The VVAD-LRS3 Dataset for Visual Voice Activity Detection

ObjectiveThis study aimed to develop and validate an Informer- Convolutional Neural Network (CNN) model to predict the gamma passing rate (GPR) for patient-specific quality assurance in volumetric modulated arc therapy (VMAT), enhancing treatment safety and efficacy by integrating multiple data sources.MethodsAnalyzing 465 VMAT treatment plans covering head & neck, chest, and abdomen, the study extracted data from 31 complexity indicators, 123 radiomics features, and 123 dosimetrics indices, along with daily linac performance data including 141 key performance indicators. A hybrid Informer-CNN architecture was used to handle both temporal and non-temporal data for predicting GPR.ResultsThe Informer-CNN model demonstrated superior predictive performance over traditional models like Convolutional Neural Networks (CNN), Long Short-Term Memory(LSTM), and Informer. Specifically, in the validation set, the model achieved a mean absolute error (MAE) of 0.0273 and a root mean square error (RMSE) of 0.0360 using the 3%/3mm criterion. In the test set, the MAE was 0.0327 and the RMSE was 0.0468. The model also showed high classification performance with AUC scores of 0.97 and 0.95 in test and validation sets, respectively.ConclusionThe developed Informer-CNN model significantly enhances the prediction accuracy and classification of gamma passing rates in VMAT treatment plans. It facilitates early integration of daily accelerator performance data, improving the assessment and verification of treatment plans for better patient-specific quality assurance.

Aims: This study aimed to explore the effect of training set diversity on the performance of deep learning models for predicting early gastric cancer (EGC) under endoscopy. Methods: Images of EGC and non-cancerous lesions under narrow-band imaging (ME-NBI) and magnifying blue laser imaging (ME-BLI) were retrospectively collected. Training set 1 was composed of 150 non-cancerous and 309 EGC ME-NBI images, training set 2 was composed of 1505 non-cancerous and 309 EGC ME-BLI images, and training set 3 was the combination of training set 1 and 2. Test set 1 was composed of 376 non-cancerous and 1052 EGC ME-NBI images, test set 2 consisted of 529 non-cancerous and 71 EGC ME-BLI images, and test set 3 was the combination of test set 1 and test set 2. Three deep learning models, convolutional neural network (CNN) 1, CNN 2 and CNN 3 (CNN 1, CNN 2 and CNN 3 were independently trained using training set 1, training set 2 and training set 3, respectively), were constructed, and their performances on each test set were respectively evaluated. One hundred and thirty-eight ME-NBI videos and 17 ME-BLI videos were further collected to evaluate and compare the performance of each model in real time. Results: On the whole, the performance of CNN 3 was the best. The accuracy (Acc), sensitivity (Sn), specificity (Sp) and area under the curve (AUC) of test set 1 in CNN 3 were 87.89% (1255/1428), 90.96% (342/376), 86.79% (913/1052) and 94.60%, respectively. The Acc, Sn, Sp and AUC of test set 2 in CNN 3 were 95% (570/600), 97.92% (518/529), 73.24% (52/71) and 90.93% respectively. The Acc, Sn, Sp and AUC of test set 3 in CNN 3 were 89.99% (1825/2028), 95.03% (860/905), 85.93% (965/1123) and 94.89%, respectively. The performance of CNN 3 was also the best in videos test set. The Acc, Sn and Sp of videos test set in CNN 3 were 91.03% (142/156), 90.58% (125/138) and 94.44% (17/18), respectively. Conclusions: The deep learning model with the most diverse training data has the best diagnostic effect.

Deep convolutional neural networks (DCNNs) for diagnosis of disease on chest radiographs (CXR) have been shown to be biased against males or females if the datasets used to train them have unbalanced sex representation. Prior work has suggested that DCNNs can predict sex on CXR, which could aid forensic evaluations, but also be a source of bias. To (1) evaluate the performance of DCNNs for predicting sex across different datasets and architectures and (2) evaluate visual biomarkers used by DCNNs to predict sex on CXRs. Chest radiographs were obtained from the Stanford CheXPert and NIH Chest XRay14 datasets which comprised of 224,316 and 112,120 CXRs, respectively. To control for dataset size and class imbalance, random undersampling was used to reduce each dataset to 97,560 images that were balanced for sex. Each dataset was randomly split into training (70%), validation (10%), and test (20%) sets. Four DCNN architectures pre-trained on ImageNet were used for transfer learning. DCNNs were externally validated using a test set from the opposing dataset. Performance was evaluated using area under the receiver operating characteristic curve (AUC). Class activation mapping (CAM) was used to generate heatmaps visualizing the regions contributing to the DCNN's prediction. On the internal test set, DCNNs achieved AUROCs ranging from 0.98 to 0.99. On external validation, the models reached peak cross-dataset performance of 0.94 for the VGG19-Stanford model and 0.95 for the InceptionV3-NIH model. Heatmaps highlighted similar regions of attention between model architectures and datasets, localizing to the mediastinal and upper rib regions, as well as to the lower chest/diaphragmatic regions. DCNNs trained on two large CXR datasets accurately predicted sex on internal and external test data with similar heatmap localizations across DCNN architectures and datasets. These findings support the notion that DCNNs can leverage imaging biomarkers to predict sex and potentially confound the accurate prediction of disease on CXRs and contribute to biased models. On the other hand, these DCNNs can be beneficial to emergency radiologists for forensic evaluations and identifying patient sex for patients whose identities are unknown, such as in acute trauma.

Text classification is one of the important applications of Natural Language Processing (NLP) in which text is classified into pre-defined categories or classes. Several state-of-the-arts have already been conducted on text classification in different Indian languages, however, limited researches on text classification performed in Odia. Moreover, the news headlines classification is a text classification technique which is performed in our study mainly focusing on Odia language. The Convolutional Neural Network (CNN) based Deep Learning technique is used to classify the Odia news headlines dataset available at Kaggle repository. The dataset consisting of approx. 15,200 Odia news headlines is split into training and testing sets of 80:20 ratio and is categorized into different classes, such as sports, business, and entertainment by using CNN. The efficacy and efficiency of the proposed CNN-based Odia news headlines classification is evaluated in contrary to other existing machine learning and deep learning models, such as Neural Networks, LSTM (Long Short-Term Memory), and Bi-LSTM (Bidirectional Long Short-Term Memory) with respect to different performance metrics like precision, accuracy, recall, F1-score, training time, and AUC-ROC (Area Under the Receiver Operating Characteristic Curve). The simulation results revealed that the proposed CNN model outperformed with respect to other models.

The study presents the results of the development and testing of deep learning neural network architectures, which demonstrate high accuracy rates in classifying neurophysiological data, in particular hemodynamic brain activation patterns obtained by functional near-infrared spectroscopy, during solving mathematical problems on spatial-numerical associations. The analyzed signal represents a multidimensional time series of oxyhemoglobin and deoxyhemoglobin dynamics. Taking the specificity of the fNIRS signal into account, a comparative analysis of 2 types of neural network architectures was carried out: (1) architectures based on recurrent neural networks: recurrent neural network with long short-term memory, recurrent neural network with long short-term memory with fully connected layers, bidirectional recurrent neural network with long short-term memory, convolutional recurrent neural network with long short-term memory; (2) architectures based on convolutional neural networks with 1D convolutions: convolutional neural network, fully convolutional neural network, residual neural network. Trained long short-term memory recurrent neural network architectures showed worse results in accuracy in comparison with 1D convolutional neural network architectures. Residual neural network (model_Resnet) demonstrated the highest accuracy rates in three experimental conditions more than 88% in detecting age-related differences in brain activation during spatial-numerical association tasks considering the individual characteristics of the respondents’ signal.

This paper presents a Gaussian data augmentation-assisted deep learning using a convolutional neural network (PCA18+GDA100+CNN LSTM) on the analysis of the state-of-the-art infrared backscatter imaging spectroscopy (IBIS) images. Both PCA and data augmentation methods were used to preprocess classification input and predict with a comparable degree of accuracy. Initially, PCA was used to reduce the number of features. We used 18 principal components based of the cumulative variance, which totaled 99.92%. GDA was also used to increase the number of samples. CNN-LSTM (long short-term memory) was then used to perform multiclass classification on the IBIS hyperspectral image. Experiments were conducted and results were collected from the K-fold cross-validation with K=20. They were analyzed with a confusion matrix and the average accuracy is 99%.

Emotion recognition in facial images is a topic that has attracted many interests. Research on emotion recognition in facial images continues to face challenges such as variations of human faces due to environments or situations. Of all the techniques that have been used, the Deep Convolution Neural Network (DCNN) method reports a significant improvement compared to traditional machine learning methods. In recent years, self-attachment mechanisms have increased in popularity in DCNN, especially in Natural Language Processing. In this study, we propose to add attention networks to DCNN method to improve the performance of emotion recognition. By incorporating self-attention mechanisms into the model architecture, the network is able to recognize which facial features are most relevant for each emotion, increasing the accuracy and interpretation of the model. This study reports an accuracy of 69.55% which is 11.15% higher than using the traditional DCNN model.

With the explosive growth in seismic data acquisition and the successful application of deep convolutional neural networks (DCNN) to various image processing tasks within multidisciplinary fields, many researchers have begun to research DCNN based automatic seismic interpretation techniques. Due to the vast number of parameters considered in deep neural networks, deep learning methods usually require a large amount of data for training. However, collecting a large number of expert interpretations is very time consuming, so related research usually uses synthetic datasets and ignores the practical problems of field datasets. In this paper, we open-source a multi-gigabyte expert-labelled field dataset in response to the challenge of accessing large-scale expert-labelled field datasets. We show that 2D fault recognition within this dataset is an image segmentation or edge detection problem in the computer vision field, that can be expressed as a pixel-level fault/non-fault binary classification. Both types of DCNNs are compared, and we propose a novel fault recognition workflow, which involves processing and screening of seismic images and labels, training DCNNs and automatic numerical evaluation. We have also demonstrated for three case study datasets that effective image augmentation methods can reduce the required labelled crosslines while maintaining satisfactory performance. Our experimental results show that our workflow not only outperforms two state-of-the-art DCNN solutions but also achieves performance comparable to humans on an expert labelled image dataset, even predicting subtle faults that an expert interpreter did not annotate. We suggest that the proposed workflow could reduce the fault interpretation life cycle from months to hours and improve the quality, and define the confidence, of fault interpretation results.

Urban air quality index forecasting using multivariate convolutional neural network based customized stacked long short-term memory model

Improving the performance of Deep Learning (DL) algorithms is a challenging problem. However, DL is applied to different types of Deep Neural Networks, and Long Short-Term Memory (LSTM) is one of them that deals with time series or sequential data. This paper attempts to overcome this problem by optimizing LSTM parameters using the Bees Algorithm (BA), which is a nature-inspired algorithm that mimics the foraging behavior of honey bees. In particular, it was used to optimize the adjustment factors of the learning rate in the forget, input, and output gates, in addition to cell candidate, in both forward and backward sides. Furthermore, the BA was used to optimize the learning rate factor in the fully connected layer. In this study, artificial porosity images were used for testing the algorithms; since the input data were images, a Convolutional Neural Network (CNN) was added in order to extract the features in the images to feed into the LSTM for predicting the percentage of porosity in the sequential layers of artificial porosity images that mimic real CT scan images of products manufactured by the Selective Laser Melting (SLM) process. Applying a Convolutional Neural Network Long Short-Term Memory (CNN-LSTM) yielded a porosity prediction accuracy of 93.17%. Although using Bayesian Optimization (BO) to optimize the LSTM parameters mentioned previously did not improve the performance of the LSTM, as the prediction accuracy was 93%, adding the BA to optimize the same LSTM parameters did improve its performance in predicting the porosity, with an accuracy of 95.17% where a hybrid Bees Algorithm Convolutional Neural Network Long Short-Term Memory (BA-CNN-LSTM) was used. Furthermore, the hybrid BA-CNN-LSTM algorithm was capable of dealing with classification problems as well. This was shown by applying it to Electrocardiogram (ECG) benchmark images, which improved the test set classification accuracy, which was 92.50% for the CNN-LSTM algorithm and 95% for both the BO-CNN-LSTM and BA-CNN-LSTM algorithms. In addition, the turbofan engine degradation simulation numerical dataset was used to predict the Remaining Useful Life (RUL) of the engines using the LSTM network. A CNN was not needed in this case, as there was no feature extraction for the images. However, adding the BA to optimize the LSTM parameters improved the prediction accuracy in the testing set for the LSTM and BO-LSTM, which increased from 74% to 77% for the hybrid BA-LSTM algorithm.

Parts of Speech (pos) tagging corresponds to techniques of tagging accurate pos tags of each word of a sentence. It is a pivotal and indispensable part of sentiment analysis, synthesizing text to speech, analyzing textual data and many more Natural Language Processing (NLP) tasks. Though there are many researches done in the field of languages like English, Chinese etc., it is still comparatively unexplored for Bangla. An excellent pos tagger in Bangla language can accelerate NLP of Bangla textual data. In this study, we have developed a novel hybrid Deep Learning (DL) model with two parallel input networks with Convolutional Neural Network (CNN), and Bidirectional Long Short Term Memory (BiLSTM), and an output network of CNN. CNN works well regarding the local dependencies of words in sentences, and BiLSTM works well with the global dependencies of words regarding the context in a sentence. No other work has been done with local features of textual data along with long term word dependencies in Bangla pos tagging studies. The model can work with any external Word Embeddings. In this study, it is implemented with both Embedding Layer and Word2Vec layer. In the evaluation process, Mathews Correlation Coefficient (MCC), Precision-Recall curve are used along with accuracy and F1-score due to the dataset's imbalanced characteristic. Our proposed model accompanied by Word2Vec layer has outperformed Vanilla Recurrent Neural Network (RNN), Gated Recurrent Units (GRU), Long Short Term Memory (LSTM), BiLSTM networks, and also one recent study using the same dataset with the highest accuracy of 0.974, F1-score of 0.883, and MCC value of 0.936.

In today's digital world, app stores have become an essential part of software distribution, providing customers with a wide range of applications and opportunities for software developers to showcase their work. This study elaborates on the importance of end-user feedback for software evolution. However, in the literature, more emphasis has been given to high-rating & popular software apps while ignoring comparatively low-rating apps. Therefore, the proposed approach focuses on end-user reviews collected from 64 low-rated apps representing 14 categories in the Amazon App Store. We critically analyze feedback from low-rating apps and developed a grounded theory to identify various concepts important for software evolution and improving its quality including user interface (UI) and user experience (UX), functionality and features, compatibility and device-specific, performance and stability, customer support and responsiveness and security and privacy issues. Then, using a grounded theory and content analysis approach, a novel research dataset is curated to evaluate the performance of baseline machine learning (ML), and state-of-the-art deep learning (DL) algorithms in automatically classifying end-user feedback into frequently occurring issues. Various natural language processing and feature engineering techniques are utilized for improving and optimizing the performance of ML and DL classifiers. Also, an experimental study comparing various ML and DL algorithms, including multinomial naive Bayes (MNB), logistic regression (LR), random forest (RF), multi-layer perception (MLP), k-nearest neighbors (KNN), AdaBoost, Voting, convolutional neural network (CNN), long short-term memory (LSTM), bidirectional long short term memory (BiLSTM), gated recurrent unit (GRU), bidirectional gated recurrent unit (BiGRU), and recurrent neural network (RNN) classifiers, achieved satisfactory results in classifying end-user feedback to commonly occurring issues. Whereas, MLP, RF, BiGRU, GRU, CNN, LSTM, and Classifiers achieved average accuracies of 94%, 94%, 92%, 91%, 90%, 89%, and 89%, respectively. We employed the SHAP approach to identify the critical features associated with each issue type to enhance the explainability of the classifiers. This research sheds light on areas needing improvement in low-rated apps and opens up new avenues for developers to improve software quality based on user feedback.

Detection of Pneumothorax with Deep Learning Models: Learning From Radiologist Labels vs Natural Language Processing Model Generated Labels

The Part 1 prequel to this review evaluated the evolution of modelling techniques used in evaluation of fruit quality over the past three decades and noted a progression towards the use of artificial neural networks (ANNs) and convolutional neural networks (CNNs). In this review, Part 2, the use of CNNs for NIR fruit quality evaluation is explored, given the success of CNNs in various other fields, such as image, video, speech, and audio processing, and the availability of large (open source) datasets of fruit spectra and reference quality attribute, which is required for the training of CNN models. The review provides an overview of deep learning and the CNN architectures and techniques used in NIR spectroscopy for regression modelling, with advantages and disadvantages identified. Studies using CNN for NIR based fruit quality evaluation are then critically examined. Eight publications have presented on models using the same open-source mango dry matter calibration and test set, enabling inter-method comparisons. CNN models have been demonstrated to be accurate, precise and robust. Techniques of transfer learning for CNN models offer an alternative solution to model updating and calibration transfer methods applied in traditional chemometrics. The review has highlighted crucial areas that require resolution and exploration in this application through future research, including, (i) data requirements for training a CNN (ii) optimal spectral pre-processing for CNN (iii) CNN architecture and hyper-parameter selection and tuning for fruit quality evaluation (iv) CNN model interpretability and explainability. Future studies must conduct clearer comparison to partial least squares (PLS) regression and shallow ANNs to better assess the prospective benefit of using CNN, a more complex model. The potential for visualisation of spectra relevance to the CNN model using techniques such as GradCam, currently employed in visualising 2D-CNN models, remains to be explored.

Rapid quantitative analysis of cotton-polyester blended fabrics using near-infrared spectroscopy combined with CNN-LSTM