Back Propagation Neural Network Classifier Research Articles

Machine Learning and Optical-Coherence-Tomography-Derived Radiomics Analysis to Predict the Postoperative Anatomical Outcome of Full-Thickness Macular Hole.

Full-thickness macular hole (FTMH) leads to central vision loss. It is essential to identify patients with FTMH at high risk of postoperative failure accurately to achieve anatomical closure. This study aimed to construct a predictive model for the anatomical outcome of FTMH after surgery. A retrospective study was performed, analyzing 200 eyes from 197 patients diagnosed with FTMH. Radiomics features were extracted from optical coherence tomography (OCT) images. Logistic regression, support vector machine (SVM), and backpropagation neural network (BPNN) classifiers were trained and evaluated. Decision curve analysis and survival analysis were performed to assess the clinical implications. Sensitivity, specificity, F1 score, and area under the receiver operating characteristic curve (AUC) were calculated to assess the model effectiveness. In the training set, the AUC values were 0.998, 0.988, and 0.995, respectively. In the test set, the AUC values were 0.941, 0.943, and 0.968, respectively. The OCT-omics scores were significantly higher in the "Open" group than in the "Closed" group and were positively correlated with the minimum diameter (MIN) and base diameter (BASE) of FTMH. Therefore, in this study, we developed a model with robust discriminative ability to predict the postoperative anatomical outcome of FTMH.

Revealing alcoholization-related volatile compounds and determining alcoholization indices in tobacco using GC-IMS coupled with chemometrics

Alcoholization is an integral part of tobacco processing and volatile compounds are key to assessing tobacco alcoholization. In this study, a total of 154 volatiles from nine categories were determined by gas chromatography-ion mobility spectrometry (GC-IMS) from four grades of tobacco, of which 114 were better identified. And then, the dynamic trends of volatile compounds with significant changes in tobacco alcoholization were analyzed. The relevant volatiles with the alcoholization indices (AIs) (R > 0.8) were screened as indicators of tobacco alcoholization. Cinnamyl isobutyrate, linolenic acid alcohol, propanoic acid-M and propanoic acid-D in all tobacco samples were highly correlated with the AIs and tended to increase during the alcoholization process. In addition, linear discriminant analysis (LDA), back-propagation neural network (BPNN) and random forest (RF) classifiers were constructed for discrimination of tobacco AIs. Three classifiers trained with a combination of 20 volatiles achieved satisfactory results with area under the curve (AUC) of 0.95 (LDA), 0.94 (BPNN) and 0.97 (RF), respectively. The RF classifier gained optimal accuracy of 100 % and 96.1 % for the training and test sets, respectively. The study confirmed that GC-IMS can be used to characterize the changes of volatile compounds in tobacco during alcoholization and combined with machine learning to achieve the determination of AIs. The results of the study may provide a new means for the tobacco industry to monitor the alcoholization process and determine the degree of alcoholization.

Two-stage fusion framework driven by domain knowledge for penetration prediction of laser welding

To solve the problem of industrial application difficulties caused by insufficient depth exploration and poor interpretability of the pure data drive neural network. A two-stage fusion framework driven by domain knowledge was proposed and focused on industrial applications: laser welding penetration monitoring. First, a multi-sensor acquisition system based on plasma and molten pool signals is developed, extracting multiple features from plasma and molten pool images. Two Back Propagation (BP) neural network information fusion models are established in the first fusion stage, leveraging the characteristics of the plasma plume and molten pool front. In the second fusion stage, the Dempster-Shafer (DS) evidence theory is employed to fuse the information from the two BP neural network models. The proposed method, achieving a high accuracy of 98.7 %, is validated by comparing its average accuracy and anti-interference capability with classical Convolutional Neural Network (CNN) and BP neural network models. The physical relationship between input and output is elucidated through an established physical model, enhancing the interpretability and anti-interference capability of the model, and thereby improving the stability and accuracy of the system.

A Study on Speech Recognition by a Neural Network Based on English Speech Feature Parameters

In this study, from the perspective of English speech feature parameters, two feature parameters, the mel-frequency cepstral coefficient (MFCC) and filter bank (Fbank), were selected to identify English speech. The algorithms used for recognition employed the classical back-propagation neural network (BPNN), recurrent neural network (RNN), and long short-term memory (LSTM) that were obtained by improving RNN. The three recognition algorithms were compared in the experiments, and the effects of the two feature parameters on the performance of the recognition algorithms were also compared. The LSTM model had the best identification performance among the three neural networks under different experimental environments; the neural network model using the MFCC feature parameter outperformed the neural network using the Fbank feature parameter; the LSTM model had the highest correct rate and the highest speed, while the RNN model ranked second, and the BPNN model ranked worst. The results confirm that the application of the LSTM model in combination with MFCC feature parameter extraction to English speech recognition can achieve higher speech recognition accuracy compared to other neural networks.

Fingerprint Identification with Fusion of Gabor and Minutiae Features Using BPNN Classifier

This study presents an innovative approach to fingerprint identification by leveraging the fusion of Gabor and Minutiae features, employing a Backpropagation Neural Network (BPNN) classifier for accurate categorization. The process involves initial handling of a fingerprint image dataset, including crucial pre-processing steps such as resizing images and implementing morphological operations like dilation, erosion, and opening. Subsequently, Gabor features and minutiae extraction are performed, followed by the fusion of these features to create a comprehensive representation. To address dimensionality concerns, Principal Component Analysis (PCA) is applied. The dataset, comprising the fused features and corresponding labels, is then loaded for the final step - classification using a BPNN. The network is configured for feed-forward backpropagation, distinguishing fingerprint patterns into categories such as arch, left loop, right loop, tented, and whorl. The evaluation metric used to measure the success of the classification process is accuracy. This approach aims to enhance fingerprint recognition by combining distinctive Gabor and minutiae features, ultimately achieving a more robust and precise identification system through the utilization of neural network-based classification. Keywords—BPNN, PCA, Finger Print Images Dataset

Machine learning and optical coherence tomography-derived radiomics analysis to predict persistent diabetic macular edema in patients undergoing anti-VEGF intravitreal therapy

BackgroundDiabetic macular edema (DME) is a leading cause of vision loss in patients with diabetes. This study aimed to develop and evaluate an OCT-omics prediction model for assessing anti-vascular endothelial growth factor (VEGF) treatment response in patients with DME.MethodsA retrospective analysis of 113 eyes from 82 patients with DME was conducted. Comprehensive feature engineering was applied to clinical and optical coherence tomography (OCT) data. Logistic regression, support vector machine (SVM), and backpropagation neural network (BPNN) classifiers were trained using a training set of 79 eyes, and evaluated on a test set of 34 eyes. Clinical implications of the OCT-omics prediction model were assessed by decision curve analysis. Performance metrics (sensitivity, specificity, F1 score, and AUC) were calculated.ResultsThe logistic, SVM, and BPNN classifiers demonstrated robust discriminative abilities in both the training and test sets. In the training set, the logistic classifier achieved a sensitivity of 0.904, specificity of 0.741, F1 score of 0.887, and AUC of 0.910. The SVM classifier showed a sensitivity of 0.923, specificity of 0.667, F1 score of 0.881, and AUC of 0.897. The BPNN classifier exhibited a sensitivity of 0.962, specificity of 0.926, F1 score of 0.962, and AUC of 0.982. Similar discriminative capabilities were maintained in the test set. The OCT-omics scores were significantly higher in the non-persistent DME group than in the persistent DME group (p < 0.001). OCT-omics scores were also positively correlated with the rate of decline in central subfield thickness after treatment (Pearson’s R = 0.44, p < 0.001).ConclusionThe developed OCT-omics model accurately assesses anti-VEGF treatment response in DME patients. The model’s robust performance and clinical implications highlight its utility as a non-invasive tool for personalized treatment prediction and retinal pathology assessment.

Multi-fault diagnosis of district heating system based on PCA_BP neural network

To diagnose multiple faults (blockage and leakage) of a district heating system (MFDHS) in a timely and accurate manner, a new multi-fault diagnosis method is proposed that considers fault diagnosis as a classification problem based on principal component analysis (PCA) and a Back Propagation (BP) neural network. Using this method, a simulation model of heating-network faults is constructed and fault experiments are conducted. Based on the simulation model and experiments, the experimental (ED) and model (MD) datasets are effectively constructed followed by the construction of substitute datasets(SD) that replaces the experimental data with model data. PCA and data standardization are then adopted to process all the datasets. Finally, three BP neural network classification models are trained and tested using the three datasets. To demonstrate the effectiveness of this method, four experimental heating networks, namely a single heat source and branch (SHB), single heat source and single loop (SHSL), double heat source and branch (DHB), and double heat source and double loop (DHDL), are discussed as case studies. The results show that the relative errors of the experimental and simulation data for the pressures and flow rates are less than 5%. The highest prediction accuracy is obtained when the fault to the normal data are 95–5% under training sets to testing sets of 70–30%, while for the four heating networks, the prediction accuracy of the method built with ED is 99.93%, 99.99%, 99.84%, and 99.66% respectively, whereas it reaches 99.99% for MD. The accuracies are 99.99%, 98.13%, 97.4%, and 99.99%, respectively, with MD serving as the training set while another 11400 sets of experimental data serve as the testing set. Accordingly, the accuracy ranges from 95.56% to 88.03%, 97.85–93.29%, 95.65–85.71%, and 100–84.82%, with SD serving as the training set and another 11400 sets of experimental data serving as the testing set.

New Face Recognition System Based on DCT Pyramid and Backpropagation Neural Network

Face recognition has emerged as a prominent biometric identification technique with applications ranging from security to human-computer interaction. This paper proposes a new face recognition system by appropriately combining techniques for improved accuracy. Specifically, it incorporates a discrete cosine transform (DCT) pyramid for feature extraction, statistical measures for dimensionality reduction of the features, and a two-layer backpropagation neural network for classification. The DCT pyramid is used to effectively capture both low- and high-frequency information from face images to improve the ability of the system to recognise faces accurately. Meanwhile, the introduction of statistical measures for dimensionality reduction helps in decreasing the computational complexity and provides better discrimination, leading to more efficient processing. Moreover, the two-layer neural network introduced, which plays a vital role in efficiently handling complex patterns, further enhances the recognition capabilities of the system. As a result of these advancements, the system achieves an outstanding 99 % recognition rate on the Olivetti Research Laboratory (ORL) data set, 98.88 % on YALE, and 99.16 % on AR. This performance demonstrates the robustness and potential of the proposed system for real-world applications in face recognition.

Construction of a digital twin model for vertical storage tank deformation assessment using terrestrial laser scanning technology

The foundational settlement and deformation of vertical storage tanks are crucial factors influencing their safe operation. To enable rapid deformation assessment of storage tanks, this paper combines point cloud data acquired through terrestrial laser scanning with relevant data processing algorithms to construct a digital twin (DT) model. This achieves high-precision automated detection of tank deformation, facilitating the digital transformation of deformation assessment and offering an integrated detection strategy. First, Euclidean distance clustering is applied to the point cloud, and the point density within clusters is statistically analyzed using a Gaussian distribution. This results in a collection of point clusters within one standard deviation, effectively filtering out outliers and noise points, which facilitates the rapid global registration of the point cloud. Second, in order to quickly segment tank point clouds in the scene, back propagation neural network classification learning based on principal component analysis information is used. The point cloud model is combined with the fitting information of slices to generate a DT model, whose deformation can be evaluated through comparison with appropriate storage tank specifications, taking radial deformation, tank inclination, and foundation settlement as indicators.

A novel fault diagnosis method for PV arrays using convolutional extension neural network with symmetrized dot pattern analysis

AbstractPV fault diagnosis remains difficult due to the non‐linear characteristic of PV output, which makes PV output to be likely disturbed by the ambient environment. This study proposes a novel convolutional extension neural network (CENN) algorithm, which is a jointed architecture based on convolutional neural network (CNN) and extension neural network (ENN), takes advantage of CNN and ENN. The CENN is combined with the symmetrized dot pattern (SDP) analysis method to diagnose the common eight PV array faults. The SDP is used to transform the measured PV signals into the point coordinate feature image; then, the CENN is trained to identify the different PV faults. Experimental results show an obvious improvement in short detection times and high accuracy compared with traditional CNN and the histogram of oriented gradient (HOG) extraction method with support vector machine (SVM), K‐nearest neighbours (KNN), and back propagation neural network (BPNN) classifiers, with 95.3%, 94%, 93.5%, and 93.3% accuracy, respectively. Using the proposed CENN, the accuracy can be raised to 97.3%. Additionally, the signals measured by various sensors are collected using programmable logic controller (PLC). The human–machine interface (HMI) and the proposed algorithm are developed using LabVIEW for graphical design. Finally, the information is transmitted to a tablet PC for performing real‐time remote monitoring.

Damage detection method of automobile hub based on image texture feature

With the rapid growth in the number of motor vehicles worldwide, the general public is beginning to attach importance to the quality inspection of wheels before they leave the factory. The current wheel defect detection systems are often cumbersome to operate and have low practical performance. Therefore, this research will use dynamic image segmentation, image texture feature extraction and Back Propagation neural network classification based on wheel image defect feature analysis algorithm to achieve automatic intelligent detection of automotive wheel defects. In this study, an intelligent detection system for automotive wheel defects is also designed, and finally the performance of the detection system is tested experimentally to illustrate its practicality. The experimental results show that the proposed intelligent detection system for automotive wheel defects based on image texture features identifies defects in wheel castings with a correct rate of 96% and a false positive rate of only 2%. This illustrates that the detection system proposed in this study has a high recognition rate and can provide a useful reference for the automotive industry inspection.

Detection of arc grounding fault based on the features of fault voltage

Arc grounding faults commonly occur in power grids, causing extensive power outages and significant economic losses. In this paper, we propose a novel method for detecting arc grounding faults that utilizes a time-frequency-phase mixed feature extraction approach. The mathematical expression for the arc grounding fault voltage is detailed in this paper, and from this, time-domain features that are unaffected by fault transition resistance are extracted. Furthermore, features with clear physical interpretation are extracted from the frequency and phase domains of the fault signal. Our method is able to avoid the subjective feature selection problem that exists in current methods. We conduct fault detection experiments using a topology designed for the purpose and a 10 kV setup to obtain voltage data of arc grounding faults. We then train a fault recognition model using a classic backpropagation neural network. Experimental results demonstrate that the proposed method achieves an accuracy of 98.22% in identifying arc grounding faults. Additionally, we apply this method to identify corona discharge and surface discharge, achieving accuracies of 90.68% for corona discharge and 92.9% for surface discharge.

Intelligent identification on cotton verticillium wilt based on spectral and image feature fusion

BackgroundVerticillium wilt is the major disease of cotton, which would cause serious yield reduction and economic losses, and the identification of cotton verticillium wilt is of great significance to cotton research. However, the traditional method is still manual, which is subjective, inefficient, and labor-intensive, and therefore, this study has proposed a novel method for cotton verticillium wilt identification based on spectral and image feature fusion. The cotton hyper-spectral images have been collected, while the regions of interest (ROI) have been extracted as samples including 499 healthy leaves and 498 diseased leaves, and the average spectral information and RGB image of each sample were obtained. In spectral feature processing, the preprocessing methods including Savitzky-Golay smoothing (SG), multiplicative scatter correction (MSC), de-trending (DT) and mean normalization (MN) algorithms have been adopted, while the feature band extraction methods have adopted principal component analysis (PCA) and successive projections algorithm (SPA). In RGB image feature processing, the EfficientNet was applied to build classification model and 16 image features have been extracted from the last convolutional layer. And then, the obtained spectral and image features were fused, while the classification model was established by support vector machine (SVM) and back propagation neural network (BPNN). Additionally, the spectral full bands and feature bands were used as comparison for SVM and BPNN classification respectively.ResultThe results showed that the average accuracy of EfficientNet for cotton verticillium wilt identification was 93.00%. By spectral full bands, SG-MSC-BPNN model obtained the better performance with classification accuracy of 93.78%. By feature bands, SG-MN-SPA-BPNN model obtained the better performance with classification accuracy of 93.78%. By spectral and image fused features, SG-MN-SPA-FF-BPNN model obtained the best performance with classification accuracy of 98.99%.ConclusionsThe study demonstrated that it was feasible and effective to use fused spectral and image features based on hyper-spectral imaging to improve identification accuracy of cotton verticillium wilt. The study provided theoretical basis and methods for non-destructive and accurate identification of cotton verticillium wilt.

A Social Media Text Analysis Technique Using ML Algorithm

Data mining and ML approaches are used for analyzing the raw data. The analysis consequences has depends on the applications requirements. The use of ML has also been become popular in analyzing the social media data based on NLP for uncovering the sentiments on the social media text post. Social media has two sides brighten and dark. There are a number of creative ways to use the social media, but there are also some users that are accomplishing their toxic intensions using social media. The use of machine learning algorithms with the social media helps us to performing prediction, classification, categorization and finding relationship among social media data attributes. Therefore, in this paper we have exploring the difference among the social media data text mining technique based data analysis and also by using NLP based text classification. In this context a publically available social media dataset from Kaggle has been collected and a data model has been presented to classify social media text using support vector machine (SVM) and backpropagation neural network (BPN) classifiers. In order to extract the features we have used the TF-IDF and in second scenario we have used the Part of speech (POS) tagger. The obtained results demonstrate the performance of BPN based classifier is higher in both the scenarios of feature classification. Additionally in simple subject classification the TF-IDF based features providing more yield as compared to POS based features. Keywords—machine learning algorithm, supervised learning, social media data, NLP based features, classification performance.

Hyperspectral Technique for Detection of Peanut Leaf Spot Disease Based on Improved PCA Loading

Leaf spot disease is a dangerous disease that affects peanut growth, and its severity can significantly impact peanut yield. Hyperspectral-based disease detection technology is a popular non-destructive technique due to its high efficiency, objectivity, and accuracy. In this study, peanut leaf spectra at different levels of severity of leaf spot disease were collected in Liaoning Province, China, in mid-August. This study analyzed the differences in wavelengths using mean spectral reflectance and sensitivity. Using improved principal component analysis loading (I-PCA loading) based on the contribution weight assignment approach, we identified three feature wavelengths of 570 nm, 671 nm, and 750 nm. We evaluated the ability of these feature wavelengths to detect the severity of leaf spot disease using k-nearest neighbor (KNN), support vector machine (SVM), and back-propagation (BP) neural network classifiers. Our experimental results showed that our improved PCA loading method achieved higher classification accuracy with fewer wavelengths than the seven commonly used feature selection methods. Among these classifiers, the SVM achieved the highest accuracy, with an overall accuracy (OA) of 96.88% and a Kappa of 95.81%. Therefore, our proposed method can accurately detect the severity of peanut leaf spot disease and provide scientific and technical support for accurately managing peanut crops.

Diagnosis of Pulmonary Edema and Covid-19 from CT slices using Squirrel Search Algorithm, Support Vector Machine and Back Propagation Neural Network

A Computer Aided Diagnosis (CAD) framework to diagnose Pulmonary Edema (PE) and covid-19 from the chest Computed Tomography (CT) slices were developed and implemented in this work. The lung tissues have been segmented using Otsu’s thresholding method. The Regions of Interest (ROI) considered in this work were edema lesions and covid-19 lesions. For each ROI, the edema lesions and covid-19 lesions were elucidated by an expert radiologist, followed by texture and shape extraction. The extracted features were stored as feature vectors. The feature vectors were split into train and test set in the ratio of 80 : 20. A wrapper based feature selection approach using Squirrel Search Algorithm (SSA) with the Support Vector Machine (SVM) classifier’s accuracy as the fitness function was used to select the optimal features. The selected features were trained using the Back Propagation Neural Network (BPNN) classifier. This framework was tested on a real-time PE and covid-19 dataset. The BPNN classifier’s accuracy with SSA yielded 88.02%, whereas, without SSA it yielded 83.80%. Statistical analysis, namely Wilcoxon’s test, Kendall’s Rank Correlation Coefficient test and Mann Whitney U test were performed, which indicates that the proposed method has a significant impact on the accuracy, sensitivity and specificity of the novel dataset considered. Comparative experimentations of the proposed system with existing benchmark ML classifiers, namely Cat Boost, Ada Boost, XGBoost, RBF SVM, Poly SVM, Sigmoid SVM and Linear SVM classifiers demonstrate that the proposed system outperforms the benchmark classifiers’ results.

Experimental analysis of machine learning methods to detect Covid-19 from x-rays

To automate the detection of covid-19 patients most have proposed deep learning neural networks to classify patients using large databases of chest x-rays. Very few used classical machine learning methods. Machine learning methods may require less computational power and perform well if the data set is small. We experiment with classical machine learning methods on three different data sources varying in size from 55 to almost 4000 samples. We experiment with four feature extraction methods of Gabor, SURF, LBP, and HOG. Backpropagation neural networks and k-nearest neighbor classifiers are combined using one of the four combining methods of bagging, RSM, ARCx4 boosting and Ada-boosting. Results show that using the proper feature extraction and feature selection methods very high performance can be reached using simple backpropagation neural network classifiers. Regardless of combiner method used, the best classification rate achieved was 99.06% for the largest data set, and 100% for the smallest data set.

Rolling Bearing Fault Diagnosis based on DWT-BPNN

For the fault diagnosis of rolling bearings, it is of great significance to improve the diagnostic accuracy. Therefore, this paper presents a rolling bearing fault diagnosis method which combines Daubechies wavelet (DW) with back propagation neural network (BPNN). Specifically, Daubechies wavelet transform is utilized to decompose the vibration signal of the original data in to different frequency components, which can be implemented to extract more prominent fault features. Then, the extracted features are input into BPNN classification model for fault diagnosis by training and testing. Finally, various experiments are carried out on the rolling bearing dataset of Western Reserve University to verify the effectiveness of this method. The results of this study demonstrate that the proposed method is able to reliably identify different fault categories with higher accuracy in comparison with the FT-BPNN methods based on Fourier transform under different loading conditions, and provides a new and effective method for the fault diagnosis of rolling bearings.

Feature selection and classification using bio‐inspired algorithms for the diagnosis of pulmonary emphysema subtypes

AbstractA computer‐assisted diagnosis framework to examine computed tomography (CT) slices for diagnosing pulmonary emphysema is designed. Partitioning of lung tissues and regions of Interest (ROIs) from the CT slices is achieved using Spatial Intuitionistic Fuzzy C‐Means (SIFCM) clustering algorithm. Shape features, texture features, and run‐length features are extracted from each ROI. Feature selection is performed as a wrapper technique by employing manta ray foraging optimization (MRFO) algorithm and random forest (RF) classifier. A backpropagation neural network (BPNN) using gradient descent is used to train the selected features. Adaptive butterfly optimization algorithm (ABOA) is used to fix the optimal topology and parameters, namely weights and biases of the neural network. The BPNN classifier is initialized with the optimized topology and parameters obtained by the ABOA. The developed framework attained an accuracy of 87.52% when tested on an emphysema dataset, which outperforms the BPNN classifier in terms of accuracy, specificity, precision, and recall.

Research on an SAE-based Substation Noise Monitoring Method

In the area of transformer noise monitoring research, condition monitoring technology based on the combination of AI technology and modern signal processing methods gradually becomes a hot research topic. General pattern recognition methods often have higher requirements for signal acquisition and processing and are often restricted by the limited generalization ability of the models. In order to solve this problem, a sparse autoencoder-based transformer state recognition method is proposed. The sparse autoencoder (SAE) is trained by FFT spectral signals, and the intrinsic features of big data are adaptively refined into simple feature functions. Intelligent diagnosis of transformer conditions is realized by the expression of feature functions. The experimental results show that, compared with BP (backpropagation) neural network and SVM (support vector machine) classification algorithms, this method can quickly and effectively improve the accuracy of transformer state recognition, which is of great significance for industrial development.