Principle Of Neural Network Research Articles

Analyzing the Predictability of an Artificial Intelligence App (Tibot) in the Diagnosis of Dermatological Conditions: A Cross-sectional Study.

Artificial intelligence (AI) aims to create programs that reproduce human cognition and processes involved in interpreting complex data. Dermatology relies on morphological features and is ideal for applying AI image recognition for assisted diagnosis. Tibot is an AI app that analyzes skin conditions and works on the principle of a convolutional neural network. Appropriate research analyzing the accuracy of such apps is necessary. This study aims to analyze the predictability of the Tibot AI app in the identification of dermatological diseases as compared to a dermatologist. This is a cross-sectional study. After taking informed consent, photographs of lesions of patients with different skin conditions were uploaded to the app. In every condition, the AI predicted three diagnoses based on probability, and these were compared with that by a dermatologist. The ability of the AI app to predict the actual diagnosis in the top one and top three anticipated diagnoses (prediction accuracy) was used to evaluate the app's effectiveness. Sensitivity, specificity, and positive predictive value were also used to assess the app's performance. Chi-square test was used to contrast categorical variables. P<.05 was considered statistically significant. A total of 600 patients were included. Clinical conditions included alopecia, acne, eczema, immunological disorders, pigmentary disorders, psoriasis, infestation, tumors, and infections. In the anticipated top three diagnoses, the app's mean prediction accuracy was 96.1% (95% CI 94.3%-97.5%), while for the exact diagnosis, it was 80.6% (95% CI 77.2%-83.7%). The prediction accuracy (top one) for alopecia, acne, pigmentary disorders, and fungal infections was 97.7%, 91.7%, 88.5%, and 82.9%, respectively. Prediction accuracy (top three) for alopecia, eczema, and tumors was 100%. The sensitivity and specificity of the app were 97% (95% CI 95%-98%) and 98% (95% CI 98%-99%), respectively. There is a statistically significant association between clinical and AI-predicted diagnoses in all conditions (P<.001). The AI app has shown promising results in diagnosing various dermatological conditions, and there is great potential for practical applicability.

Enhancing Multimodal Patterns in Neuroimaging by Siamese Neural Networks with Self-Attention Mechanism.

The combination of different sources of information is currently one of the most relevant aspects in the diagnostic process of several diseases. In the field of neurological disorders, different imaging modalities providing structural and functional information are frequently available. Those modalities are usually analyzed separately, although a joint of the features extracted from both sources can improve the classification performance of Computer-Aided Diagnosis (CAD) tools. Previous studies have computed independent models from each individual modality and combined them in a subsequent stage, which is not an optimum solution. In this work, we propose a method based on the principles of siamese neural networks to fuse information from Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET). This framework quantifies the similarities between both modalities and relates them with the diagnostic label during the training process. The resulting latent space at the output of this network is then entered into an attention module in order to evaluate the relevance of each brain region at different stages of the development of Alzheimer's disease. The excellent results obtained and the high flexibility of the method proposed allow fusing more than two modalities, leading to a scalable methodology that can be used in a wide range of contexts.

Ideological and Political Function Modeling of College Music Education Based on Improved Neural Network

In the new situation, the state attaches great importance to the ideological and political work in colleges and universities. In order to optimize the function of ideological and political education in colleges and universities, this paper puts forward a modeling method based on improved neural network. Combining ideological and political education with music education organically, this paper studies and calculates the influence relationship and optimizes the structure of ideological and political function model of university music education combined with neural network principle. It puts forward effective strategies to promote the integration of the two.

Fire alarm systems construction on artificial intelligence principles

Coverage of modern views on security trends and directions for developing security systems in the context of global automation and integration into unified automated control systems. An overview of the prospects for using artificial intelligence in security systems to identify difficult to classify situations using signs obtained from monitoring data. Assessment of the prospects for the use of fire alarm systems based on the principles of artificial intelligence based on neural networks (building on the principles of combinatorics of methods of neural networks and odd logic systems). Construction of a mathematical model that describes the process of generating a reliable formation of a signal about the transition of the system to the "fire" state and a description of the problem of correctly setting the system response threshold for signal generation, solving the problem of minimizing the number of false positives by introducing an additional channel that receives initial information about the state of the object in the optical range. Investigation of the reliability of the characteristics of gas fire detectors declared by the manufacturer to determine the possibility of using them as initial elements of fire alarm systems built on the principles of neural networks.

Retracted] Wireless Sensor Network and AI Application for Educational Technology Course

Over the course of its long development, the modern educational technology curriculum has undergone several changes and amassed a lot of information. Theoretically speaking, the state places a strong priority on the use of IT in schools. Students majoring in education should take educational technology courses so that they can learn the characteristics and application techniques of core current information‐based teaching media and incorporate them into their own lesson plans and classroom activities. This will help them meet the information needs of today’s classrooms as they evolve with the advent of educational modernization and availability of educational information. Thus, this research employs a wireless sensor network (WSN) to gather and send data on ed tech classes and then employs AI to assess those classes’ quality and guide real‐time changes to how they are taught and complete the following tasks: (1) The development status of educational technology courses and WSN at home and abroad is introduced. (2) The application of WSN in teaching is introduced, the basic principle of GRU neural network and related optimization algorithms is expounded, and the quality evaluation system of educational technology courses is constructed. (3) The IPSO‐Adam‐GRU evaluation model improves the GRU neural network’s hyperparameters with the help of the improved PSO approach and Adam gradient descent. The model is fed test data for evaluation, and the findings are compared to those from an expert’s evaluation to determine how well the model performs. The results demonstrate that the model established for this article is superior to others since it provides a more accurate assessment.

Nonlinear neural networks adaptive control for a class of fractional-order tuberculosis model.

The problem of nonlinear adaptive control for a class of fractional-order tuberculosis (TB) model is studied in this paper. By analyzing the transmission mechanism of TB and the characteristics of fractional calculus, a fractional-order TB dynamical model is established with media coverage and treatment as control variables. With the help of universal approximation principle of radial basis function neural networks and the positive invariant set of established TB model, the expressions of control variables are designed and the stability of error model is analyzed. Thus, the adaptive control method can guarantee that the number of susceptible and infected individuals can be kept close to the corresponding control targets. Finally, the designed control variables are illustrated by numerical examples. The results indicate that the proposed adaptive controllers can effectively control the established TB model and ensure the stability of controlled model, and two control measures can protect more people from tuberculosis infection.

Enhancing Emotion Recognition through Deep Learning and Brain-Computer Interface Technology

Deep learning, a specialized branch of machine learning, uses complex neural networks for advanced data processing tasks in which feature extraction and progression require more than one layer of processing using a Neural Network. The Brain Computer Interface is a direct communication link between the brain's electrical activity and an external equipment. The actual brain data is gathered using the BCI technology with electrodes fitted in Brain cells and the research focuses on the recognition of multimodal emotions after feature extraction and multilayer processing with an Artificial Neural Network (ANN). Several organizations have lately released enormous datasets containing experimental data on people's physiological signals (EEGs, eye movements) while they experience different emotions. These datasets collections are being exclusively designed to encourage the development of successful deep-learning emotion identification algorithms where the volunteers had to do some set of tasks regarding the research purpose. In this study, we evaluate deep learning approaches more specifically by the Artificial Neural Network (ANN) using the SEED-IV Electroencephalogram (EEG) dataset. In this paper, we briefly represented the implementation of Emotion Recognition using Artificial Neural Network (ANN), which is one of the topics of our research. Furthermore, we explore the operating principles of artificial neural networks, which are our primary model for identifying emotions. Finally, we offer a few proposals for implementing neural networks on EEG inputs. With the ANN and 3D convolutional layers, we finally achieved a 63.425% accuracy rate.

A multi-ion interference decoupling model based on ion-selective electrode arrays.

Compared to traditional liquid-junction ion-selective electrodes, solid-contact ion-selective electrodes (SC-ISEs) have attracted much attention and undergone rapid development due to their compactness and ease of integration. However, the application and widespread use of SC-ISEs are limited by their non-ideal selectivity and susceptibility to signal drift. Although the principles of artificial neural network (ANN) methods have shown significant progress in partially resolving the selectivity issue, they generally require extensive calibration steps and computational resources to implement. As a result, numerical computation models are more practical and economical, but existing approaches often overlook experimental phenomena and have relatively complex modeling principles. In this study, we propose a proportional factor model based on the trend of SC-ISEs affected by the multiple ions, along with a scalable dynamic correction procedure to improve its robustness. This model utilizes an estimated response surface method to solve nonlinear equations, requiring fewer calibration experiments. It accurately extracts the concentrations of multiple target ions in the presence of multi-ion interference and dynamically adjusts the model parameters for different types of ISEs. Additionally, we design a multi-channel SC-ISE array as a carrier for theoretical validation. In a case study, we demonstrate the feasibility of decoupling multiple ions using the SC-ISE array, obtaining concentrations of calcium, magnesium, sodium, and potassium ions, and verify the accuracy of the multi-ion detection system for real-world water samples.

Photovoltaic Bypass Diode Fault Detection Using Artificial Neural Networks

Due to the importance of determining faulty bypass diodes in photovoltaic systems, faulty bypass diodes have been of widespread interest in recent years due to their importance to improving PV system durability, operation, and overall safety. This paper presents new work in developing an artificial intelligence (AI) based model using the principles of artificial neural networks (ANN) to detect short and open PV bypass diodes fault conditions. With only three inputs from the PV system, namely the output power, short-circuit current, and open-circuit voltage, the developed ANN model can determine whether the PV bypass diodes are defective. In the experimentally validated case of short and open bypass diodes, 93.6% and 93.3% of faulty bypass diodes can be detected. Furthermore, the developed ANN model has an average precision and sensitivity of 96.4% and 92.6%, respectively.

Система для розпізнавання елементів одягу та їх кольорів на зображенні

The paper introduces a system for recognizing clothing items and their colors, developed using the Java programming language in the form of a web application, employing the Model-View-Controller (MVC) architecture. For development and project assembly convenience, the Gradle automatic build system was utilized. The system operates based on artificial neural network principles on the Clarifai platform, thoroughly examining the process of object detection in images and the system's real-world performance. InceptionV2 serves as the base model for image processing, incorporating the FPN technique to analyze images at different scales, resizing images up to 512 pixels, and training using stochastic gradient descent with hard negative mining. The system employs a median cut algorithm to determine the dominant color of clothing items. Determining the dominant color of each clothing item is done separately, with a focus on achieving 100 % recognition accuracy. An approach to determining the nearest color name based on the dominant color is implemented. This approach is based on calculating the Euclidean distance between two points in a three-dimensional space, iterating through 140 colors with RGB color model names to find the nearest color name. However, there are certain limitations in finding the nearest color name that result in a reduced accuracy of 60 %. The impact of various factors, such as lighting and image quality, is thoroughly examined in the context of their influence on the system's performance. The user interface is designed as an intuitive tool for interacting with the system, allowing users to check recognized clothing items and identified colors. Additionally, recommendations on basic color combinations that users can employ to enhance their clothing style are provided. The work includes an example of the system's application on a real image, visually demonstrating the results and describing the quality of clothing item and color recognition. One of the key features of the presented system is its flexibility and scalability. Recognition quality can be further improved through additional model training on larger datasets. The paper also addresses data processing and result analysis optimization. The novelty of the research results lies in the development of a comprehensive clothing item and color recognition system using artificial neural networks and the implementation of an approach to finding the nearest color with name. Completeness of the system refers to the entire image processing cycle within a single system, including clothing item location determination, clothing item classification, dominant color determination, and finding the nearest color with name for visual display to the user, along with providing basic color combinations for color correction or reassurance of color combinations' correctness. The practical significance of the research results lies in the development of the application's structure and algorithms, the software implementation of the system, which changes the approach to selecting and verifying color combinations. This approach provides users with clear color names, allowing them to create their own accurate color combinations for clothing sets or effectively search for clothing items online based on received colors. The system can also be used as a "data labeling" tool for preparing materials used in training machine learning models or neural networks. Future prospects include integration with other systems, expanding the image database to improve accuracy, and utilizing additional data sources to enhance the system's functionality.

Research Progress of Artificial Neural Network and Its Application in Fault Diagnosis of Chemical Industry

Many characteristics exhibited by artificial neural networks, such as nonlinearity, large scale, strong parallel processing ability, as well as robustness, fault tolerance, and strong self-learning ability, make it attractive for fault detection and diagnosis in complex systems. The relationship between the complex process, cumbersome process, and measurable process variable failure causes of chemical process is very complicated. Once a failure occurs, it will cause huge economic losses and casualties. The emergence of artificial neural network provides a new chemical fault diagnosis technology, which can carry out early and accurate fault detection and diagnosis for chemical process and equipment, so as to improve the efficiency and safety of production. This paper introduces the basic principle and development history of artificial neural network, as well as several typical artificial neural networks, such as back propagation algorithm (BP network), radial basis network (RBF network), and their application in chemical process fault diagnosis.

Development of CNN-LSTM combinational architecture for COVID-19 detection.

The world has been under extreme pressure due to the spread of the coronavirus. The urgency to eradicate the virus has caused distress amongst civilians and medical agencies to an equal extent. Due to anomalies observed in the results from reverse transcription-polymerase chain reaction (RTPCR) tests, more reliable options like computed tomography (CT) scan-based tests are being researched upon. In this paper, a novel combinational architecture is built upon the principles of Convolution Neural Networks (CNN) and Long Short Term Memory (LSTM) Networks to detect COVID-19 virus. This method uses chest X-ray images as inputs to combinational architecture for the classification of samples. The CNN part of the network will be used to extract features that help in the classification, and the LSTM part will be used for classification based on the extracted features. A total of 8 convolutional layers and 4 pooling layers are used for CNN and 4 LSTM layers of 64 and 128 cells respectively. Instead of the sigmoid function, a rectified linear unit function is used as an activation function. This provides non-linearity to the CNN and better accuracies in comparison. The proposed model employs a padding layer to prevent the loss of information. Accuracy, loss, F1 score, and Matthew's Correlation Coefficient (MCC) are calculated to analyse the effectiveness of the proposed architecture. The proposed model is validated using a relatively larger dataset of 7292 images. The combinational architecture provides a more informative and truthful result in the evaluation of classification as it caters to both the size of positive elements and negative elements in the dataset. The proposed CNN-LSTM model gives an accuracy of 98.91% and an MCC value of 97.84% respectively. The model is also compared with models employing transfer learning methods for similar applications.

RETRACTED: Application of Convolutional Neural Network in Modern Technology Field and Improvement by Time-space Version

Based on the basic principle of convolutional neural networks, this paper gradually introduces the development of convolutional neural networks and its related application fields. We also analyze and compare the iterations and developments of convolutional neural networks in three most commonly used areas: medical treatment, face recognition and transportation. A large number of articles have been written to understand the variations and differences of convolutional neural networks in these three areas, such as the use of different training methods or different structures, such as the time-space convolutional neural networks that will be mentioned in the article. And the last part is the summary.

Face Recognition Based on Convolutional Neural Networks

Since science and technology have been progressing steadily in recent years, deep learning's potential applications have expanded greatly. From unlocking the screen of a phone with a human face to driverless technology, which has emerged in recent years. Facial recognition is proving to be a boon to life. Among various deep learning algorithms, the appearance of convolutional neural network (CNN)has made unprecedented progress in image recognition. In this paper, the basic principles of convolutional neural networks are explained, and the most important concepts are introduced. The convolutional neural network is used for experiments. The input layer, convolution layer, pooling layer, fully connected layer, and output layer are the nine layers that make up the traditional and complete convolutional neural network model, which is used as the experimental foundation. LFW dataset is used for training, and the experimental results are given. At the end of the paper, the accuracy and loss functions are analyzed and the accurate results of facial recognition are achieved.

Optimization prediction and experimental verification of cyclone performance based on BP neural network

As a kind of centrifugal separation equipment, hydrocyclone has developed rapidly with the advantages of high economic benefits, good compactness, high processing efficiency, simple process and so on. Hydrocyclone has broad practical significance and application prospects in chemical, light industry, oil mining and refining, environmental protection, medicine, food processing, ship transportation and surface oil spill treatment. BP neural networks are mainly used in the following four aspects: function approximation, pattern recognition, classification, and data compression. In this paper, we introduce the principle of BP neural network, establish an analytical model, apply BP neural network to predict the performance of hydrocyclone, and conduct experimental verification to optimize the hydrocyclone under different working conditions according to the experimental results.

Nigeria Generating Station Capacity and Load Demand Requirement Using Regression Analysis (Least Square) and Artificial Neuro-Fuzzy Inference System (Anfis) for Reliable Power Supply

The increasing demand of electrical energy is growing exponential rate while there is mismatched between power supply and power demand. This means that there is strong record to determine the existing state of the system for purpose of future projection of power generation to avoid system collapse. Evidently, to conduct the analysis of system forecast for energy demand, many mathematical techniques and artificial machine learning which is based on artificial intelligence. This paper considered the application of least – square regression technique and artificial neuro-fuzzy inference system (ANFIS) and artificial neural network (ANN) principle, on the view to train the data set for Nigeria energy consumption (residential. Commercial and industrial) on the view to determine total system consumption pattern for reliable power supply. The data used ranges from (2000-2012) as input for the projection plan of 2013-2035, was determined for Nigeria power system. This is modelled and simulated in matlab software tool (version 19.0). the purposed artificial intelligence system based on Artificial neural fuzzy inference system (ANFIS) which provided high machine learning fixtures to predicts close to the actual value as compared to the traditional least square algorithms. The results were evaluated using mean absolute percentage error (MAPE) of 5.73%. The proposed technique ANFIS predicted load forecast (2013-2035) to provide energy data for electric utilities and to plan successfully for efficient power generation. Application of Adaptive Neuron – fuzzy inference in artificial intelligence involved both neural network (NN) and fuzzy logic (FL) principle together an adaptive neuro-fuzzy inference system (ANFIS), is an artificial neural network that is based on a fuzzy inference system. ANFIS is a very useful system to extract numerical model from numerical data since it integrates both neural network and fuzzy logic principle together, it is capable of adapting the benefits of both in a single framework. An adaptive neuro-fuzzy inference system (ANFIS) is an application of adaptive neuro-fuzzy logic that uses frame work of artificial intelligence (AI). Hence its inference system corresponds to a code of IF – THEN RULE in fuzzy neuro network. It has a learning capability to approximate non-linear functions hence adaptive neurofuzzy inference system (ANFIS) deserves to be considered as good estimator.

Application of Artificial Neural Network in Optimal Design of Reactor

Reactor is widely used in biology, chemical industry, metallurgy, environmental protection and other fields, playing an irreplaceable role. With the development of science and technology and the concept of green development, the application of artificial neural network to optimize the reactor reaction conditions has become a trend. Artificial neural network plays an important role in reactor optimization because of its strong fault tolerance, the ability to express complex nonlinear relations and perform complex operations. This paper will briefly describe the basic principle and research progress of artificial neural network, and its application in reactor design.

Artificial Intelligence-Based Interactive Art Design under Neural Network Vision Valve

Interactive art design (IAD) is an organic integration of art and technology. From the perspective of the development of AI machines from ancient times to the present, it has gone through the stages of command interface, graphical interface, and multimedia interface. The development of interactive art has been around for some years. As an art form, it not only brings optimization and enjoyment to people’s quality of life but also meets the needs of human-computer interaction, improves the efficiency of art design, and realizes human-human interaction. The purpose is to bring different feelings and experiences to the works and people’s psychology. In particular, the application of AI in IAD has not only brought great changes to designers. It also derives the interaction of behavioral limbs, which brings a greater experience and interaction to the audience, thereby creating a better interactive effect. Therefore, this paper completes the following work: (1) the research progress of AI in IAD at home and abroad is introduced. (2) The combination of AI and IAD is proposed, and the basic principle of RBF neural network and intelligent optimization algorithm are introduced, and the evaluation index of IAD is constructed. (3) Using the constructed dataset to test two intelligent optimization algorithms, the results show that the PSO-RBF model is more excellent in evaluating the quality of IAD. The trained model is used for experiments, and the output of the model is compared with the expert evaluation results, and the error is very small. Comparing the quality indicators of IAD before and after the integration of AI, the results show that AI has an excellent improvement effect on IAD.

Sparse phase retrieval using a physics-informed neural network for Fourier ptychographic microscopy.

In this paper, we report a sparse phase retrieval framework for Fourier ptychographic microscopy using the recently proposed principle of physics-informed neural networks. The phase retrieval problem is cast as training bidirectional mappings from the measured image space with random noise and the object space to be reconstructed, in which the image formation physics and convolutional neural network are integrated. Meanwhile, we slightly modify the mean absolute error loss function considering the signal characteristics. Two datasets are used to validate this framework. The results indicate that the proposed framework is able to reconstruct sparsely sampled data using a small aperture overlapping rate without additional data driving whereas conventional methods cannot.

Evaluation of the Modern Value of Imperial Examination Culture in the Context of Cultural Confidence Based on Deep Learning Models

National prosperity, national spiritual independence, and cultural security greatly impact the cultural self-confidence. It is also a new era issue that must be answered in building a socialist cultural power. In this paper, we scientifically define the core concepts of the imperial examination system, imperial examination culture, and imperial examination cultural heritage. We systematically discuss the connotation and extension of the imperial examination cultural heritage, analyze the positive energy and negative effects of the imperial examination culture, propose the construction of a new type of examination culture system, and innovate the content and form of the examination. We conduct an in-depth study on the modern value of the imperial examination culture in the context of cultural confidence and then construct appropriate evaluation indicators to use the deep learning model to evaluate the value. The technical principle and performance optimization method of DBN neural network are introduced and then the modern value evaluation system of imperial examination culture is constructed. The parameters of the DBN model are optimally selected through experiments. After the optimal model is trained, the sample data are input to obtain the results. The results are compared with the evaluation results of human experts to judge the performance of the model. In many cases, the results are very close to, or the same as the expert results, in some of the experiments, the accuracy of our model varied from the expert results by a small margin of up to 0.4%.