Search

Objectives: Aggressive driving behavior can lead to potential traffic collision risks, and abnormal weather conditions can exacerbate this behavior. This study aims to develop recognition models for aggressive driving under various climate conditions, addressing the challenge of collecting sufficient data in abnormal weather. Methods: Driving data was collected in a virtual environment using a driving simulator under both normal and abnormal weather conditions. A model was trained on data from normal weather (source domain) and then transferred to foggy and rainy weather conditions (target domains) for retraining and fine-tuning. The K-means algorithm clustered driving behavior instances into three styles: aggressive, normal, and cautious. These clusters were used as labels for each instance in training a CNN model. The pre-trained CNN model was then transferred and fine-tuned for abnormal weather conditions. Results: The transferred models showed improved recognition performance, achieving an accuracy score of 0.81 in both foggy and rainy weather conditions. This surpassed the non-transferred models' accuracy scores of 0.72 and 0.69, respectively. Conclusions: The study demonstrates the significant application value of transfer learning in recognizing aggressive driving behaviors with limited data. It also highlights the feasibility of using this approach to address the challenges of driving behavior recognition under abnormal weather conditions. [ABSTRACT FROM AUTHOR]

Summary: Palpebral conjunctival hue alteration is used in non‐invasive screening for anaemia, whereas it is a qualitative measure. This study constructed machine/deep learning models for predicting haemoglobin values using 150 palpebral conjunctival images taken by a smartphone. The median haemoglobin value was 13.1 g/dL, including 10 patients with <11 g/dL. A segmentation model using U‐net was successfully constructed. The segmented images were subjected to non‐convolutional neural network (CNN)‐based and CNN‐based regression models for predicting haemoglobin values. The correlation coefficients between the actual and predicted haemoglobin values were 0.38 and 0.44 in the non‐CNN‐based and CNN‐based models, respectively. The sensitivity and specificity for anaemia detection were 13% and 98% for the non‐CNN‐based model and 20% and 99% for the CNN‐based model. The performance of the CNN‐based model did not improve with a mask layer guiding the model's attention towards the conjunctival regions, however, slightly improved with correction by the aspect ratio and exposure time of input images. The gradient‐weighted class activation mapping heatmap indicated that the lower half area of the conjunctiva was crucial for haemoglobin value prediction. In conclusion, the CNN‐based model had better results than the non‐CNN‐based model. The prediction accuracy would improve by using more input data with anaemia. [ABSTRACT FROM AUTHOR]

The human brain comprises a complex interconnection of nerve cells and vital organs, which regulates crucial bodily processes. Although neurons commonly undergo developmental stages, they may occasionally experience abnormalities, leading to abnormal growths known as brain tumors. The objective of brain tumor segmentation is to produce precise boundaries of brain tumor regions. This study extensively analyzes deep learning methods for brain tumor detection, evaluating their effectiveness across diverse datasets. It introduces a hybrid model, which is proposed by the name HybriCSF: hybrid convolutional-SVM-fuzzy C-means model combining convolutional neural network (CNN) with the classifier support vector machine (SVM) and clustering technique fuzzy C-means (FCM). The proposed model was implemented on Br35H, BraTs 2020 and BraTs2021 datasets. The suggested model outperformed the existing methods by achieving 98.6% of accuracy on Br35H dataset and dice score of 0.63, 0.87, 0.81 on BraTs 2020 dataset for enhancing tumor (ET), whole tumor (WT), and tumor core (TC), respectively. The achieved dice scores on the BraTs 2021datasets are 0.89, 0.95, and 0.89 for ET, WT, and TC, respectively. The results show that the suggested model HybriCSF outperforms the other CNN-based models in terms of accuracy. [ABSTRACT FROM AUTHOR]

Brain tumors are the result of irregular development of cells. It is a major cause of adult demise worldwide. Several deaths can be avoided with early brain tumor detection. Magnetic resonance imaging (MRI) for earlier brain tumor diagnosis may improve the chance of survival for patients. The most common method of diagnosing brain tumors is MRI. The improved visibility of malignancies in MRI makes therapy easier. The diagnosis and treatment of brain cancers depend on their identification and treatment. Numerous deep learning models are proposed over the last decade including Alexnet, VGG, Inception, ResNet, DenseNet, etc. All these models are trained on a huge dataset, ImageNet. These general models have many parameters, which become irrelevant when implementing these models for a specific problem. This study uses a custom deep-learning model for the classification of brain MRIs. The proposed Disease and Spatial Attention Model (DaSAM) has two modules; (a) the Disease Attention Module (DAM), to distinguish between disease and non-disease regions of an image, and (b) the Spatial Attention Module (SAM), to extract important features. The experiments of the proposed model are conducted on two multi-class datasets that are publicly available, the Figshare and Kaggle datasets, where it achieves precision values of 99% and 96%, respectively. The proposed model is also tested using cross-dataset validation, where it achieved 85% accuracy when trained on the Figshare dataset and validated on the Kaggle dataset. The incorporation of DAM and SAM modules enabled the functionality of feature mapping, which proved to be useful for the highlighting of important features during the decision-making process of the model. [ABSTRACT FROM AUTHOR]

Aims: The aim of this study was to create artificial intelligence (AI) software with the purpose of providing a second opinion to physicians to support distal radius fracture (DRF) detection, and to compare the accuracy of fracture detection of physicians with and without software support. Methods: The dataset consisted of 26,121 anonymized anterior-posterior (AP) and lateral standard view radiographs of the wrist, with and without DRF. The convolutional neural network (CNN) model was trained to detect the presence of a DRF by comparing the radiographs containing a fracture to the inconspicuous ones. A total of 11 physicians (six surgeons in training and five hand surgeons) assessed 200 pairs of randomly selected digital radiographs of the wrist (AP and lateral) for the presence of a DRF. The same images were first evaluated without, and then with, the support of the CNN model, and the diagnostic accuracy of the two methods was compared. Results: At the time of the study, the CNN model showed an area under the receiver operating curve of 0.97. AI assistance improved the physician’s sensitivity (correct fracture detection) from 80% to 87%, and the specificity (correct fracture exclusion) from 91% to 95%. The overall error rate (combined false positive and false negative) was reduced from 14% without AI to 9% with AI. Conclusion: The use of a CNN model as a second opinion can improve the diagnostic accuracy of DRF detection in the study setting. Cite this article: Bone Joint Res 2024;13(10):588–595.

Cataract, an age-related eye disease, poses a significant ophthalmological public health challenge in both developed and developing nations. Tailoring treatment or surgery plans helps accurately categorise the cataract's developmental stage. Precise cataract grading helps in diagnosing cataracts and subsequently scheduling surgical intervention. In this project endeavour, a solution is presented to automate the cataract grading process utilizing slit lamp bio-microscope data sets acquired through smartphones. This innovation is particularly valuable for novice practitioners and non-specialist doctors/experts who may struggle with proficiently interpreting cataract progression, leading to potential misdiagnoses. To address this challenge, a Neural Network model is harnessed to automatically predict the grade of cataracts. The study employs multi-class image classification models, including the Convolutional neural network (CNN) model, the Efficient Net B0 model, and the ResNet50 model, for this purpose. Notably, the ResNet50 model outperforms the other models in terms of accuracy and prediction capability for the provided data set. Achieving an accuracy rate of 0.8611, the ResNet50 model demonstrates superior performance in classifying cataract grades, after augmenting the data set with 544 images. This performance comparison establishes the ResNet50 model as the most robust choice among the considered models and data sets.

In this study, a novel method combining contour analysis with deep CNN is applied for fire detection. The method was made for fire detection using two main algorithms: one which detects the color properties of the fires, and another which analyzes the shape through contour detection. To overcome the disadvantages of previous methods, we generate a new labeled dataset, which consists of small fire instances and complex scenarios. We elaborated the dataset by selecting regions of interest (ROI) for enhanced fictional small fires and complex environment traits extracted through color characteristics and contour analysis, to better train our model regarding those more intricate features. Results of the experiment showed that our improved CNN model outperformed other networks. The accuracy, precision, recall and F1 score were 99.4%, 99.3%, 99.4% and 99.5%, respectively. The performance of our new approach is enhanced in all metrics compared to the previous CNN model with an accuracy of 99.4%. In addition, our approach beats many other state-of-the-art methods as well: Dilated CNNs (98.1% accuracy), Faster R-CNN (97.8% accuracy) and ResNet (94.3%). This result suggests that the approach can be beneficial for a variety of safety and security applications ranging from home, business to industrial and outdoor settings. [ABSTRACT FROM AUTHOR]

With the rapid development of artificial intelligence technology, image recognition has become a key technology in many fields. This paperaims to analyze the performance of image recognition tasks through the implementation of deep learning algorithms using PyTorch. This paperintroduces the selection and preprocessing techniques of datasets, including normalization, data augmentation, size adjustment, and batch processing, which provide high-quality input data for model training. In terms of deep learning model design, Convolutional Neural Network was adopted, CNN was used as the infrastructure and the cross entropy loss function and Adam optimizer were selected for model training. The implementation details of PyTorch demonstrate the model construction, training process, and debugging, as well as how to use PyTorch's DataLoader class to improve data loading efficiency. The Results and Discussion section presents the training results of the model, where the CNN model achieves the highest accuracy of 99% in image recognition tasks, and the loss function value ultimately stabilizes at 0.02, indicating that the model has efficient learning and generalization capabilities. Performance comparison experiments show that, the CNN implemented by PyTorch outperforms MXNet and Caffe in terms of frame rate and processing delay, with a frame rate between 51FPS-79FPS and a minimum processing delay of 308ms. The analysis of practical application scenarios further explores the potential application of the model in fields such as medical image analysis, autonomous driving, retail industry, and safety monitoring, and points out challenges such as data privacy, model generalization ability, and computational resource limitations. [ABSTRACT FROM AUTHOR]

Many industries, including healthcare, the military, finance, and more, need extra protection for the interchange of picture data since images are now sent across open channels that might be attacked. In order to protect the system against differential and brute force assaults, the security aspects are crucial. The transmission of multimedia, including digital pictures, text, audio, and video, relies heavily on encryption to maintain secrecy, integrity, and confidentiality while preventing unwanted access to critical information. Even while chaos-based cryptosystems aren't as widely used as AES, DES, or RSA, they've been a hot topic of study recently and can enhance the security of public key cryptosystems when combined with them. With the rise of deep convolutional neural networks (CNNs) as the go-to machine learning technology for many uses, there have been several efforts to use CNNs to decipher encrypted data. On the other hand, prior research has paid little attention to protecting model parameters and has instead concentrated on protecting data. Additionally, they provide high-level implementations without thoroughly analyzing the trade-offs between speed, security, and accuracy in the ECC implementation of common CNN basic operators like non-linear activation, convolution, along with pooling. The goal of this research is to develop and construct a cryptosystem based on Chaos that can effectively encrypt images and withstand differential assaults. In order to create the first layer of encryption, the system first divides the original picture into smaller pieces and rearranges them. A logistic map is used to generate a one-dimensional sequence, which is then multiplied over the highest pixel value and processed bit by bit as part of the encryption process. We use the outcome to encrypt the picture, and then apply the same procedure to decode it. For efficient key generation during picture encryption, the suggested model uses the ECC approach to produce a Dung Beetle optimization (DBO). For improved security performance, the chaotic map notion is introduced to the robust optimization approach. The results of the investigation demonstrate that the suggested approach provides significantly improved security performance while leaving picture quality unaffected. The histogram, Pearson's correlation analysis, peak signal-to-noise ratio (PSNR), entropy, number of pixels change rate (NPCR), and unified average fluctuation in intensity (UACI) are used to assess the encryption outcomes. Our findings prove that the suggested strategy is safe, dependable, efficient, and adaptable. [ABSTRACT FROM AUTHOR]

Although atmospheric CO2 concentrations collected by satellites play a crucial role in understanding global greenhouse gases, the sparse geographic distribution greatly affects their widespread application. In this paper, a hybrid CNN and spatiotemporal Kriging (CNN-STK) model is proposed to generate a monthly spatiotemporal continuous XCO2 dataset over China at 0.25° grid-scale from 2015 to 2020, utilizing OCO-2 XCO2 and geographic covariates. The validations against observation samples, CAMS XCO2 and TCCON measurements indicate the CNN-STK model is effective, robust, and reliable with high accuracy (validation set metrics: R2 = 0.936, RMSE = 1.3 ppm, MAE = 0.946 ppm; compared with TCCON: R2 = 0.954, RMSE = 0.898 ppm and MAE = 0.741 ppm). The accuracy of CNN-STK XCO2 exhibits spatial inhomogeneity, with higher accuracy in northern China during spring, autumn, and winter and lower accuracy in northeast China during summer. XCO2 in low-value-clustering areas is notably influenced by biological activities. Moreover, relatively high uncertainties are observed in the Qinghai-Tibet Plateau and Sichuan Basin. This study innovatively integrates deep learning with the geostatistical method, providing a stable and cost-effective approach for other countries and regions to obtain regional scales of atmospheric CO2 concentrations, thereby supporting policy formulation and actions to address climate change. [ABSTRACT FROM AUTHOR]

This article mainly studies the method of deep recognition analysis of power technology standard images using Convolutional Neural Network (CNN). In this study, an efficient CNN model was constructed to quickly and accurately classify and recognize common technical standard images in the power industry. The experimental results indicate that the research method has high accuracy and good practicality in the recognition of power technology standard images. In the experimental stage, this study designed four experiments to explore the application of CNN models in power technology standard image recognition, with a focus on evaluating the impact of different network designs and data processing strategies on model performance. In the benchmark model comparison experiment, the AUC (Area Under the ROC Curve) value of the CNN model in the power technology standard image recognition task was 0.96, the AUC value of the Support Vector Machine (SVM) was 0.88, and the AUC value of the decision tree model was 0.82. In experiments on the impact of different data augmentation strategies, the CNN model using scaling augmentation strategy achieved an accuracy of 90%. In the experiment on the impact of network depth on performance, the accuracy of the CNN model was highest improved to 95%. In real-time recognition capability testing, although the CNN model has the highest accuracy, the average processing time for each image is also relatively long. In the above data conclusion, the effectiveness of the CNN model in power technology standard image recognition tasks has been demonstrated, and possible paths for optimizing model performance in practical applications have also been revealed. [ABSTRACT FROM AUTHOR]

Cytological evaluation through microscopic image analysis of fine needle aspiration cytology (FNAC) is pivotal in the initial screening of breast cancer. The sensitivity of FNAC as a screening tool relies on both image quality and the pathologist’s expertise. To enhance diagnostic accuracy and alleviate the pathologist’s workload, a computer-aided diagnosis (CAD) system was developed. A comparative study was conducted, assessing twelve candidate pre-trained models. Utilizing a locally gathered FNAC image dataset, three superior models-MobileNet-V2, DenseNet-121, and Inception-V3-were selected based on their training, validation, and testing accuracies. Further, these models underwent evaluation in four transfer learning scenarios to enhance testing accuracy. While the outcomes were promising, they left room for improvement, motivating us to create a novel deep convolutional neural network (CNN). The newly proposed model exhibited robust performance with testing accuracy at 85%. Our research concludes that the most lightweight, high-accuracy model is the one we propose. We’ve integrated it into our user-friendly Android App, “Breast Cancer Detection System,” in TensorFlow Lite format, with cloud database support, showcasing its effectiveness. Implementing an artificial intelligent (AI)-based diagnosis system with a user-friendly interface holds the potential to enhance early breast cancer detection using FNAC. [ABSTRACT FROM AUTHOR]

For manufacturing components with thick plates, such as in the heavy equipment and shipbuilding industries, the gas metal arc welding (GMAW) process is applied. Among the components that apply the thick plate GMAW process, there are groove butt joints, which are fabricated through multi-pass welding. Various welding qualities are managed in multi-pass welding, and the root-pass weld is controlled to ensure complete joint penetration (CJP). Currently, the state of complete joint penetration during root-pass welding is managed visually, making it difficult to confirm the penetration condition in real time. Therefore, there is a need to predict the penetration condition in real time. In this study, we propose a convolutional neural network (CNN)-based prediction model that can classify penetration conditions using welding current and voltage data from the root pass of V-groove butt joints. The root gap of the joints was varied between 1.0 and 2.0 mm, and the wire feed rate was adjusted. During welding, the current and voltage were measured. The welding current and voltage are transformed into a short-time Fourier transform (STFT) representation depicting the arc and wire extension lengths. The transformed dynamic resistance STFT information serves as the input variable for the CNN model. Preprocessing steps, including thresholding, are applied to optimize the input variables. The CNN architecture comprises three convolutional layers and two pooling layers. The model classifies penetration conditions as partial joint penetration (PJP), CJP, and burn-through, achieving a high accuracy of 97.8%. The proposed method facilitates the non-destructive evaluation of the root-pass welding quality without expensive monitoring equipment, such as vision cameras. It is expected to be immediately applied to the thick plate welding process using readily available welding data. [ABSTRACT FROM AUTHOR]

Vision-based human gesture detection is the task of forecasting a gesture, namely clapping or sign language gestures, or waving hello, utilizing various video frames. One of the attractive features of gesture detection is that it makes it possible for humans to interact with devices and computers without the necessity for an external input tool like a remote control or a mouse. Gesture detection from videos has various applications, like robot learning, control of consumer electronics computer games, and mechanical systems. This study leverages the Lion Swarm optimizer with a deep convolutional neural network (LSO-DCNN) for gesture recognition and classification. The purpose of the LSO-DCNN technique lies in the proper identification and categorization of various categories of gestures that exist in the input images. The presented LSO-DCNN model follows a three-step procedure. At the initial step, the 1D-convolutional neural network (1D-CNN) method derives a collection of feature vectors. In the second step, the LSO algorithm optimally chooses the hyperparameter values of the 1D-CNN model. At the final step, the extreme gradient boosting (XGBoost) classifier allocates proper classes, i.e., it recognizes the gestures efficaciously. To demonstrate the enhanced gesture classification results of the LSO-DCNN approach, a wide range of experimental results are investigated. The brief comparative study reported the improvements in the LSO-DCNN technique in the gesture recognition process.

[Display omitted] • Imaging Flow Cytometry based understanding of cellular heterogeneity. • Application of a CNN based image classifier to differentiate between hMSCs. • Development of machine-learning aided quality assurance platform for hMSCs. The multipotent stem cells of our body have been largely harnessed in biotherapeutics. However, as they are derived from multiple anatomical sources, from different tissues, human mesenchymal stem cells (hMSCs) are a heterogeneous population showing ambiguity in their in vitro behavior. Intra-clonal population heterogeneity has also been identified and pre-clinical mechanistic studies suggest that these cumulatively depreciate the therapeutic effects of hMSC transplantation. Although various biomarkers identify these specific stem cell populations, recent artificial intelligence-based methods have capitalized on the cellular morphologies of hMSCs, opening a new approach to understand their attributes. A robust and rapid platform is required to accommodate and eliminate the heterogeneity observed in the cell population, to standardize the quality of hMSC therapeutics globally. Here, we report our primary findings of morphological heterogeneity observed within and across two sources of hMSCs namely, stem cells from human exfoliated deciduous teeth (SHEDs) and human Wharton jelly mesenchymal stem cells (hWJ MSCs), using real-time single-cell images generated on immunophenotyping by imaging flow cytometry (IFC). We used the ImageJ software for identification and comparison between the two types of hMSCs using statistically significant morphometric descriptors that are biologically relevant. To expand on these insights, we have further applied deep learning methods and successfully report the development of a Convolutional Neural Network-based image classifier. In our research, we introduced a machine learning methodology to streamline the entire procedure, utilizing convolutional neural networks and transfer learning for binary classification, achieving an accuracy rate of 97.54%. We have also critically discussed the challenges, comparisons between solutions and future directions of machine learning in hMSC classification in biotherapeutics. [ABSTRACT FROM AUTHOR]

Presently issues related to the heart are mostly considered as the deadliest diseases resulting in the increasing death rate across the globe irrespective of gender and age. Accurate and early prediction of these heart abnormalities require enormous experience along with advanced technology. Artificial Intelligence and Machine learning have contributed extensively as an emerging technology for the prediction of the diseases including heart abnormalities. However, still high accuracy of prediction and less computational complexities remains challenge for the researchers. Currently the prevailing scenario existing post pandemic and new variants coming in, it is very essential to come up with an amalgamation of technology and techniques as assessment for doctors in diagnosis. Proposed article makes use of Convolutional Neural networks (CNN) for detecting and classifying the heart abnormality into three labels as mild, moderate or severe which helps in appropriate treatment of the heart diseases. CNN model is built. The model is trained with datasets corresponding to mild, moderate and severe conditions of heart abnormality. The model is tested extensively for the test data and the performance indices are obtained are with reference to accuracy, precision, recall, specificity, F1-Score. Proposed work is implemented in the Keras libraries with Tensorflow 2.1.1 as backend. Furthermore, prediction performance and complexity overhead are correlated with the other existing cutting-edge algorithms in predicting the heart abnormalities. Results demonstrates that the proposed model proved to be efficient than the others with high prediction accuracy (99%) and less computational complexities. [ABSTRACT FROM AUTHOR]

This paper provides a method for improving the photovoltaic conversion efficiency and optical attributes of silicon solar cells manufactured from as‐cut boron doped p‐type multi‐crystalline silicon wafers using acid‐based chemical texturization via machine learning. A decreased reflectance, which can be attained by the right chemical etching conditions, is one of the key elements for raising solar cell efficiency. In this work, the mc‐Silicon wafer surface reflectance is obtained under (<2%) after optimization of wet chemical etching. The HF + HNO3 + CH3COOH chemical etchant is used in the ratio 1:3:2 at different conditions of the etching duration of 1 min, 2 min, 3 min, and 4 min, respectively. The as‐cut boron doped p‐type mc‐silicon wafers are analysed with ultraviolet–visible spectroscopy, optical microscopy, Fourier transforms infrared spectroscopy, thickness profilometer, and scanning electron microscopy before and after etching. The chemical etching solution produces good results in 3 min etched wafer, with a reflectivity value of <2%.The reflectivity and optical images are inputs to the convolutional neural network model and the linear regression model to obtain the etching rate for better reflectivity. The classification model provides 99.6% accuracy and the regression model results in the minimum mean squared error (MSE) of 0.062. [ABSTRACT FROM AUTHOR]

Cataract, an age-related eye disease, poses a significant ophthalmological public health challenge in both developed and developing nations. Tailoring treatment or surgery plans helps accurately categorise the cataract's developmental stage. Precise cataract grading helps in diagnosing cataracts and subsequently scheduling surgical intervention. In this project endeavour, a solution is presented to automate the cataract grading process utilizing slit lamp bio-microscope data sets acquired through smartphones. This innovation is particularly valuable for novice practitioners and non-specialist doctors/experts who may struggle with proficiently interpreting cataract progression, leading to potential misdiagnoses. To address this challenge, a Neural Network model is harnessed to automatically predict the grade of cataracts. The study employs multi-class image classification models, including the Convolutional neural network (CNN) model, the Efficient Net B0 model, and the ResNet50 model, for this purpose. Notably, the ResNet50 model outperforms the other models in terms of accuracy and prediction capability for the provided data set. Achieving an accuracy rate of 0.8611, the ResNet50 model demonstrates superior performance in classifying cataract grades, after augmenting the data set with 544 images. This performance comparison establishes the ResNet50 model as the most robust choice among the considered models and data sets. [ABSTRACT FROM AUTHOR]

Vision-based human gesture detection is the task of forecasting a gesture, namely clapping or sign language gestures, or waving hello, utilizing various video frames. One of the attractive features of gesture detection is that it makes it possible for humans to interact with devices and computers without the necessity for an external input tool like a remote control or a mouse. Gesture detection from videos has various applications, like robot learning, control of consumer electronics computer games, and mechanical systems. This study leverages the Lion Swarm optimizer with a deep convolutional neural network (LSO-DCNN) for gesture recognition and classification. The purpose of the LSO-DCNN technique lies in the proper identification and categorization of various categories of gestures that exist in the input images. The presented LSO-DCNN model follows a three-step procedure. At the initial step, the 1D-convolutional neural network (1D-CNN) method derives a collection of feature vectors. In the second step, the LSO algorithm optimally chooses the hyperparameter values of the 1D-CNN model. At the final step, the extreme gradient boosting (XGBoost) classifier allocates proper classes, i.e., it recognizes the gestures efficaciously. To demonstrate the enhanced gesture classification results of the LSO-DCNN approach, a wide range of experimental results are investigated. The brief comparative study reported the improvements in the LSO-DCNN technique in the gesture recognition process. [ABSTRACT FROM AUTHOR]

Software defect identification (SDI) is a key part of improving the quality of software projects and lowering the risks that along with maintenance. It does identify the software defect causes that have not been reached yet to get sufficient results. On the other hand, many researchers have recently developed several models, including NN, ML, DL, advanced CNN, and LSTM, to enhance the effectiveness of defect prediction. Due to an insufficient dataset size, repeated investigations, and no longer appropriate baseline selection, the research on the CNN model was unable to produce reliable results. In addition, XAI a well-known explainability approach creates deep models in computer vision, as well as successfully handles the software defect prediction that is easy for humans to understand. To address these issues, firstly we have used SMOTE for preprocessing which was collected from the NASA repository; categorical and numerical data. Secondly, we have experimented with software defect prediction using 1D-CNN and 2D-CNN named lightweight CNN (LCNN). Subsequently, evaluation we have employed a 100-repetition holdout validation. For the cross-validation setup, we utilized the 1D-CNN model was $20\times 1$ , and for the 2D-CNN model, it was $4\times 5 \times 1$ . After that, the results of the experiment were compared and assessed in terms of accuracy, MSE, and AUC. The result shows that 2D-CNN shows 1.36% better contrast with 1D-CNN. Thirdly, we have conducted research on the identification of software defect features via LIME and SHAP in XAI stand as state-of-the-art techniques. However, we cannot use 2D-CNN because it involves more complex relationships, making it challenging to create transparent explanations. That is why we have realized that 1D-CNN will superior result to explain the root cause of software feature identifications. Finally, LIME provides accurate visualization of software defect features in contrast with SHAP, as well as it helps the stakeholders of the software industry easily find actual root causes of software defect identification.

Online customer feedback is essential for promoting online buying. The e-commerce sector has experienced exponential growth since COVID-19. Now days the growth of any business in e-commerce industry is highly influenced by the online consumer reviews and a lot of research has been conducted by numerous researchers in this regard to determine the reviews helpfulness for experience and search based products. In the study, the author's aims to develop a convolutional neural network based binary classification model for assessing the usefulness of the products reviews through the analysis of consumer evaluations with reference to products and services in e-commerce. In this experiment, predicting helpfulness is considered as binary research problem in order to determine review helpfulness in association with the structural, emotional, linguistic, emotive, lexical, and voting feature sets. In this study, various machine learning algorithms viz., K-nearest neighbor (KNN), Linear regression (LR), Gaussian naive bays (GNB), linear discriminant analysis (LDA), support vector machine (SVM) and convolution neural networks (CNN) were used to build the classification models and their results were compared with each other and other existing state of art models. From the simulation results, it was observed that CNN outperformed over the above stated algorithms and other existing state-of-the-art classification models, achieving 99.72% and 99.97% accuracy for two different search and experience based datasets. Furthermore, the performance of these models were also evaluated in terms of precision, recall, and F1 score. The findings presented in this paper reveals the importance of machine learning models in selecting the quality products. [ABSTRACT FROM AUTHOR]

Structural damages are expected to be detected from the changes in shear‐wave propagation in the vertical direction of the building. In the previous study, we constructed a convolutional neural network (CNN) model to recognize the changes in impulse responses at the inter‐floor with the virtual source at the top. However, it is difficult to tell the exact damaged stories, because it contains the information of all the stories between the virtual source and the receiver. In this study, to perform the story‐by‐story damage detection, wave propagation between two adjacent floors was proposed. The virtual source of the impulse response was changed from the top floor to the determined inter‐floor of the building. By recognizing the changes in the wavefield between the determined floor where the virtual source is located and its upper and lower adjacent floor, damages between two adjacent stories can be identified. The CNN model was used to automatically recognize the changes in the visualized impulse responses over time and validated using the data of a shake‐table test on a one‐third scaled 18‐story steel frame building. [ABSTRACT FROM AUTHOR]

Brain tumors are the result of irregular development of cells. It is a major cause of adult demise worldwide. Several deaths can be avoided with early brain tumor detection. Magnetic resonance imaging (MRI) for earlier brain tumor diagnosis may improve the chance of survival for patients. The most common method of diagnosing brain tumors is MRI. The improved visibility of malignancies in MRI makes therapy easier. The diagnosis and treatment of brain cancers depend on their identification and treatment. Numerous deep learning models are proposed over the last decade including Alexnet, VGG, Inception, ResNet, DenseNet, etc. All these models are trained on a huge dataset, ImageNet. These general models have many parameters, which become irrelevant when implementing these models for a specific problem. This study uses a custom deep-learning model for the classification of brain MRIs. The proposed Disease and Spatial Attention Model (DaSAM) has two modules; (a) the Disease Attention Module (DAM), to distinguish between disease and non-disease regions of an image, and (b) the Spatial Attention Module (SAM), to extract important features. The experiments of the proposed model are conducted on two multi-class datasets that are publicly available, the Figshare and Kaggle datasets, where it achieves precision values of 99% and 96%, respectively. The proposed model is also tested using cross-dataset validation, where it achieved 85% accuracy when trained on the Figshare dataset and validated on the Kaggle dataset. The incorporation of DAM and SAM modules enabled the functionality of feature mapping, which proved to be useful for the highlighting of important features during the decision-making process of the model.

In this study, a novel method combining contour analysis with deep CNN is applied for fire detection. The method was made for fire detection using two main algorithms: one which detects the color properties of the fires, and another which analyzes the shape through contour detection. To overcome the disadvantages of previous methods, we generate a new labeled dataset, which consists of small fire instances and complex scenarios. We elaborated the dataset by selecting regions of interest (ROI) for enhanced fictional small fires and complex environment traits extracted through color characteristics and contour analysis, to better train our model regarding those more intricate features. Results of the experiment showed that our improved CNN model outperformed other networks. The accuracy, precision, recall and F1 score were 99.4%, 99.3%, 99.4% and 99.5%, respectively. The performance of our new approach is enhanced in all metrics compared to the previous CNN model with an accuracy of 99.4%. In addition, our approach beats many other state-of-the-art methods as well: Dilated CNNs (98.1% accuracy), Faster R-CNN (97.8% accuracy) and ResNet (94.3%). This result suggests that the approach can be beneficial for a variety of safety and security applications ranging from home, business to industrial and outdoor settings.

Although atmospheric CO2 concentrations collected by satellites play a crucial role in understanding global greenhouse gases, the sparse geographic distribution greatly affects their widespread application. In this paper, a hybrid CNN and spatiotemporal Kriging (CNN-STK) model is proposed to generate a monthly spatiotemporal continuous XCO2 dataset over China at 0.25° grid-scale from 2015 to 2020, utilizing OCO-2 XCO2 and geographic covariates. The validations against observation samples, CAMS XCO2 and TCCON measurements indicate the CNN-STK model is effective, robust, and reliable with high accuracy (validation set metrics: R2 = 0.936, RMSE = 1.3 ppm, MAE = 0.946 ppm; compared with TCCON: R2 = 0.954, RMSE = 0.898 ppm and MAE = 0.741 ppm). The accuracy of CNN-STK XCO2 exhibits spatial inhomogeneity, with higher accuracy in northern China during spring, autumn, and winter and lower accuracy in northeast China during summer. XCO2 in low-value-clustering areas is notably influenced by biological activities. Moreover, relatively high uncertainties are observed in the Qinghai-Tibet Plateau and Sichuan Basin. This study innovatively integrates deep learning with the geostatistical method, providing a stable and cost-effective approach for other countries and regions to obtain regional scales of atmospheric CO2 concentrations, thereby supporting policy formulation and actions to address climate change.

For manufacturing components with thick plates, such as in the heavy equipment and shipbuilding industries, the gas metal arc welding (GMAW) process is applied. Among the components that apply the thick plate GMAW process, there are groove butt joints, which are fabricated through multi-pass welding. Various welding qualities are managed in multi-pass welding, and the root-pass weld is controlled to ensure complete joint penetration (CJP). Currently, the state of complete joint penetration during root-pass welding is managed visually, making it difficult to confirm the penetration condition in real time. Therefore, there is a need to predict the penetration condition in real time. In this study, we propose a convolutional neural network (CNN)-based prediction model that can classify penetration conditions using welding current and voltage data from the root pass of V-groove butt joints. The root gap of the joints was varied between 1.0 and 2.0 mm, and the wire feed rate was adjusted. During welding, the current and voltage were measured. The welding current and voltage are transformed into a short-time Fourier transform (STFT) representation depicting the arc and wire extension lengths. The transformed dynamic resistance STFT information serves as the input variable for the CNN model. Preprocessing steps, including thresholding, are applied to optimize the input variables. The CNN architecture comprises three convolutional layers and two pooling layers. The model classifies penetration conditions as partial joint penetration (PJP), CJP, and burn-through, achieving a high accuracy of 97.8%. The proposed method facilitates the non-destructive evaluation of the root-pass welding quality without expensive monitoring equipment, such as vision cameras. It is expected to be immediately applied to the thick plate welding process using readily available welding data.

Abstract Background Cardiomyocytes derived from human iPS cells (hiPSCs) include cells showing SAN‐ and non‐SAN‐type spontaneous APs. Objectives To examine whether the deep learning technology could identify hiPSC‐derived SAN‐like cells showing SAN‐type‐APs by their shape. Methods We acquired phase‐contrast images for hiPSC‐derived SHOX2/HCN4 double‐positive SAN‐like and non‐SAN‐like cells and made a VGG16‐based CNN model to classify an input image as SAN‐like or non‐SAN‐like cell, compared to human discriminability. Results All parameter values such as accuracy, recall, specificity, and precision obtained from the trained CNN model were higher than those of human classification. Conclusions Deep learning technology could identify hiPSC‐derived SAN‐like cells with considerable accuracy.

In this study, we identified the different causes of odor problems and their associated discomfort. We also recognized the significance of public health and environmental concerns. To address odor issues, it is vital to conduct precise analysis and comprehend the root causes. We suggested a hybrid model of a Convolutional Neural Network (CNN) and Transformer called the CNN–Transformer to tackle this challenge and assessed its effectiveness. We utilized a dataset containing 120,000 samples of odor to compare the performance of CNN+LSTM, CNN, LSTM, and ELM models. The experimental results show that the CNN+LSTM hybrid model has an accuracy of 89.00%, precision of 89.41%, recall of 91.04%, F1-score of 90.22%, and RMSE of 0.28, with a large prediction error. The CNN+Transformer hybrid model had an accuracy of 96.21%, precision and recall of 94.53% and 94.16%, F1-score of 94.35%, and RMSE of 0.27, showing a low prediction error. The CNN model had an accuracy of 87.19%, precision and recall of 89.41% and 91.04%, F1-score of 90.22%, and RMSE of 0.23, showing a low prediction error. The LSTM model had an accuracy of 95.00%, precision and recall of 92.55% and 94.17%, F1-score of 92.33%, and RMSE of 0.03, indicating a very low prediction error. The ELM model performed poorly with an accuracy of 85.50%, precision and recall of 85.26% and 85.19%, respectively, and F1-score and RMSE of 85.19% and 0.31, respectively. This study confirms the suitability of the CNN–Transformer hybrid model for odor analysis and highlights its excellent predictive performance. The employment of this model is expected to be advantageous in addressing odor problems and mitigating associated public health and environmental concerns. [ABSTRACT FROM AUTHOR]

Many people have been severely affected by the COVID-19 outbreak, which has left them anxious, terrified, and other difficult feelings. Since the introduction of coronavirus vaccinations, people's emotional spectrum has broadened and become more sophisticated.We want to observe and interpret their sentiments using deep learning techniques in this work. The most efficient way to convey one's thoughts and feelings right now is via social media, and using Twitter may help one better understand what is popular and what is going through other people's minds. Analyzing and visualization of data play a vital role in Data Science; as customers over e-commerce increase, feedback/reviews shared by them increase significantly, and decisions by a new customer to buy a product or not rely on these reviews; reviews might falsely be displayed which may be involving in controlling if any products demand and supply so, reviews analyzing and visualizationto understand they are genuinely playing an important role over e-commerce nowadays. Our primary objective in conducting this study was to understand better the various perspectives individuals held on the vaccination process and reviews of products purchased online. As shown by the presented study, analysis and visualization approaches may be used to facilitate rapid and easy comprehension of e-commerce data, despite its high dimensionality.All correlation and non-correlation factors were mapped and examined, providing a comprehensive picture of the proposed data and its connection to other parameters.The proposed work provides an overview of sentiment observations across arguments and the relationships between parameters; it opens the door for modeling to extract some decision-making insights from the data, which can be used to improve the efficiency of application areas like product quality and customer satisfaction. [ABSTRACT FROM AUTHOR]