by ontogenetic development of gastrointestinal tube with emphasis on the functional stomach in Sobaity larvae (Sparidentex hasta) from hatching to 40 days post- hatching (DPH) reared at 21 °C. The exocrine pancreas including zymogens granules was detected coincident with the gradual yolk-sac absorption (1–3 DPH). The mouth and anus opened at 4 DPH simultaneously with the early convolution of the gut and its segmentation in anterior and posterior intestine, and the buccopharyngeal cavity. The most obvious differentiation in the digestive tract was the formation of an incipient stomach at 12 DPH, surrounded by striated muscle fibers and the development of the gastric glands. The functional stomach appeared 20 DPH onward, where it increased in size and the gastric folds elongated. The stomach was fully developed at 25 DPH with the transition from striated to smooth muscle cells and appearance of gastric glands in the posterior region of the stomach. Pepsinogen was not expressed before 12 DPH and significantly detected coincident with an increase in growth performance and also before the onset of weaning. It increased exponentially parallel to the larval stomach development and concomitant to the change in the diet from live to microbound food. The pepsin enzyme activity, even at a low level, was observed two days after pepsinogen gene expression (14 DPH), increasing sharply after the onset of weaning. Moreover, according to a phylogenetic tree based on the amino acid, the Sobaity’ sequence of pepsinogen was included in the subtype A2, when expressed during late larval development. These results enhanced our understanding of ontogenetic development of pepsinogen in this species, as the pepsinogen expressed earlier than when is expected, and suggest that farmers can initiate weaning at an earlier larval age than is currently practiced, regarding the fact that Sobaity larvae developed a functional stomach 25 DPH onwards.

Let G be a graph with the vertex set { v 1 , … , v n } . The Seidel matrix of G is an n × n matrix whose diagonal entries are zero, i j -th entry is − 1 if v i and v j are adjacent and otherwise is 1. The Seidel energy of G , denoted by E ( S ( G ) ) , is defined to be the sum of absolute values of all eigenvalues of the Seidel matrix of G . Haemers conjectured that the Seidel energy of any graph of order n is at least 2 n − 2 and , up to Seidel equivalence, the equality holds for K n . Recently, this conjecture was proved for n ≤ 12 . We establish the validity of Haemers’ Conjecture in general.