Final Classification Accuracy Research Articles

Early Prediction of Reading Risk in Fourth Grade: A Combined Latent Class Analysis and Classification Tree Approach

ABSTRACT Purpose Fourth grade typically involves shifting the instruction from learning to read to reading to learn, which can cause students to struggle. However, early reading intervention guided by assessment has demonstrated effectiveness in preventing later reading difficulties (RD). This study presents a classification and regression tree (CART) model predicting fourth-grade reading groups using first-grade measures. Method Students were assessed in first and fourth grade (N = 452). Fourth-grade groups were determined using latent class analysis based on word reading and reading comprehension measures with a cut-point at the 15th percentile. A CART model was trained to determine the best decision rules to classify students at risk of developing later RD and compared to a logistic regression model. Results Important first-grade predictors included a mix of oral language and foundational word-reading skills with final classification accuracy estimates of .90 AUC, .91 sensitivity, and .75 specificity. Conclusion While the CART and logistic regression models’ classification accuracy was similar, CART has the advantage of offering a more intuitive way for practitioners to determine risk. Multivariate screening can be time-consuming, but CART models offer the potential to reduce false positives and guide targeted interventions, leading to better use of school resources.

Tobacco shred varieties classification using Multi-Scale-X-ResNet network and machine vision.

The primary task in calculating the tobacco shred blending ratio is identifying the four tobacco shred types: expanded tobacco silk, cut stem, tobacco silk, and reconstituted tobacco shred. The classification precision directly affects the subsequent determination of tobacco shred components. However, the tobacco shred types, especially expanded tobacco silk and tobacco silk, have no apparent differences in macro-scale characteristics. The tobacco shreds have small size and irregular shape characteristics, creating significant challenges in their recognition and classification based on machine vision. This study provides a complete set of solutions aimed at this problem for screening tobacco shred samples, taking images, image preprocessing, establishing datasets, and identifying types. A block threshold binarization method is used for image preprocessing. Parameter setting and method performance are researched to obtain the maximum number of complete samples with acceptable execution time. ResNet50 is used as the primary classification and recognition network structure. By increasing the multi-scale structure and optimizing the number of blocks and loss function, a new tobacco shred image classification method is proposed based on the MS-X-ResNet (Multi-Scale-X-ResNet) network. Specifically, the MS-ResNet network is obtained by fusing the multi-scale Stage 3 low-dimensional and Stage 4 high-dimensional features to reduce the overfitting risk. The number of blocks in Stages 1–4 are adjusted from the original 3:4:6:3 to 3:4:N:3 (A-ResNet) and 3:3:N:3 (B-ResNet) to obtain the X-ResNet network, which improves the model’s classification performance with lower complexity. The focal loss function is selected to reduce the impact of identification difficulty for different sample types on the network and improve its performance. The experimental results show that the final classification accuracy of the network on a tobacco shred dataset is 96.56%. The image recognition of a single tobacco shred requires 103 ms, achieving high classification accuracy and efficiency. The image preprocessing and deep learning algorithms for tobacco shred classification and identification proposed in this study provide a new implementation approach for the actual production and quality detection of tobacco and a new way for online real-time type identification of other agricultural products.

Grassland Use Intensity Classification Using Intra-Annual Sentinel-1 and -2 Time Series and Environmental Variables

Detailed spatial data on grassland use intensity is needed in several European policy areas for various applications, e.g., agricultural management, supporting nature conservation programs, improving biodiversity strategies, etc. Multisensory remote sensing is an efficient tool to collect information on grassland parameters. However, there is still a lack of studies on how to process, combine, and implement large radar and optical image datasets in a joint observation framework to map grassland types on large heterogeneous study areas. In our study, we assessed the usefulness of 2521 Sentinel-1 and 586 Sentinel-2 satellite images and topographic data for mapping grassland use intensity. We focused on the distinction between intensively and extensively managed permanent grassland in a large heterogeneous study area in Slovenia. We provided dense Satellite Image Time Series (SITS) for 2017, 2018 and 2019 to identify important differences, e.g., management practices, between the two grassland types analysed. We also investigated the effectiveness of combining two different remote-sensing products, the optical Normalised Difference Vegetation Index (NDVI) and radar coherence. Grassland types were distinguished using an object-based approach and the Random Forest classification. With the use of SITS only, the models achieved poor performance in the case of cloudy years (2018). However, the performance improved with additional features (environmental variables). The feature selection method based on Mean Decrease Accuracy (MDA) provided a deeper insight into the high-dimensional multisensory SITS. It helped select the most relevant features (acquisition dates, environmental variables) that distinguish between intensive and extensive grassland types. The addition of environmental variables improved the overall classification accuracy by 7–15%, while the feature selection additionally improved the final overall classification accuracy (using all available features) by 2–3%. Although the reference dataset was limited (1259 training samples), the final overall classification accuracy was above 88% in all years analysed. The results show that the proposed Random Forest classification using combined multisensor data and environmental variables can provide better and more stable information on grasslands than single optical or radar data SITS on large heterogeneous areas. Therefore, a combined approach is recommended to distinguish different grassland types.

Convolutional neural network-based automatic cervical vertebral maturation classification method.

This study aimed to develop a fully automated artificial intelligence-aided cervical vertebral maturation (CVM) classification method based on convolutional neural networks (CNNs) to provide an auxiliary diagnosis for orthodontists. This study consisted of cephalometric images from patients aged between 5 and 18 years. After grouping them into six cervical stages (CSs) by orthodontists, a data set was constructed for analyzing CVM using CNNs. The data set was divided into training, validation, and test sets in the ratio of 70, 15, and 15%. Four CNN models namely, VGG16, GoogLeNet, DenseNet161, and ResNet152 were selected as the candidate models. After training and validation, the models were evaluated to determine which of them is most suitable for CVM analysis. Heat maps were analyzed for a deeper understanding of what the CNNs had learned. The final classification accuracy ranking was ResNet152>DenseNet161>GoogLeNet>VGG16, as evaluated on the test set. ResNet152 proved to be the best model among the four models for CVM classification with a weighted κ of 0.826, an average AUC of 0.933 and total accuracy of 67.06%. The F1 score rank for each subgroup was: CS6>CS1>CS4>CS5>CS3>CS2. The area of the third (C3) and fourth (C4) cervical vertebrae were activated when CNNs were assessing the images. CNN models proved to be a convenient, fast and reliable method for CVM analysis. CNN models have the potential to provide automatic auxiliary diagnostic tools in the future.

Urban Land Use and Land Cover Change Analysis Using Random Forest Classification of Landsat Time Series

Efficient implementation of remote sensing image classification can facilitate the extraction of spatiotemporal information for land use and land cover (LULC) classification. Mapping LULC change can pave the way to investigate the impacts of different socioeconomic and environmental factors on the Earth’s surface. This study presents an algorithm that uses Landsat time-series data to analyze LULC change. We applied the Random Forest (RF) classifier, a robust classification method, in the Google Earth Engine (GEE) using imagery from Landsat 5, 7, and 8 as inputs for the 1985 to 2019 period. We also explored the performance of the pan-sharpening algorithm on Landsat bands besides the impact of different image compositions to produce a high-quality LULC map. We used a statistical pan-sharpening algorithm to increase multispectral Landsat bands’ (Landsat 7–9) spatial resolution from 30 m to 15 m. In addition, we checked the impact of different image compositions based on several spectral indices and other auxiliary data such as digital elevation model (DEM) and land surface temperature (LST) on final classification accuracy based on several spectral indices and other auxiliary data on final classification accuracy. We compared the classification result of our proposed method and the Copernicus Global Land Cover Layers (CGLCL) map to verify the algorithm. The results show that: (1) Using pan-sharpened top-of-atmosphere (TOA) Landsat products can produce more accurate results for classification instead of using surface reflectance (SR) alone; (2) LST and DEM are essential features in classification, and using them can increase final accuracy; (3) the proposed algorithm produced higher accuracy (94.438% overall accuracy (OA), 0.93 for Kappa, and 0.93 for F1-score) than CGLCL map (84.4% OA, 0.79 for Kappa, and 0.50 for F1-score) in 2019; (4) the total agreement between the classification results and the test data exceeds 90% (93.37–97.6%), 0.9 (0.91–0.96), and 0.85 (0.86–0.95) for OA, Kappa values, and F1-score, respectively, which is acceptable in both overall and Kappa accuracy. Moreover, we provide a code repository that allows classifying Landsat 4, 5, 7, and 8 within GEE. This method can be quickly and easily applied to other regions of interest for LULC mapping.

Identification of antioxidants in edible oil by two-dimensional correlation spectroscopy combined with deep learning

The quality of edible oil is an essential part of food safety which is highly concerned by people. In this study, perturbation Raman spectroscopy combined with deep learning was used to identify antioxidants in edible oils. Convolutional neural network (CNN) and recurrent neural network (RNN) are two classical network structures in deep learning. First of all, we explored the identification effect of antioxidants in edible oils using one-dimensional Raman data combined with one-dimensional CNN and RNN. At the same time, we also compared the identification effect of the data set under a single heating time disturbance. Then two-dimensional correlation spectroscopy combined with a two-dimensional CNN model was used to identify the types of antioxidants. It was found that the final classification accuracy reached 97%, which was nearly 10% higher than the one-dimensional CNN model. This showed that the two-dimensional correlation spectral analysis based on external disturbance can “amplify” the subtle differences in the spectral data. In addition, the traditional chemometric method, partial least squares discriminant analysis (PLS-DA), was used as a control experiment. According to this study, it can be seen that the perturbation spectrum combined with deep learning was feasible in the detection of trace substances in edible oils.

A Review of Recent Developments in Driver Drowsiness Detection Systems.

Continuous advancements in computing technology and artificial intelligence in the past decade have led to improvements in driver monitoring systems. Numerous experimental studies have collected real driver drowsiness data and applied various artificial intelligence algorithms and feature combinations with the goal of significantly enhancing the performance of these systems in real-time. This paper presents an up-to-date review of the driver drowsiness detection systems implemented over the last decade. The paper illustrates and reviews recent systems using different measures to track and detect drowsiness. Each system falls under one of four possible categories, based on the information used. Each system presented in this paper is associated with a detailed description of the features, classification algorithms, and used datasets. In addition, an evaluation of these systems is presented, in terms of the final classification accuracy, sensitivity, and precision. Furthermore, the paper highlights the recent challenges in the area of driver drowsiness detection, discusses the practicality and reliability of each of the four system types, and presents some of the future trends in the field.

Improved Binary Adaptive Wind Driven Optimization Algorithm-Based Dimensionality Reduction for Supervised Classification

Typically, Feature Selection (FS) is adopted as a critical preprocessing step in most pattern recognition and data mining tasks. It helps to avoid the acute impact of irrelevant and redundant features on the performance of the classification model under consideration. To tackle this problem, researchers have proposed different methods for selecting the most significant features to improve the overall classification accuracy for a given dataset by extracting relevant information. Realistically, for a given dataset with a large number of features, conventional methods usually struggle to find good solutions. Therefore, in this study, a meta-heuristic algorithm called the Wind Driven Optimization (WDO) is enhanced and then cloned into a binary variant, the improved Binary Adaptive WDO (iBAWDO). The proposed iBAWDO would manage to select the most relevant (near-optimal) features while reducing the computational cost and enhancing (or even maintaining) the final classification accuracy. An evolutionary crossover technique as well as the Simulated Annealing algorithm (SA) are incorporated into the original WDO algorithm to enhance its search ability for feasible regions as well as exploitation within these regions, respectively. To assess the relevance of the features selected, two popular classifiers, k-Nearest Neighbor (k-NN) and Support Vector Machine (SVM), are adopted as fitness evaluators. The proposed iBAWDO algorithm was validated on 18 multi-scale benchmark datasets against binary versions of 11 well-known meta-heuristic approaches: Binary version of the original WDO (BWDO), Binary Particle Swarm Optimization (BPSO), Binary Bat Algorithm (BBA), Binary Grey Wolf Optimization (BGWO), Binary Whale Optimization Algorithm (BWOA), Binary Grasshopper Optimization Algorithm (BGOA), Binary Sailfish Optimizer (BSFO), Binary Harris Hawks optimization (BHHO), Binary Bird Swarm Algorithm (BBSA), Binary Atom Search Optimization (BASO), and Binary Henry Gas Solubility Optimization (BHGSO). A Wilcoxon’s rank-sum non-parametric test was conducted at a 5% significance level to statistically affirm the high competitiveness of the proposed method. Overall, the experimental results revealed that the proposed method is significantly effective on both small- and large-dimensional datasets.

Text Data Processing and Classification Algorithm Based on Data Fusion and Granular Computing

With the rapid expansion of network information and the emergence of a large number of electronic texts, how to organize and manage this massive information has become a major challenge. Automatic text categorization technology is to study how to let the machine classify unknown text through self-learning, thus solving the difficulties encountered in manual classification. Because granular computing can reduce the knowledge in solving complex problems, it is more convenient to summarize and acquire knowledge. It has become a hotspot in recent years, and it also provides new ideas for text classification research. The rough set model of granular computing can acquire knowledge by mining decision rules. The decision process is more transparent and easy to understand. It has been paid attention to and applied in text classification research. Based on the research of existing achievements, this paper makes a further study on the application of granular computing in text categorization. After analyzing the existing feature selection methods, the feature distribution is proposed based on the relationship between feature words and categories. By calculating the distribution distance between any two feature words, the feature words with similar distribution distances are aggregated, which effectively reduces the dimension of the feature space and also avoids the individual samples caused by the existing feature selection algorithm. A phenomenon that is discarded due to features. The experimental results show that the clustering method can obtain higher classification accuracy than other feature selection methods when using SVM as the classifier. SVM performs best, and the final text classification accuracy rate can reach 85.46%. According to the correlation principle of the rough set, feature selection is made for each information granularity, the selected feature is used as the condition attribute and the coordination matrix is constructed, and the most similar sample is heuristically searched to obtain the attribute reduction set.

Objective Recognition of Tinnitus Location Using Electroencephalography Connectivity Features.

Purpose: Tinnitus is a common but obscure auditory disease to be studied. This study will determine whether the connectivity features in electroencephalography (EEG) signals can be used as the biomarkers for an efficient and fast diagnosis method for chronic tinnitus.Methods: In this study, the resting-state EEG signals of tinnitus patients with different tinnitus locations were recorded. Four connectivity features [including the Phase-locking value (PLV), Phase lag index (PLI), Pearson correlation coefficient (PCC), and Transfer entropy (TE)] and two time-frequency domain features in the EEG signals were extracted, and four machine learning algorithms, included two support vector machine models (SVM), a multi-layer perception network (MLP) and a convolutional neural network (CNN), were used based on the selected features to classify different possible tinnitus sources.Results: Classification accuracy was highest when the SVM algorithm or the MLP algorithm was applied to the PCC feature sets, achieving final average classification accuracies of 99.42 or 99.1%, respectively. And based on the PLV feature, the classification result was also particularly good. And MLP ran the fastest, with an average computing time of only 4.2 s, which was more suitable than other methods when a real-time diagnosis was required.Conclusion: Connectivity features of the resting-state EEG signals could characterize the differentiation of tinnitus location. The connectivity features (PCC and PLV) were more suitable as the biomarkers for the objective diagnosing of tinnitus. And the results were helpful for clinicians in the initial diagnosis of tinnitus.

W2wnet: A Two-Module Probabilistic Convolutional Neural Network with Embedded Data Cleansing Functionality

Convolutional Neural Networks (CNNs) are supposed to be fed with only high-quality annotated datasets. Nonetheless, in many real-world scenarios, such high quality is very hard to obtain, and datasets may be affected by any sort of image degradation and mislabelling issues. This negatively impacts the performance of standard CNNs, both during the training and the inference phase. To address this issue we propose Wise2WipedNet (W2WNet), a new two-module Convolutional Neural Network, where a Wise module exploits Bayesian inference to identify and discard spurious images during the training, and a Wiped module takes care of the final classification while broadcasting information on the prediction confidence at inference time. The goodness of our solution is demonstrated on a number of public benchmarks addressing different image classification tasks, as well as on a real-world case study on histological image analysis. Overall, our experiments demonstrate that W2WNet is able to identify image degradation and mislabelling issues both at training and at inference time, with a positive impact on the final classification accuracy.

Brain Image Classification Using Time Frequency Extraction with Histogram Intensity Similarity

Brain medical image classification is an essential procedure in Computer-Aided Diagnosis (CAD) systems. Conventional methods depend specifically on the local or global features. Several fusion methods have also been developed, most of which are problem-distinct and have shown to be highly favorable in medical images. However, intensity-specific images are not extracted. The recent deep learning methods ensure an efficient means to design an end-to-end model that produces final classification accuracy with brain medical images, compromising normalization. To solve these classification problems, in this paper, Histogram and Time-frequency Differential Deep (HTF-DD) method for medical image classification using Brain Magnetic Resonance Image (MRI) is presented. The construction of the proposed method involves the following steps. First, a deep Convolutional Neural Network (CNN) is trained as a pooled feature mapping in a supervised manner and the result that it obtains are standardized intensified pre-processed features for extraction. Second, a set of time-frequency features are extracted based on time signal and frequency signal of medical images to obtain time-frequency maps. Finally, an efficient model that is based on Differential Deep Learning is designed for obtaining different classes. The proposed model is evaluated using National Biomedical Imaging Archive (NBIA) images and validation of computational time, computational overhead and classification accuracy for varied Brain MRI has been done.

A Deep Learning Ensemble Approach for Automated COVID-19 Detection from Chest CT Images.

Background: The aim of this study was to evaluate the performance of an automated COVID-19 detection method based on a transfer learning technique that makes use of chest computed tomography (CT) images. Method: In this study, we used a publicly available multiclass CT scan dataset containing 4171 CT scans of 210 different patients. In particular, we extracted features from the CT images using a set of convolutional neural networks (CNNs) that had been pretrained on the ImageNet dataset as feature extractors, and we then selected a subset of these features using the Information Gain filter. The resulting feature vectors were then used to train a set of k Nearest Neighbors classifiers with 10-fold cross validation to assess the classification performance of the features that had been extracted by each CNN. Finally, a majority voting approach was used to classify each image into two different classes: COVID-19 and NO COVID-19. Results: A total of 414 images of the test set (10% of the complete dataset) were correctly classified, and only 4 were misclassified, yielding a final classification accuracy of 99.04%. Conclusions: The high performance that was achieved by the method could make it feasible option that could be used to assist radiologists in COVID-19 diagnosis through the use of CT images.

Segment-based bag of visual words model for urban land cover mapping using polarimetric SAR data

Producing high-resolution urban land cover maps is essential for decision-making and urban management. In this regard, Synthetic Aperture Radar (SAR), especially Polarimetric SAR (PolSAR), has served as a valuable data source to fulfill this task. Direct conversion of low-level features into high-level Land Cover (LC) concept may reduce the final classification accuracy. Therefore, mid-level representation models, such as Bag of Visual Words (BOVW), were employed to resolve the existing semantic gap challenge. In this paper, a Segment-based BOVW (Seg-BOVW) model was developed for urban land cover classification using PolSAR data. To this end, two PolSAR data over San Francisco Bay (SFB) and Flevoland (FL) acquired by RADARSAT-2 were employed to comprehensively evaluate the Seg-BOVW model's performance. First, to exploit the full potential of PolSAR data, 169 low-level features in four categories: (1) original, (2) polarimetric, (3) texture, and (4) decomposition features were extracted. Afterward, a Multi-Objective Genetic Algorithm (MOGA) was implemented to investigate the importance of low-level features for urban land cover mapping. This step resulted in selecting 14 and 6 low-level features, as the high contributing features, for SFB and FL datasets, respectively. The Seg-BOVW model achieved significant overall accuracies of 96.02% and 98.82% for SFB and FL, respectively, indicating the high potential of the proposed method for urban land cover classification. Furthermore, a comparison with other well-known algorithms of Support Vector Machine (SVM), Artificial Neural Networks (ANN), and Convolutional Neural Networks (CNN) was made, suggesting the capability of the Seg-BOVW model to improve the urban land cover classification results. Finally, the Seg-BOVW model was tested with two other PolSAR datasets acquired with different sensors over SFB to examine its applicability with different datasets.

Classification modeling method for hyperspectral stamp-pad ink data based on one-dimensional convolutional neural network.

In the questioned document, the examination of stamp-pad ink is crucial scientific evidence to discern the difference between genuine and forged documents. In this study, a new method for rapid and non-destructive identification of types of stamp-pad inks by combining hyperspectral imaging (HSI) technology and deep learning was developed. Twenty stamp-pad inks of different brands and models were collected and numbered in turn, and then, each of them was sealed six times repeatedly on the A4 printing paper for the test. After that, the hyperspectral imager was used to collect the hyperspectral images and the reflectance spectral data were obtained after pixel fusion. Principal component analysis (PCA) and non-negative matrix factorization (NMF) were used to deal with the dataset, but visual results were not good. Then, back propagation neural network (BPNN) and one-dimensional convolutional neural network (1D-CNN) were constructed and their merits and drawbacks were compared. The final loss function of the BPNN of training set and validation set was stable at 0.27 and 0.42, and the classification accuracy of the training set and validation set reached 90.02% and 83.99%, respectively. Compared with the BPNN, the 1D-CNN had better stability and efficiency for the classification. The loss function of the training set and validation set was as low as 0.068 and 0.075, and the final classification accuracy reached 98.30% and 97.94%, respectively. Therefore, the combination of hyperspectral imaging technology and 1D-CNN represents a potentially simple, non-destructive, and rapid method for stamp-pad inks detection and classification.

Multi-scale guided feature extraction and classification algorithm for hyperspectral images

To solve the problem that the traditional hyperspectral image classification method cannot effectively distinguish the boundary of objects with a single scale feature, which leads to low classification accuracy, this paper introduces the idea of guided filtering into hyperspectral image classification, and then proposes a multi-scale guided feature extraction and classification (MGFEC) algorithm for hyperspectral images. Firstly, the principal component analysis theory is used to reduce the dimension of hyperspectral image data. Then, guided filtering algorithm is used to achieve multi-scale spatial structure extraction of hyperspectral image by setting different sizes of filtering windows, so as to retain more edge details. Finally, the extracted multi-scale features are input into the support vector machine classifier for classification. Several practical hyperspectral image datasets were used to verify the experiment, and compared with other spectral feature extraction algorithms. The experimental results show that the multi-scale features extracted by the MGFEC algorithm proposed in this paper are more accurate than those extracted by only using spectral information, which leads to the improvement of the final classification accuracy. This fully shows that the proposed method is not only effective, but also suitable for processing different hyperspectral image data.

Classification of Channel Coding Structure Based on 1-D Inception Architecture

The channel coding structure classification based on the traditional method usually requires the coding type and prior knowledge to be known, which is completely invalid for the unknown signals that are usually intercepted in the current communication environment. In order to solve this problem, this paper proposes a method of blind classification of channel coding structure based on one-dimensional Inception. First, we use MATLAB to generate signal datasets with different channel coding structures under different signal-to-noise ratios (SNR), and then construct a Convolutional Neural Network (CNN) model based on the one-dimensional Inception, and use the datasets to train the network model. The final classification accuracy rate is 96%. This experimental result demonstrates the effectiveness of the proposed method.

A PLA2R-IgG4 Antibody-Based Predictive Model for Assessing Risk Stratification of Idiopathic Membranous Nephropathy.

Background Known as an autoimmune glomerular disease, idiopathic membranous nephropathy (IMN) is considered to be associated with phospholipase A2 receptor (PLA2R) in terms of the main pathogenesis. The quantitative detection of serum PLA2R-IgG and PLA2R-IgG4 antibodies by time-resolved fluoroimmunoassay (TRFIA) was determined, and the value of them, both in the clinical prediction of risk stratification in IMN, was observed in this study. Methods 95 patients with IMN proved by renal biopsy were enrolled, who had tested positive for serum PLA2R antibodies by ELISA, and the quantitative detection of serum PLA2R-IgG and PLA2R-IgG4 antibodies was achieved by TRFIA. All the patients were divided into low-, medium-, and high-risk groups, respectively, which were set as dependent variables, according to proteinuria and renal function. Random forest (RF) was used to estimate the value of serum PLA2R-IgG and PLA2R-IgG4 in predicting the risk stratification of progression in IMN. Results Out-of-bag estimates of variable importance in RF were employed to evaluate the impact of each input variable on the final classification accuracy. The variable of albumin, PLA2R-IgG, and PLA2R-IgG4 had high values (>0.3) of 0.3156, 0.3981, and 0.7682, respectively, which meant that these three were more important for the risk stratification of progression in IMN. In order to further assess the contribution of PLA2R-IgG and PLA2R-IgG4 to the model, we built four different models and found that PLA2R-IgG4 played an important role in improving the predictive ability of the model. Conclusions In this study, we established a random forest model to evaluate the value of serum PLA2R-IgG4 antibodies in predicting risk stratification of IMN. Compared with PLA2R-IgG, PLA2R-IgG4 is a more efficient biomarker in predicting the risk of progression in IMN.

A generic intelligent tomato classification system for practical applications using DenseNet-201 with transfer learning

A generic intelligent tomato classification system based on DenseNet-201 with transfer learning was proposed and the augmented training sets obtained by data augmentation methods were employed to train the model. The trained model achieved high classification accuracy on the images of different quality, even those containing high levels of noise. Also, the trained model could accurately and efficiently identify and classify a single tomato image with only 29 ms, indicating that the proposed model has great potential value in real-world applications. The feature visualization of the trained models shows their understanding of tomato images, i.e., the learned common and high-level features. The strongest activations of the trained models show that the correct or incorrect target recognition areas by a model during the classification process will affect its final classification accuracy. Based on this, the results obtained in this study could provide guidance and new ideas to improve the development of intelligent agriculture.

Measurement and prediction of work engagement under different indoor lighting conditions using physiological sensing

Employee productivity is of paramount importance to most organizations. Studies have shown that a suitable indoor lighting condition is key to help employees remain productive and comfortable in their office spaces. However, it is very difficult to monitor and quantify productivity, which limits our ability to select indoor conditions that maximize the performance of the building occupants. Instead, work engagement is a measurable parameter that is directly related to productivity. Therefore, this paper proposes a method to investigate the effect of lighting level on occupants’ work engagement by studying the frontal asymmetry index (FAI) measured by electroencephalography (EEG). Statistical analysis is performed to investigate the work engagement of the occupants under three typical lighting levels (i.e., 200 lux, 500 lux, and 1000 lux) while they are performing cognitive tasks. The results show that the effect of lighting level on work engagement varies across individuals, which highlights the necessity of developing personalized models. Therefore, this study also proposes a method to predict engagement level for different individuals based on the lighting level and their galvanic skin response (GSR), heart rate (HR), and skin temperature (ST) using the Random Forest (RF) and Artificial Neural Network (ANN). The results show that RF outperforms ANN in most of the prediction cases, and the final classification accuracies are 83.3% for the 3-scale case and 62.2% for the 5-scale case. This opens the possibility of using easily measurable physiological parameters to estimate human brain activities and predict their work engagement under different lighting scenarios.