Diagnosis of dermatophytosis in cats using artificial neural networks

BACKGROUND: Currently, artificial intelligence in the form of artificial neural networks is being actively implemented in a number of areas of our lives, including medicine. In particular, in otorhinolaryngology, artificial neural networks are used to analyze images obtained during endoscopic examinations of patients (e.g., videolaryngoscopy) [1–3]. The interpretation of laryngoscopic images often presents significant difficulties for practicing physicians, which reduces the frequency of detection of precancerous laryngeal diseases and contributes to the increase in the number of patients with stage III–IV laryngeal cancer [4, 5]. This underscores the significance of prompt performance and accurate interpretation of the findings of endoscopic examinations of patients with laryngeal disorders. Artificial neural networks can be employed to analyze the results of videolaryngoscopy, furnishing the physician with supplementary information that can enhance diagnostic accuracy and diminish the probability of error [6, 7]. AIM: The study aims to develop and train an artificial neural network for recognizing characteristic features of laryngeal neoplasms and variants of laryngeal normality. MATERIALS AND METHODS: The study was conducted under the grant of the Moscow Center for Innovative Technologies in Healthcare (grant No. 2112-1/22) entitled “Using Neural Networks (Artificial Intelligence Algorithms) for Control and Improving the Quality of Diagnosis and Treatment of Diseases of Laryngeal and Ear Structures through Digital Technologies”.The following methods were used during the course of the study: data collection for the creation of a photobank (dataset) of medical images obtained during videolaryngoscopy; data partitioning for the formation of datasets for individual nosologies and groups of diseases; the method of consilium; analysis of the accuracy of recognition and classification of digital endoscopic images; and training of classification neural networks. Consequently, a dataset comprising 1,471 laryngeal images in digital formats (JPEG, BMP) was assembled, labelled, and uploaded for the purpose of training the artificial neural network. Of the total number of images, 410 were classified as pertaining to laryngeal formation, while 1061 were classified as variants of normality. Subsequently, the neural network was trained and tested to identify the signs of normal and laryngeal masses. RESULTS: The results of the testing of the artificial neural network indicated the formation of an inaccuracy matrix, the calculation of the value of recognition accuracy, the calculation of the quality indicators of the model performance, and the construction of the ROC curve. The developed and trained artificial neural network demonstrated an accuracy of 86% in recognizing the signs of laryngeal masses and norms. CONCLUSIONS: This study demonstrates that a trained artificial neural network can successfully distinguish between signs of normal and laryngeal masses in endoscopic photographs. With further training of the neural network and achievement of high accuracy, this technology can be used in clinical practice as an assistant in the interpretation of laryngoscopic images and early diagnosis of laryngeal masses. It can also be employed to control and improve the quality of diagnosis and treatment of diseases of the throat, nose, and ears by primary care physicians.

<abstract><p>The training of artificial neural networks (ANNs) with rectified linear unit (ReLU) activation via gradient descent (GD) type optimization schemes is nowadays a common industrially relevant procedure. GD type optimization schemes can be regarded as temporal discretization methods for the gradient flow (GF) differential equations associated to the considered optimization problem and, in view of this, it seems to be a natural direction of research to <italic>first aim to develop a mathematical convergence theory for time-continuous GF differential equations</italic> and, thereafter, to aim to extend such a time-continuous convergence theory to implementable time-discrete GD type optimization methods. In this article we establish two basic results for GF differential equations in the training of fully-connected feedforward ANNs with one hidden layer and ReLU activation. In the first main result of this article we establish in the training of such ANNs under the assumption that the probability distribution of the input data of the considered supervised learning problem is absolutely continuous with a bounded density function that every GF differential equation admits for every initial value a solution which is also unique among a suitable class of solutions. In the second main result of this article we prove in the training of such ANNs under the assumption that the target function and the density function of the probability distribution of the input data are piecewise polynomial that every non-divergent GF trajectory converges with an appropriate rate of convergence to a critical point and that the risk of the non-divergent GF trajectory converges with rate 1 to the risk of the critical point. We establish this result by proving that the considered risk function is <italic>semialgebraic</italic> and, consequently, satisfies the <italic>Kurdyka-Łojasiewicz inequality</italic>, which allows us to show convergence of every non-divergent GF trajectory.</p></abstract>

Artificial neural networks (ANN) have been widely used in the field of data classification. Normally, training of neural network is applied with the traditional back propagation technique. As, this approach has various drawbacks, training of neural network is done with Particle Swarm Optimization (PSO). PSO has been widely used to solve the diverse kind of optimization problems. Population initialization performs a significant role in meta-heuristic algorithms. This paper describes a new initialization population approach Log Logistic termed as PSOLL-NN to create the initialization of the swarm. The proposed algorithm has been tested for weight optimization of feed forward neural network; and compared with back propagation Algorithm (BPA), standard PSO (PSO-NN), PSO initialized with Halton Sequence (PSOH-NN), Torus sequence (PSOT-NN) and Sobol sequence (PSOS-NN). The experimental results show that the proposed technique performed exceptionally better than the other traditional techniques. Moreover, the outcome of our work presents a foresight that how the proposed initialization technique can be used as an efficient alternative to standard training approaches for the data classification problems.

Comparing large number of metaheuristics for artificial neural networks training to predict water temperature in a natural river

Background. There are a large number of neural networks that have their advantages and disadvantages, for example, simple, fast and easy to use single-stranded perceptrons are suitable for linear and linearized regression tasks, and more complicated neural networks are expendable in training and prediction time. Therefore, the problem arises for the development of fast and efficient algorithms for training artificial neural networks. An additional factor for researching new methods for training neural networks is finding the smallest training and prediction errors.Objective. The aim of the paper is to search and analyze the properties of the most effective method of training artificial neural networks using a combined approximation of the response surface. Another step is to perform computational experiments on proposed artificial neural networks and compare the results of experiments with known and developed methods.Methods. Analysis of known methods of combined approximation of the response surface was used. New algorithms for training neural networks, based on clustering of data using k-means method were developed. The algorithm with the smallest errors of artificial neural network learning and data prediction is chosen.Results. The results of research of different methods of training of artificial neural networks are given. Peculiarities of the methods of combined approximation of the response surface are analyzed. It is shown that the two methods of combined approximation of the response surface for training of artificial neural networks and prediction confirm the effectiveness of the proposed approach. Combined approximation algorithm is selected, which provides the lowest learning and forecasting errors.Conclusions. It was investigated that developed methods of combined approximation of the response surface allow training neural networks and predicting data with less error than when using autoregressive model with moving average, multilayer perceptron or artificial neural networks of models of geometric transformations without additional data processing.

In order to obtain optimal control of a real object, it is necessary to know the precise mathematical model of this control object. In the present study an artificial neural network is used for building a mathematical model of the control object. First, some forms of control are defined, and with the help of these controls, the control object is modeled. The obtained values of the controls and the space state vector are stored to create a training sample. The artificial neural network is then trained on this training set. For a trained neural network, a set of optimal control problems is solved. The optimal control obtained by the trained artificial neural network is applied to a real control object. The accuracy of the approximation of the mathematical model by an artificial neural network can be estimated based on the proximity of the functional values of the control object and the trained neural network.

This paper presents a novel technique for the supervised training of feed-forward artificial neural networks (ANN) using the Harmony Search (HS) algorithm. HS is a stochastic meta-heuristic that is inspired from the improvisation process of musicians. Unlike Backpropagation, HS is non-trajectory driven. By modifying an existing improved version of HS & adopting a suitable ANN data representation, we propose a training technique where two of HS probabilistic parameters are determined dynamically based on the best-to-worst (BtW) harmony ratio in the current harmony memory instead of the improvisation count. This would be more suitable for ANN training since parameters and termination would depend on the quality of the attained solution. We have empirically tested and verified our technique by training an ANN with a benchmarking problem. In terms of overall training time and recognition, our results have revealed that our method is superior to both the original improved HS and standard Backpropagation.

Purpose of research. The goal of the work is to develop and justify a cognitive peer-to-peer infrastructure that will improve the conditions for collective work on projects based on agile methodology. Cognitive architecture is defined as a structure that ensures the implementation of anthropomorphic and neuromorphic functions in natural or artificial systems. The proposed approach is based on organizing the interaction of the collective intelligence of members of an agile team and artificial intelligence, represented by trained artificial neural networks. When forming an agile team, it is proposed to take into account the structure of the cognitive sphere in the structure of the mental processes of a human cognitive agent.Methods. Domain knowledge is determined based on the collective intelligence of the agile team members and the training of artificial neural networks. It is assumed that artificial neural networks are available to all members of an agile team and implement the functions of collective artificial intelligence, provided that their training uses the professionalism and experience of a person in a natural social environment. Mental operations such as analysis, partitioning (modularization), comparison, abstraction, synthesis, generalization, classification, concretization, known from general psychology courses, are interpreted not only as a result of human activity, but also as the functionality of a program. Some elements of the cognitive sphere processes “memory” and “speech” are realized in a similar way.Results. The system is implemented on the basis of a peer-to-peer computer network that provides communications between all artificial and natural participants in the cognitive process during the design process. A conceptual model of a cognitive collective intelligence cell is proposed, combining elements of the actual collective intelligence of agile agents with the collective artificial intelligence of agents based on neural networks. In an expert assessment of the quality of individual design stages, it was proposed to use tagging based on the emotional-volitional and motivational mental processes of individuals.Conclusion. Cognitive information processing is based on the idea of modeling human thinking processes in computer systems. In the system under consideration, this includes natural language processing, written speech recognition, associated with understanding information through software imitation of human intelligence. The accepted concept involves the implementation of collective intelligence not only artificially, but also by organizing convenient interaction between participants in an intellectual chat. Artificial intelligence, also collective, is implemented using initially trained and further trained neural networks.

Tailor-made blanks are sheet metal assemblies with different thicknesses and/or materials and/or surface coatings. A monolithic sheet can be machined to make the required thickness variations that is referred as tailor machined blanks. Due to the thickness variation in tailor machined blanks, laser bending of these blanks is more complicated than monolithic plates. In this article, laser forming of tailor machined blanks is investigated and an artificial neural network (ANN) will be configured to predict the bending angle of laser formed tailor machined blanks. The input parameters of neural network are selected as start point of scan path, laser irradiating method, laser beam diameter, laser output power and number of radiation passes. The results show that a 5×8×1 trained neural network can predict the bending angle with acceptable accuracy. Comparison of the randomly selected tests with experimental results shows 1.1% error in the prediction of bending angle by trained artificial neural network.

Through the success of deep learning in various domains, artificial neural networks are currently among the most used artificial intelligence methods. Taking inspiration from the network properties of biological neural networks (e.g. sparsity, scale-freeness), we argue that (contrary to general practice) artificial neural networks, too, should not have fully-connected layers. Here we propose sparse evolutionary training of artificial neural networks, an algorithm which evolves an initial sparse topology (Erdős–Rényi random graph) of two consecutive layers of neurons into a scale-free topology, during learning. Our method replaces artificial neural networks fully-connected layers with sparse ones before training, reducing quadratically the number of parameters, with no decrease in accuracy. We demonstrate our claims on restricted Boltzmann machines, multi-layer perceptrons, and convolutional neural networks for unsupervised and supervised learning on 15 datasets. Our approach has the potential to enable artificial neural networks to scale up beyond what is currently possible.

Surface deflection is a phenomenon that causes fine wrinkles on the outer surfaces of sheet metal and deteriorates product external appearance. It is quantitatively defined as the difference between the section curve of the sheet and the ideal curve. In this study, using neural networks, a prediction model for surface deflection according to material properties was constructed and combined with a genetic algorithm; the combination of the material properties was studied to predict the minimum surface deflection. Because of the limited number of simulation data, neural networks were developed using several sampling methods such as central composite design, Latin hypercube sampling, and random sampling. In the training of the neural networks, the optimal hyper-parameter of the neural network was found automatically using Latin hypercube sampling. In conclusion, for prediction of surface deflection in rectangular embossing, neural networks made by central composite design showed the best performance. In addition, it was confirmed that the procedure of combining automatic training of a neural network and the genetic algorithm accurately predicted the set of material properties that generates the minimum surface deflection. Also, the quantity of surface deflection predicted by the neural network was very close to that predicted by finite element analysis.

Approaches presented today in the scientific literature suggest that there are no methodological solutions based on the training of artificial neural networks to predict the direction of industrial development, taking into account a set of factors—innovation, environmental friendliness, modernization and production growth. The aim of the study is to develop a predictive model of performance management of innovative industrial systems by building neural networks. The research methods were correlation analysis, training of neural networks (species—regression), extrapolation, and exponential smoothing. As a result of the research, the estimation efficiency technique of an innovative industrial system in a complex considering the criteria of technical modernization, development, innovative activity, and ecologization is developed; the prognostic neural network models allow to optimize the contribution of signs to the formation of target (set) values of indicators of efficiency for macro and micro-industrial systems that will allow to level a growth trajectory of industrial systems; the priority directions of their development are offered. The following conclusions: the efficiency of industrial systems is determined by the volume of sales of goods, innovative products and waste recycling, which allows to save resources; the results of forecasting depend significantly on the DataSet formulated. Although multilayer neural networks independently select important features, it is advisable to conduct a correlation analysis beforehand, which will provide a higher probability of building a high-quality predictive model. The novelty of the research lies in the development and testing of a unique methodology to assess the effectiveness of industrial systems: it is based on a multidimensional system approach (takes into account factors of innovation, environmental friendliness, modernization and production growth); it combines a number of methodological tools (correlation, ranking and weighting); it expands the method of effectiveness assessment in terms of the composition of variables (previously presented approaches are limited to the aspects considered).

The paper describes an evolutionary approach to artificial neural network (NN) training, which is used to determine the state of oil-production equipment. A new artificial NN weight coefficient coding method using multi- chromosomes is proposed. The genetic operators of crossingover and mutation applied to multi-chromosomes are examined. A genetic algorithm structure of artificial NN training based on the developed genetic operators is proposed. A comparison of the proposed approach to NN training with existing ones has been carried out.

This paper presents a real time feedforward control scheme of a squirrel cage induction motor. This scheme uses an artificial neural network (ANN). The objective of this controller is to force the rotor speed to follow an arbitrarily prescribed trajectory. The proposed neural network structure is first trained to identify the inverse dynamics of the drive system. Then the trained neural network is used as a feedforward controller to generate both the input voltage and frequency for the motor to follow the desired trajectory. The training data is obtained from a laboratory setup which implements an LSI circuit (HEF4752V), a PWM inverter, and an induction motor. The main advantage of the proposed scheme is that it does not need a detailed and elaborate model of the drive system. The proposed system is capable of achieving accurate tracking control of the speed even when the nonlinear parameters of the motor and the load are unknown. These unknown nonlinear parameters are captured by the trained artificial neural network. The architecture and the training algorithm of the neural network are presented and discussed. The effectiveness of the proposed drive system is investigated using a laboratory model. Laboratory results showed a very simple and reliable tracking control system.

Efficient reliability analysis considering uncertainty in random field parameters: Trained neural networks as surrogate models