Neural Learning Algorithm Research Articles

Classification and Severity Level Assessment of Fusarium Wilt Disease in Chickpeas using Convolutional Neural Network

Background: The fusarium wilt disease of chickpea leaves is a common illness that leads to economic problems for farmers due decreased crop yield. Early disease detection and the implementation of suitable precautions can help to increase the yield of chickpeas. This study offers an improved method for Fusarium wilt disease prediction based on severity level using a convolutional neural learning algorithm. Methods: The Convolutional Neural Network (CNN) model is utilized in this work to identify leaf disease due to wilting. The dataset contains 4,339 images of chickpea leaves that were obtained from Kaggle. After preprocessing, the data is sent into the network model for training. The model shows acceptable classification and accuracy metrics. Result: Deep learning methods are very useful tools for tracking leaf diseases at their early stages and can help farmers with the use of controlling methods. The proposed work looks for changes in the shape and color of chickpea leaves in order to predict severe fusarium disease. Training and validation accuracies show a balanced trade-off by giving satisfactory outcomes. The model shows an overall accuracy of 74.79%. The confusion matrix and classification parameters increase the model’s performance.

Distributed Neural Learning Algorithms for Multiagent Reinforcement Learning

In this paper, the fully distributed neural learning algorithms by neural network approximation for networked multi-agent reinforcement learning (NMARL) are studied. To tackle the convergence analysis of methods in NMARL with tremendous state-action space, most of the existing distributed algorithms are designed by linear function approximation, which however would fall into a situation of poor expression. To conquer such limitation, the distributed neural learning algorithms are developed by using a novel neural network approximation that bridges the theory and practice of deep multi-agent reinforcement learning (DMARL). Specifically, inspired by the overparametrization method for minimizing mean-squared projected bellman error (MSPBE), the distributed neural learning algorithms with population semigradients and stochastic semigradients are respectively proposed to solve the NMARL problem. Furthermore, the convergence of the proposed algorithms are strictly given by employing the overparametrization method to establish the approximate stationary point of MSPBE to characterizes the algorithms toward the global optimum. Finally, some numerical simulations demonstrate the effectiveness of the distributed neural learning algorithms.

Study on the Relationship between Resistivity and the Physical Properties of Seafloor Sediments Based on the Deep Neural Learning Algorithm

The occurrence of deep-sea geohazards is accompanied by dynamic changes in the physical properties of seafloor sediments. Therefore, studying the physical properties is helpful for monitoring and early warnings of deep-sea geohazards. Existing physical property inversion methods have problems regarding the poor inversion accuracy and limited application scope. To address these issues, we establish a deep learning model between the resistivity of seafloor sediment and its density, water content, and porosity. Compared with empirical formulas, the deep learning model has the advantages of a more concentrated prediction range and a higher prediction accuracy. This algorithm was applied to invert the spatial distribution characteristics and temporal variation of the seafloor sediment density, water content, and porosity in the South China Sea hydrate test area for 12 days. The study reveals that the dynamic changes in the physical properties of seafloor sediments in the South China Sea hydrate zone exhibit obvious stratification characteristics. The dynamic changes in the physical properties of seafloor sediments are mainly observed at depths of 0–0.9 m below the seafloor, and the sediment properties remain stable at depths of 0.9–1.8 m below the seafloor. This study achieves the monitoring and early warning of dynamic changes in the physical properties of seafloor sediments and provides a guarantee for the safe construction of marine engineering.

An Observer-Based Stiffness Estimation for Space Target Capture by Space Robot with Controllable Damping Mechanism

The space target capturing task using the spacecraft-manipulator system (SMS) has special significance in on-orbit servicing due to its theoretical challenges and practical value. The contact force between the end effector (gripper) and the target exerted by the tumbling motion of the space target destabilizes the spacecraft base. A full-dimensional controllable damping mechanism (FDCDM) with a cross-axis structure was designed to buffer the transient impact force on the end joint. The introduction of a damping mechanism gives the space robot a variable stiffness and damping system, and a stiffness estimation algorithm is proposed to calibrate the system stiffness, as stiffness cannot be measured directly. The full-dimensional controllable damping mechanism (FDCDM) with a cross-axis structure is equivalent to a four-DOF tandem joint, and the whole-body dynamic model of the SMS endowed with a full-dimensional controllable damping mechanism (FDCDM) was established using the Kane equation. Then, an unknown input observer (UIO)-based identification theory is proposed to precisely estimate the internal flexibility torque and the corresponding joint stiffness. A model-based neural learning algorithm is proposed to update the variable parameter matrix of the observer. The simulation experiment results demonstrate that the flexibility torque and joint stiffness could be accurately estimated within the expected error, illustrating the feasibility and effectiveness of the proposed method.

Correction: A Lightweight Neural Learning Algorithm for Real-Time Facial Feature Tracking System via Split-Attention and Heterogeneous Convolution

Correction: A Lightweight Neural Learning Algorithm for Real-Time Facial Feature Tracking System via Split-Attention and Heterogeneous Convolution

A Lightweight Neural Learning Algorithm for Real-Time Facial Feature Tracking System via Split-Attention and Heterogeneous Convolution

A Lightweight Neural Learning Algorithm for Real-Time Facial Feature Tracking System via Split-Attention and Heterogeneous Convolution

AI-based automatic segmentation of craniomaxillofacial anatomy from CBCT scans for automatic detection of pharyngeal airway evaluations in OSA patients

This study aims to generate and also validate an automatic detection algorithm for pharyngeal airway on CBCT data using an AI software (Diagnocat) which will procure a measurement method. The second aim is to validate the newly developed artificial intelligence system in comparison to commercially available software for 3D CBCT evaluation. A Convolutional Neural Network-based machine learning algorithm was used for the segmentation of the pharyngeal airways in OSA and non-OSA patients. Radiologists used semi-automatic software to manually determine the airway and their measurements were compared with the AI. OSA patients were classified as minimal, mild, moderate, and severe groups, and the mean airway volumes of the groups were compared. The narrowest points of the airway (mm), the field of the airway (mm2), and volume of the airway (cc) of both OSA and non-OSA patients were also compared. There was no statistically significant difference between the manual technique and Diagnocat measurements in all groups (p > 0.05). Inter-class correlation coefficients were 0.954 for manual and automatic segmentation, 0.956 for Diagnocat and automatic segmentation, 0.972 for Diagnocat and manual segmentation. Although there was no statistically significant difference in total airway volume measurements between the manual measurements, automatic measurements, and DC measurements in non-OSA and OSA patients, we evaluated the output images to understand why the mean value for the total airway was higher in DC measurement. It was seen that the DC algorithm also measures the epiglottis volume and the posterior nasal aperture volume due to the low soft-tissue contrast in CBCT images and that leads to higher values in airway volume measurement.

Construction of a Landscape Design and Greenery Maintenance Scheduling System Based on Multimodal Intelligent Computing and Deep Neural Networks.

Digitalization brings challenges and new opportunities to the development of landscape gardening, “smart gardening,” which is a product of landscape gardening in response to the development of the digital era. Based on the multimodal intelligent computing method and deep neural network machine learning algorithm, this paper adopts “digital landscape design logic” to analyze and research smart gardens and digital design. The digital landscape design process and methods are discussed based on design logic, design basis, environment analysis, and results presentation, and the greenery maintenance scheduling system is constructed. The paper focuses on the digital implementation of the environmental analysis of the site and uses Rhino software and Grasshopper visual programming language to build parametric logic, establish parametric analysis models, and conduct a comprehensive analysis of the current environment. The main theme of the whole paper is a logical approach to digital landscape design for smart gardens, using digital technology tools from the perspective of smart garden thinking, combining quantitative analysis and qualitative design, and intervening in digital landscape garden planning and design to explore the application of digital technology and tools.

Functional mimicry of Ruffini receptors with fibre Bragg gratings and deep neural networks enables a bio-inspired large-area tactile-sensitive skin

Collaborative robots are expected to physically interact with humans in daily living and the workplace, including industrial and healthcare settings. A key related enabling technology is tactile sensing, which currently requires addressing the outstanding scientific challenge to simultaneously detect contact location and intensity by means of soft conformable artificial skins adapting over large areas to the complex curved geometries of robot embodiments. In this work, the development of a large-area sensitive soft skin with a curved geometry is presented, allowing for robot total-body coverage through modular patches. The biomimetic skin consists of a soft polymeric matrix, resembling a human forearm, embedded with photonic fibre Bragg grating transducers, which partially mimics Ruffini mechanoreceptor functionality with diffuse, overlapping receptive fields. A convolutional neural network deep learning algorithm and a multigrid neuron integration process were implemented to decode the fibre Bragg grating sensor outputs for inference of contact force magnitude and localization through the skin surface. Results of 35 mN (interquartile range 56 mN) and 3.2 mm (interquartile range 2.3 mm) median errors were achieved for force and localization predictions, respectively. Demonstrations with an anthropomorphic arm pave the way towards artificial intelligence based integrated skins enabling safe human–robot cooperation via machine intelligence.

Role of Artificial Intelligence in Parenteral Formulation: A Review

Artificial Intelligence is defined as the division of computer science which tackles the simulation of human intelligence processes by machines and refers to the usage of automatic algorithms, machine vision, and technologies in various sectors of the pharmaceutical field. As this intelligent system has a handful of applications it is utilized by pharma industries for various pharmaceutical product developments mainly in sterile preparations, such as parenterals, vaccines, seras, and other blood products. In our country, parenterals are the second most widely used means for drug delivery after oral dosage forms. Therefore, the quality of the parenterals is crucial. In the present scenario, inspection of parenterals products during the manufacturing is quite important. Aiming at that difficulty hence it is considered for combining the manual process with automated technologies like artificial neural networks, deep learning, and clustering algorithms. In this way, liquid preparations can be more accurately identified and the accuracy of detection can be greatly improved.

Characteristics, Main Controlling Factors and Prediction of Complex Fracture–Cavity Reservoirs of Buried Carbonate Hills in the Weixinan Depression, Western South China Sea

Buried hill oil reservoirs have become a key area for offshore oil and gas exploration. In this paper, a typical oil field in the western South China Sea is used as the research object, and a study on the characterization, cause of formation and prediction of the fracture–cavity reservoir distribution is carried out. The reservoir in the study area is a complex fracture–cavity reservoir that developed due to weathering and leaching, tectonic movement and dissolution reconstruction on the limestone skeleton. The reservoir spaces are composed of karst caves, fractures and pores. The main controlling factors include lithological changes, karst landforms, tectonic deformation and faulting. To address the controlling mechanisms of the lithological changes on the formation of fracture–cavity reservoirs, a new parameter, the lithology standard deviation, to evaluate lithological changes is proposed based on the characteristics of the lithological changes, and the distribution of these lithological changes is portrayed in combination with the seismic attributes. The tectonic deformation principal curvature inversion algorithm is used to simulate the distribution of the tectonic principal curvature at the top of the Carboniferous. The larger the tectonic principal curvature is, the stronger the deformation of the rock formation and the more favorable the conditions are for fracture formation. The karst geomorphology controls the overall reservoir distribution, and the karst highlands and karst slope areas are the zones with the most–developed secondary pore space (or fractures and karst caves). The faulting control area is the fracture and dissolution pore development area, major faults control the distribution of the karst cave reservoirs, and secondary faults influence the formation of fractures in the faulted area. The study predicts and evaluates the distribution of fracture–cavity reservoirs from the perspective of fracture–cavity genesis quantification, and by gridding and normalizing the four major genesis quantification evaluation parameters and fusing the geological factors that control the formation of fractures and karst caves by using Back–Propagation neural network deep learning algorithms, a method for predicting the distribution of fracture–cavity reservoirs constrained by geological genesis analysis is developed.

Artificial intelligence methods for improving the inventive design process, application in lattice structure case study

Abstract Nowadays, firms are constantly looking for methodological approaches that help them to decrease the time needed for the innovation process. Among these approaches, it is worth mentioning the TRIZ-based frameworks such as the Inventive Design Methodology (IDM), where the Problem Graph method is used to formulate a problem. However, the application of IDM is time-consuming due to the construction of a complete map to clarify a problem situation. Therefore, the Inverse Problem Graph (IPG) method has been introduced within the IDM framework to enhance its agility. Nevertheless, the manual gathering of essential information, including parameters and concepts, requires effort and time. This paper integrates the neural network doc2vec and machine learning algorithms as Artificial Intelligence methods into a graphical method inspired by the IPG process. This integration can facilitate and accelerate the development of inventive solutions by extracting parameters and concepts in the inventive design process. The method has been applied to develop a new lattice structure solution in the material field.

Development of a weed detection system using machine learning and neural network algorithms

The detection of weeds at the stages of cultivation is very important for detecting and preventing plant diseases and eliminating significant crop losses, and traditional methods of performing this process require large costs and human resources, in addition to exposing workers to the risk of contamination with harmful chemicals. To solve the above tasks, also in order to save herbicides and pesticides, to obtain environmentally friendly products, a program for detecting agricultural pests using the classical K-Nearest Neighbors, Random Forest and Decision Tree algorithms, as well as YOLOv5 neural network, is proposed. After analyzing the geographical areas of the country, from the images of the collected weeds, a proprietary database with more than 1000 images for each class was formed. A brief review of the researchers' scientific papers describing the methods they developed for identifying, classifying and discriminating weeds based on machine learning algorithms, convolutional neural networks and deep learning algorithms is given. As a result of the research, a weed detection system based on the YOLOv5 architecture was developed and quality estimates of the above algorithms were obtained. According to the results of the assessment, the accuracy of weed detection by the K-Nearest Neighbors, Random Forest and Decision Tree classifiers was 83.3 %, 87.5 %, and 80 %. Due to the fact that the images of weeds of each species differ in resolution and level of illumination, the results of the neural network have corresponding indicators in the intervals of 0.82–0.92 for each class. Quantitative results obtained on real data demonstrate that the proposed approach can provide good results in classifying low-resolution images of weeds.

Biometric Authentication from Photic Stimulated EEG Records

ABSTRACT Studies on brain biometrics have shown that electroencephalogram (EEG) signals encourage more secure authentication. The uniqueness, persistence, universality and robustness of EEG signals against fraud attacks offer potential for highly secure biometric systems. However, more studies are needed to improve collectibility, stability, performance, and acceptability of brain biometrics. EEG signals of healthy subjects were generally used in previous studies. However, adequate studies have not been conducted on subjects who are not mentally healthy. Moreover, EEG signals were usually recorded from healthy subjects at resting, thinking, visual stimulation, and imagery states. In this study, unlike other studies, photic stimuli with EEG data were used for the first time in order to identify both subjects with attention deficit hyperactivity disorder (ADHD) and healthy subjects. In the proposed method, power densities of 1–49 Hz frequencies of EEG segments were obtained by applying periodogram power spectral density estimation method using Kaiser Window to raw EEG data. These values were used for training of 1D Convolutional Neural Network deep learning algorithms. Classification success of the proposed method was measured as 97.17%. These results proved that EEG data obtained from subjects by applying photic stimuli can be used in EEG-based identification systems.

APPLICATION OF MACHINE LEARNING ALGORITHMS FOR CLASSIFICATION OF WEED VARIETIES

The article considers the relevance of the introduction of intelligent weed detection systems, in order to save herbicides and pesticides, as well as to obtain environmentally friendly products. A brief review of the researchers' scientific works is carried out, which describes the methods of identification, classification and discrimination of weeds developed by them based on machine learning algorithms, convolutional neural networks and deep learning algorithms. This research paper presents a program for detecting pests of agricultural land using the algorithms K-Nearest Neighbors, Random Forest and Decision Tree. The data set is collected from 4 types of weeds, such as amaranthus, ambrosia, bindweed and bromus. According to the results of the assessment, the accuracy of weed detection by the classifiers K-Nearest Neighbors, Random Forest and Decision Tree was 83.3%, 87.5%, and 80%. Quantitative results obtained on real data demonstrate that the proposed approach can provide good results in classifying low-resolution images of weeds.

A bagging dynamic deep learning network for diagnosing COVID-19

COVID-19 is a serious ongoing worldwide pandemic. Using X-ray chest radiography images for automatically diagnosing COVID-19 is an effective and convenient means of providing diagnostic assistance to clinicians in practice. This paper proposes a bagging dynamic deep learning network (B-DDLN) for diagnosing COVID-19 by intelligently recognizing its symptoms in X-ray chest radiography images. After a series of preprocessing steps for images, we pre-train convolution blocks as a feature extractor. For the extracted features, a bagging dynamic learning network classifier is trained based on neural dynamic learning algorithm and bagging algorithm. B-DDLN connects the feature extractor and bagging classifier in series. Experimental results verify that the proposed B-DDLN achieves 98.8889% testing accuracy, which shows the best diagnosis performance among the existing state-of-the-art methods on the open image set. It also provides evidence for further detection and treatment.

Transaction Fraud Detection (Anomaly detection) using Machine Learning

Now a day’s online shopping crazy thing for peoples. As well as many people uses Online transaction for many purposes. Transaction fraud is growing seriously. Therefore, the study on fraud detection is interesting and significant and we can say necessary. An important way of detecting fraud is to extract the behavior profiles (BPs) of users based on their historical transaction records, and then to verify if an incoming transaction is a fraud or not in view of Their BPs. Also we apply SVM, Adaboost, and Neural Network machine learning algorithm to see which one is giving the best result.

Design of MRI structured spiking neural networks and learning algorithms for personalized modelling, analysis, and prediction of EEG signals

This paper proposes a novel method and algorithms for the design of MRI structured personalized 3D spiking neural network models (MRI-SNN) for a better analysis, modeling, and prediction of EEG signals. It proposes a novel gradient-descent learning algorithm integrated with a spike-time-dependent-plasticity algorithm. The models capture informative personal patterns of interaction between EEG channels, contrary to single EEG signal modeling methods or to spike-based approaches which do not use personal MRI data to pre-structure a model. The proposed models can not only learn and model accurately measured EEG data, but they can also predict signals at 3D model locations that correspond to non-monitored brain areas, e.g. other EEG channels, from where data has not been collected. This is the first study in this respect. As an illustration of the method, personalized MRI-SNN models are created and tested on EEG data from two subjects. The models result in better prediction accuracy and a better understanding of the personalized EEG signals than traditional methods due to the MRI and EEG information integration. The models are interpretable and facilitate a better understanding of related brain processes. This approach can be applied for personalized modeling, analysis, and prediction of EEG signals across brain studies such as the study and prediction of epilepsy, peri-perceptual brain activities, brain-computer interfaces, and others.

1131-P: Clinical Prediction Models for MODY Genetic Testing in the United States Monogenic Diabetes Registry

The University of Chicago National Monogenic Diabetes Registry houses a novel dataset of people presenting with possible or known monogenic diabetes and has over a decade of data points, despite the rarity of the disease relative to T1DM and especially T2DM. Currently, determination of genetic testing, including which specific panel or genes is most appropriate to sequence, is a manual process introducing variability and subjectivity. Aim: To test and compare performance of several models against the manual process and each other in identifying maturity-onset diabetes of the young (MODY) from available clinical information. Methods: Statistical models for prediction of monogenic diabetes were applied to clinical data for cases suspected of having MODY. The models tested in the study with this dataset ranged from the traditional methods of regression to more modern approaches of neural nets, classification trees, clustering, and other machine learning algorithms for classification of MODY against type 1 and type 2 and a possible subset of monogenic diabetes subtypes. Results: Using a sample size of N=330, with 70 testing positive for either GCK, HNF1A, or HNF4A, exploratory data models were able to achieve 75-80% validation accuracy (random validation data set at 10-20% of N) while internal human classification accuracy was roughly 42% over the same samples. The traditional logistic regression and neural net had equivalent performance. Conclusions: The 3750 participants within the Registry and more than a decade of follow-up presents a unique opportunity to assess which clinical features are most predictive in identifying those with monogenic diabetes. The results of this work will direct future testing efforts within the Registry to assist in efficient diagnosis of these uncommon yet clinically important forms of diabetes. Disclosure R. P. Mulligan: None. L. R. Letourneau-freiberg: None. T. L. Bowden: None. P. Tian: None. B. Kandasamy: None. L. H. Philipson: Advisory Panel; Self; Nevro Corp., Research Support; Self; Provention Bio, Inc. S. W. Greeley: None. R. N. Naylor: None. Funding National Institute of Diabetes and Digestive and Kidney Diseases (R01DK104942, P30DK020595)

Intelligent Learning Algorithm for English Flipped Classroom Based on Recurrent Neural Network

Reading and writing are the foundations of English learning as well as an important method of instruction. With the advancement of network technology and the onset of the information age, an increasing number of students have lost interest in traditional English reading and writing instruction in the classroom. Flipped classrooms have emerged as a result of this situation and have become the focus of research in one fell swoop. As a result, flipped classroom research at home and abroad has primarily focused on the theory and practical application of flipped classrooms, and flipped classroom application practice is primarily based on the overall classroom, with few separate discussions on the effects of flipped classroom students’ self‐learning. As a result, we developed a recurrent neural network‐based intelligent assisted learning algorithm for English flipped classrooms. There are two main characteristics of the model. First, it is a gated recurrent unit based on a variant structure of the recurrent neural network. The double‐gating mechanism fully considers the context and selects memory through weight assignment, and on this basis, it integrates the novel LeakyReLU function to improve the model’s training convergence efficiency. Second, by overcoming time‐consuming problems in the medium, the adoption of the connection sequence classification algorithm eliminates the need for prior alignment of speech and text data, resulting in a direct boost in model training speed. The experimental results show that in the English flipped classroom’s intelligent learning mode, students explore and discover knowledge independently, their enthusiasm and interest in learning are greatly increased, and the flipped classroom’s teaching effect is greatly improved.