Component Convolutional Neural Networks Research Articles

Exposing Deep Fake Face Detection using LSTM and CNN

The rapid advancement of deep learning techniques, creating realistic multimedia content has become increasingly accessible, leading to the proliferation of DeepFake technology. DeepFake utilizes generative deep learning algorithms to produce or modify face features in a highly realistic manner, often making it challenging to differentiate between real and manipulated media. This technology, while beneficial in fields such as entertainment and education, also poses significant threats, including misinformation and identity theft. Consequently, detecting DeepFakes has become a critical area of research. In this paper, we propose a novel approach to DeepFake face detection by integrating Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) networks. Our method leverages the strengths of CNNs in spatial feature extraction and LSTMs in temporal sequence modeling to enhance detection accuracy. The CNN component captures intricate facial features, while the LSTM analyzes the temporal dynamics of video frames. We evaluate our model on several benchmark datasets, including Celeb-DF (v2), DeepFake Detection Challenge Preview, and FaceForensics++. Experimental results demonstrate that our hybrid CNN-LSTM model achieves state-of-the-art performance, surpassing existing methods in both accuracy and robustness. This study highlights the potential of combining CNN and LSTM architectures for effective DeepFake detection, contributing to the ongoing efforts to safeguard against digital media manipulation

A Novel Hybrid Deep Learning Model for Complex Systems: A Case of Train Delay Prediction

Predicting the status of train delays, a complex and dynamic problem, is crucial for railway enterprises and passengers. This paper proposes a novel hybrid deep learning model composed of convolutional neural networks (CNN) and temporal convolutional networks (TCN), named the CNN + TCN model, for predicting train delays in railway systems. First, we construct 3D data containing the spatiotemporal characteristics of real-world train data. Then, the CNN + TCN model employs a 3D CNN component, which is fed into the constructed 3D data to mine the spatiotemporal characteristics, and a TCN component that captures the temporal characteristics in railway operation data. Furthermore, the characteristic variables corresponding to the two components are selected. Finally, the model is evaluated by leveraging data from two railway lines in the United Kingdom. Numerical results show that the CNN + TCN model has greater accuracy and convergence performance in train delay prediction.

Deepfake Detection System

Abstract: The emergence of deepfake technology presents a profound challenge to the integrity and trustworthiness of multimedia content online. To address this issue, this study proposes a novel deepfake detection system that integrates a hybrid Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) architecture. The proposed system employs a multi-faceted approach to training by utilizing several diverse datasets encompassing real and synthetic videos. This comprehensive training strategy ensures that the model learns robust features representative of both authentic and manipulated content across various contexts and scenarios. By incorporating a range of datasets, including those specifically curated to represent different deepfake generation techniques and quality levels, the model gains a more comprehensive understanding of the intricate nuances present in manipulated videos. During the training phase, the CNN component extracts high-level spatial features from individual frames of the input videos, effectively capturing visual patterns indicative of deepfake manipulation.

Convolutional neural network for improved event-based Shack-Hartmann wavefront reconstruction

Shack-Hartmann wavefront sensing is a technique for measuring wavefront aberrations, whose use in adaptive optics relies on fast position tracking of an array of spots. These sensors conventionally use frame-based cameras operating at a fixed sampling rate to report pixel intensities, even though only a fraction of the pixels have signal. Prior in-lab experiments have shown feasibility of event-based cameras for Shack-Hartmann wavefront sensing (SHWFS), asynchronously reporting the spot locations as log intensity changes at a microsecond time scale. In our work, we propose a convolutional neural network (CNN) called event-based wavefront network (EBWFNet) that achieves highly accurate estimation of the spot centroid position in real time. We developed a custom Shack-Hartmann wavefront sensing hardware with a common aperture for the synchronized frame- and event-based cameras so that spot centroid locations computed from the frame-based camera may be used to train/test the event-CNN-based centroid position estimation method in an unsupervised manner. Field testing with this hardware allows us to conclude that the proposed EBWFNet achieves sub-pixel accuracy in real-world scenarios with substantial improvement over the state-of-the-art event-based SHWFS. An ablation study reveals the impact of data processing, CNN components, and training cost function; and an unoptimized MATLAB implementation is shown to run faster than 800 Hz on a single GPU.

In-Season Cotton Yield Prediction with Scale-Aware Convolutional Neural Network Models and Unmanned Aerial Vehicle RGB Imagery.

In the pursuit of sustainable agriculture, efficient water management remains crucial, with growers relying on advanced techniques for informed decision-making. Cotton yield prediction, a critical aspect of agricultural planning, benefits from cutting-edge technologies. However, traditional methods often struggle to capture the nuanced complexities of crop health and growth. This study introduces a novel approach to cotton yield prediction, leveraging the synergy between Unmanned Aerial Vehicles (UAVs) and scale-aware convolutional neural networks (CNNs). The proposed model seeks to harness the spatiotemporal dynamics inherent in high-resolution UAV imagery to improve the accuracy of the cotton yield prediction. The CNN component adeptly extracts spatial features from UAV-derived imagery, capturing intricate details related to crop health and growth, modeling temporal dependencies, and facilitating the recognition of trends and patterns over time. Research experiments were carried out in a cotton field at the USDA-ARS Cropping Systems Research Laboratory (CSRL) in Lubbock, Texas, with three replications evaluating four irrigation treatments (rainfed, full irrigation, percent deficit of full irrigation, and time delay of full irrigation) on cotton yield. The prediction revealed that the proposed CNN regression models outperformed conventional CNN models, such as AlexNet, CNN-3D, CNN-LSTM, ResNet. The proposed CNN model showed state-of-art performance at different image scales, with the R2 exceeding 0.9. At the cotton row level, the mean absolute error (MAE) and mean absolute percentage error (MAPE) were 3.08 pounds per row and 7.76%, respectively. At the cotton grid level, the MAE and MAPE were 0.05 pounds and 10%, respectively. This shows the proposed model's adaptability to the dynamic interplay between spatial and temporal factors that affect cotton yield. The authors conclude that integrating UAV-derived imagery and CNN regression models is a potent strategy for advancing precision agriculture, providing growers with a powerful tool to optimize cultivation practices and enhance overall cotton productivity.

Short-term photovoltaic power forecasting with feature extraction and attention mechanisms

The uncertainty of weather conditions has always been a major challenge limiting the performance of photovoltaic (PV) power prediction. Enhancing the accuracy and stability of PV power prediction is crucial for optimizing grid operation. In response to this challenge, this study introduces a hybrid model that integrates an attention mechanism with the Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory Network (BiLSTM). This model aims to mitigate the adverse impact of weather variability on the accuracy of PV power prediction by effectively extracting key features from multidimensional time series data. Additionally, through ablation experiments, this research further assesses the contribution of each model component to performance. Validated with actual data collected from a 1 MW PV power station in China, our proposed model demonstrates significant performance advantages compared to eight advanced prediction models. Ablation study results reveal that removing the CNN component led to a 58.2% increase in MAE and a 53.9% increase in RMSE, while the removal of the attention mechanism resulted in an 83.8% increase in maximum error. These findings underscore the substantial enhancement in prediction accuracy achieved through the integration of CNN and BiLSTM, and the introduction of the attention mechanism significantly boosts the model's prediction stability.

A review of convolutional neural networks in computer vision

In computer vision, a series of exemplary advances have been made in several areas involving image classification, semantic segmentation, object detection, and image super-resolution reconstruction with the rapid development of deep convolutional neural network (CNN). The CNN has superior features for autonomous learning and expression, and feature extraction from original input data can be realized by means of training CNN models that match practical applications. Due to the rapid progress in deep learning technology, the structure of CNN is becoming more and more complex and diverse. Consequently, it gradually replaces the traditional machine learning methods. This paper presents an elementary understanding of CNN components and their functions, including input layers, convolution layers, pooling layers, activation functions, batch normalization, dropout, fully connected layers, and output layers. On this basis, this paper gives a comprehensive overview of the past and current research status of the applications of CNN models in computer vision fields, e.g., image classification, object detection, and video prediction. In addition, we summarize the challenges and solutions of the deep CNN, and future research directions are also discussed.

DDOS ATTACK DETECTION USING HYBRID (CCN AND LSTM) ML MODEL

LSTM (Long Short-Term Memory) and CNN (Convolutional Neural Networks) are two types of deep learning algorithms; by combining the strengths of LSTM and CNN, researchers have developed deep learning models that can effectively detect SDN (Software-Defined Network) attacks including Distributed Denial of Service. These models effectively analyze network traffic, encompassing temporal and spatial characteristics, resulting in precise identification of malicious traffic.In this research, a hybrid model composed of CNN and LSTM is used to detect the DDoS attack in SDN network. Where the CNN component of the model can identify spatial patterns in network traffic, such as the characteristics of individual packets, while the LSTM component can capture temporal patterns in traffic over time, such as the timing and frequency of traffic bursts. The proposed model has been trained on a labeled network traffic dataset, with one class representing normal traffic and another class representing DDoS attack traffic. During the training process, the model adjusts its weights and biases to minimize the difference between its predicted output and the actual output for each input sample. Once trained, the hybrid model classifies incoming network traffic in the dataset as either normal or malicious with an initial accuracy of (78.18%) and losses of (39.77%) at the 1st epoch till it reaches an accuracy of (99.99%) with losses of (9.29×10-5) at the epoch number 500. It should be mentioned that the hybrid model of CNN and LSTM for DDoS detection is implemented using Python Anaconda platform with an ETA 28ms/step.

DEEPFAKE DETECTION USING DEEP LEARNING (CNN+LSTM)

Artificial intelligence advancements have led to the development of deepfake technology, which seriously jeopardises the integrity of visual media material. Robust detection algorithms are becoming more and more necessary as deepfake creation techniques become more complex. This study combines Long Short-Term Memory (LSTM) networks with Convolutional Neural Networks (CNNs) to present a novel method for deepfake identification. The suggested CNN+LSTM architecture makes use of LSTMs' temporal modelling capabilities and CNNs' spatial feature extraction capabilities. While the LSTM component analyses the temporal connections between frames to identify patterns suggestive of deepfake manipulation, the CNN component concentrates on capturing local features and patterns in individual frames. The combination of these two networks improves the model's capacity to identify minute anomalies and inconsistencies that are indicative of deepfake content. To extract frame-level characteristics, we use Res-Next Convolutional Neural Networks. These attributes are then used to train a Recurrent Neural Network (RNN) based on Long Short-Term Memory (LSTM) to determine whether a video has been manipulated, i.e., whether it is a deepfake or a genuine video. We intend to train our deepfake detection model on a varied set of public datasets in order to improve its real-time performance. We improve the model's adaptability by learning features from different photos. Face-Forensic++, Deepfake Detection Challenge, and Celeb-DF datasets are used to extract videos. Furthermore, to assure competitive performance in real-world scenarios, our model will be assessed against a large amount of real-time data, including the YouTube dataset. Key Words: Temporal modelling, Deepfake technology , Long Short-Term Memory (LSTM) networks, Convolutional Neural Networks (CNNs)

Comparison of models of deep convolutional neural networks

Recently, deep learning has gained considerable success and acceptance in a variety of fields, attracting an increasing number of researchers who are delving deeper and gaining a broader perspective on the subject. It provides more sustainability and opportunities to advance the development of society and transform the lives of individuals. Consequently, it is crucial for individuals to understand the neural network development path. This paper provides a concise overview of the structure and components of Convolutional Neural Network, as well as some of the most well-known and influential learning models in the history of its development. Through an analysis of various models of convolutional neural network, the workings of convolutional neural networks were investigated. The paper discovered that the structure of neural networks is becoming deeper and more complex in order to achieve greater efficacy and avoid the overfitting issue. For researchers to enhance and advance neural network performance, there are still numerous parameters and perspectives to improve and advance.

OR32-02 Thyroid Cancer Polygenic Risk Score Improves Risk Stratification Of Thyroid Nodules When Added To Ultrasound Imaging

Abstract Disclosure: N. Pozdeyev: None. M. Dighe: None. M. Barrio: None. C. Raeburn: None. H.A. Smith: None. M. Fisher: None. S. Chavan: None. N. Rafaels: None. J. Shortt: None. M. Leu: None. T. Clark: None. C. Marshall: None. B.R. Haugen: None. D. Subramanian: None. K. Crooks: None. C. Gignoux: None. T.A. Cohen: None. Purpose. Evaluating thyroid nodules to rule out malignancy is a common clinical task. Image-based risk stratification schemas rely on the presence of high-risk thyroid nodule sonographic features and, therefore, are less suitable for the diagnosis of malignant thyroid nodules that have a benign appearance on the ultrasound. To mitigate the deficiency of thyroid nodule evaluation relying solely on the sonographic characteristics, we used thyroid cancer polygenic risk score (PRS) to complement deep learning analysis of ultrasound images. Methods. A supervised deep learning classifier of thyroid nodules was trained on 32,545 thyroid US images from 621 nodules and tested on an independent set of 232 nodules from patients genotyped on the Illumina's MEGAEX platform. The deep-learning thyroid nodule classifier was developed by fine-tuning a BiT-M ResNet-50x1 convolutional neural network (CNN) pre-trained on the ImageNet-21k dataset. A polygenic risk score (PRS) was calculated using thyroid cancer genome-wide association meta-analysis summary statistics from the Global Biobank Meta-analysis Initiative. The thyroid cancer PRS was defined as a weighted sum of five alleles with the strongest association with thyroid cancer. CNN predictions and PRS were combined into a meta-classifier using logistic regression with or without genetic ancestry and demographic covariates. Results. The CNN classifier achieved an area under the receiver operating characteristic curve (AUC) of 0.83 on the out-of-sample test set of 232 thyroid nodules. The CNN classifier incorrectly classified thyroid nodules without suspicious sonographic characteristics belonging to difficult-to-diagnose subtypes such as follicular thyroid cancer and follicular variant of papillary thyroid cancer. Combining predictions from the CNN classifier with PRS into a cross-validated classifier improved AUC to 0.868 (DeLong test, p = 0.05). Incorporating genetic ancestry in the form of five genetic principal components further improved AUC of the benign vs. malignant CNN + PRS thyroid nodule classifier to 0.885 (p = 0.007). Finally, when age, sex, and nodule dimensions were considered, the AUC of the meta-classifier increased to 0.915 (p = 2.3e-4). The meta-classifier including predictions from CNN, PRS, and genetic principal components showed a sensitivity of 0.95, specificity of 0.61, NPV of 0.97, and PPV of 0.5. This performance was superior to that of the clinical Thyroid Imaging Reporting and Data System (TI-RADS) as reported by radiologists in ultrasound reports. Conclusions. For the first time, we showed PRS provides an orthogonal thyroid cancer risk assessment complementary to the ultrasound image-based thyroid nodule risk evaluation. This proof-of-concept study opens an opportunity for developing next-generation schemas for thyroid nodule evaluation incorporating clinical, imaging, and genetic data. Presentation: Sunday, June 18, 2023

Lung cancer disease identification using hybrid models

Using hybrid models, we present a novel method for detecting lung cancer in this study. Our method uses the random forest and convolutional neural network (CNN) techniques to incorporate machine learning and deep learning advantages. The proposed composite method combines structured clinical data with unprocessed imaging data for a more complete lung cancer diagnosis. The CNN component of our hybrid model excels at extracting features from images of lung cancer, while the random forest component excels at capturing complex relationships in structured data. For greater precision and consistency, the results of the two models may be averaged. The hybrid model outperforms the existing methods. The hybrid model acquired an accuracy rate of 98%. Future lung cancer detection will be rapid and accurate due to the hybrid model’s improved performance and decreased inference periods.

Energy efficiency optimization and carbon emission reduction targets of resource-based cities based on BiLSTM-CNN-GAN model

IntroductionEnergy consumption and carbon emissions are major global concerns, and cities are responsible for a significant portion of these emissions. To address this problem, deep learning techniques have been applied to predict trends and influencing factors of urban energy consumption and carbon emissions, and to help formulate optimization programs and policies.MethodsIn this paper, we propose a method based on the BiLSTM-CNN-GAN model to predict urban energy consumption and carbon emissions in resource-based cities. The BiLSTMCNN-GAN model is a combination of three deep learning techniques: Bidirectional Long Short-Term Memory (BiLSTM), Convolutional Neural Networks (CNN), and Generative Adversarial Networks (GAN). The BiLSTM component is used to process historical data and extract time series information, while the CNN component removes spatial features and local structural information in urban energy consumption and carbon emissions data. The GAN component generates simulated data of urban energy consumption and carbon emissions and optimizes the generator and discriminator models to improve the quality of generation and the accuracy of discrimination.Results and discussionThe proposed method can more accurately predict future energy consumption and carbon emission trends of resource-based cities and help formulate optimization plans and policies. By addressing the problem of urban energy efficiency and carbon emission reduction, proposed method contributes to sustainable urban development and environmental protection.

On the Analyses of Medical Images Using Traditional Machine Learning Techniques and Convolutional Neural Networks

Convolutional neural network (CNN) has shown dissuasive accomplishment on different areas especially Object Detection, Segmentation, Reconstruction (2D and 3D), Information Retrieval, Medical Image Registration, Multi-lingual translation, Local language Processing, Anomaly Detection on video and Speech Recognition. CNN is a special type of Neural Network, which has compelling and effective learning ability to learn features at several steps during augmentation of the data. Recently, different interesting and inspiring ideas of Deep Learning (DL) such as different activation functions, hyperparameter optimization, regularization, momentum and loss functions has improved the performance, operation and execution of CNN Different internal architecture innovation of CNN and different representational style of CNN has significantly improved the performance. This survey focuses on internal taxonomy of deep learning, different models of vonvolutional neural network, especially depth and width of models and in addition CNN components, applications and current challenges of deep learning.

An enhancement of transformer-based architecture with randomized regularization for wind speed prediction

The wind power is considered as a potential renewable energy resource which requires less management cost and effort than the others like as tidal, geothermal, etc. However, the natural randomization and volatility aspects of wind in different regions have brought several challenges for efficiently as well as reliably operating the wind-based power supply grid. Thus, it is necessary to have centralized monitoring centers for managing as well as optimizing the performance of wind power farms. Among different management task, wind speed prediction is considered as an important task which directly support for further wind-based power supply resource planning/optimization, hence towards power shortage risk and operating cost reductions. Normally, considering as traditional time-series based prediction problem, most of previous deep learning-based models have demonstrated significant improvement in accuracy performance of wind speed prediction problem. However, most of recurrent neural network (RNN) as well as sequential auto-encoding (AE) based architectures still suffered several limitations related to the capability of sufficient preserving the spatiotemporal and long-range time dependent information of complex time-series based wind datasets. Moreover, previous RNN-based wind speed predictive models also perform poor prediction results within high-complex/noised time-series based wind speed datasets. Thus, in order to overcome these limitations, in this paper we proposed a novel integrated convolutional neural network (CNN)-based spatiotemporal randomization mechanism with transformer-based architecture for wind speed prediction problem, called as: RTrans-WP. Within our RTrans-WP model, we integrated the deep neural encoding component with a randomized CNN learning mechanism to softy align temporal feature within the long-range time-dependent learning context. The utilization of randomized CNN component at the data encoding part also enables to reduce noises and time-series based observation uncertainties which are occurred during the data representation learning and wind speed prediction-driven fine-tuning processes.

Seizure Detection and Prediction by Parallel Memristive Convolutional Neural Networks.

During the past two decades, epileptic seizure detection and prediction algorithms have evolved rapidly. However, despite significant performance improvements, their hardware implementation using conventional technologies, such as Complementary Metal-Oxide-Semiconductor (CMOS), in power and area-constrained settings remains a challenging task; especially when many recording channels are used. In this paper, we propose a novel low-latency parallel Convolutional Neural Network (CNN) architecture that has between 2-2,800x fewer network parameters compared to State-Of-The-Art (SOTA) CNN architectures and achieves 5-fold cross validation accuracy of 99.84% for epileptic seizure detection, and 99.01% and 97.54% for epileptic seizure prediction, when evaluated using the University of Bonn Electroencephalogram (EEG), CHB-MIT and SWEC-ETHZ seizure datasets, respectively. We subsequently implement our network onto analog crossbar arrays comprising Resistive Random-Access Memory (RRAM) devices, and provide a comprehensive benchmark by simulating, laying out, and determining hardware requirements of the CNN component of our system. We parallelize the execution of convolution layer kernels on separate analog crossbars to enable 2 orders of magnitude reduction in latency compared to SOTA hybrid Memristive-CMOS Deep Learning (DL) accelerators. Furthermore, we investigate the effects of non-idealities on our system and investigate Quantization Aware Training (QAT) to mitigate the performance degradation due to low Analog-to-Digital Converter (ADC)/Digital-to-Analog Converter (DAC) resolution. Finally, we propose a stuck weight offsetting methodology to mitigate performance degradation due to stuck [Formula: see text] memristor weights, recovering up to 32% accuracy, without requiring retraining. The CNN component of our platform is estimated to consume approximately 2.791 W of power while occupying an area of 31.255 mm2 in a 22 nm FDSOI CMOS process.

Algorithm/Hardware Codesign for Real-Time On-Satellite CNN-Based Ship Detection in SAR Imagery

Recently, the convolutional neural network (CNN)-based approach for on-satellite ship detection in synthetic aperture radar (SAR) images has received increasing attention since it does not rely on predefined imagery features and distributions that are required in conventional detection methods. To achieve high detection accuracy, most of the existing CNN-based methods leverage complex off-the-shelf CNN models for optical imagery. Unfortunately, this usually leads to expensive computational cost, which is hard to process in real time using resource-constrained devices deployed in the harsh satellite environment. In this article, we propose OSCAR-RT, the first end-to-end algorithm/hardware codesign framework for real-time on-satellite CNN-based SAR ship detection, which can simultaneously produce an accurate and hardware-friendly CNN model and an ultraefficient field-programmable gate array (FPGA)-based hardware accelerator that can be deployed on satellites. With the real-time on-satellite processing speed in mind, we start from a state-of-the-art compact CNN model for optical imagery. To eliminate the sharp decrease in the detection accuracy for SAR imagery, we analyze the discrepancy between the SAR domain and optical domain and propose to adapt the model by adjusting the output feature size to better detect relatively smaller objects in SAR imagery. To improve the detection speed, we propose to develop a fully pipelined interlayer streaming accelerator architecture, where all the layers of the CNN model can be concurrently processed using on-chip FPGA resources. To achieve this architecture, we first propose a hardware-guided, progressive, and structural pruning strategy, which is guided by our modeled hardware metrics and applies state-of-the-art coarse-grained and fine-grained filter pruning as well as mixed-precision quantization techniques. Moreover, to improve the reusability and portability of the hardware accelerator design, we develop a library of highly optimized CNN components in high-level synthesis, together with their performance and resource models. Finally, we map the pruned CNN model onto these hardware library components in a fully pipelined interlayer streaming fashion, by adjusting their parallelism factors to balance the execution of each layer and fit into the resource constraint. Experimental results using the adapted MobileNetV1, MobileNetV2, and SqueezeNet models on the widely used SAR ship detection dataset (SSDD) demonstrate the effectiveness of OSCAR-RT; for the MobileNetV1 model, it achieves an average precision of 94%, a detection speed of 652 frames/s on the Xilinx VC709 FPGA evaluation board while consuming about 5.8-W power.

Deep Learning for Classification and Selection of Cine CMR Images to Achieve Fully Automated Quality-Controlled CMR Analysis From Scanner to Report.

Introduction: Deep learning demonstrates great promise for automated analysis of CMR. However, existing limitations, such as insufficient quality control and selection of target acquisitions from the full CMR exam, are holding back the introduction of deep learning tools in the clinical environment. This study aimed to develop a framework for automated detection and quality-controlled selection of standard cine sequences images from clinical CMR exams, prior to analysis of cardiac function.Materials and Methods: Retrospective study of 3,827 subjects that underwent CMR imaging. We used a total of 119,285 CMR acquisitions, acquired with scanners of different magnetic field strengths and from different vendors (1.5T Siemens and 1.5T and 3.0T Phillips). We developed a framework to select one good acquisition for each conventional cine class. The framework consisted of a first pre-processing step to exclude still acquisitions; two sequential convolutional neural networks (CNN), the first (CNNclass) to classify acquisitions in standard cine views (2/3/4-chamber and short axis), the second (CNNQC) to classify acquisitions according to image quality and orientation; a final algorithm to select one good acquisition of each class. For each CNN component, 7 state-of-the-art architectures were trained for 200 epochs, with cross entropy loss and data augmentation. Data were divided into 80% for training, 10% for validation, and 10% for testing.Results: CNNclass selected cine CMR acquisitions with accuracy ranging from 0.989 to 0.998. Accuracy of CNNQC reached 0.861 for 2-chamber, 0.806 for 3-chamber, and 0.859 for 4-chamber. The complete framework was presented with 379 new full CMR studies, not used for CNN training/validation/testing, and selected one good 2-, 3-, and 4-chamber acquisition from each study with sensitivity to detect erroneous cases of 89.7, 93.2, and 93.9%, respectively.Conclusions: We developed an accurate quality-controlled framework for automated selection of cine acquisitions prior to image analysis. This framework is robust and generalizable as it was developed on multivendor data and could be used at the beginning of a pipeline for automated cine CMR analysis to obtain full automatization from scanner to report.

A Model of Convolutional Neural Network Combined with External Knowledge to Measure the Question Similarity for Community Question Answering Systems

Automatically determining similar questions and ranking the obtained questions according to their similarities to each input question is a very important task to any community Question Answering system (cQA). Various methods have applied for this task including conventional machine learning methods with feature extraction and some recent studies using deep learning methods. This paper addresses the problem of how to combine advantages of different methods into one unified model. Moreover, deep learning models are usually only effective for large data, while training data sets in cQA problems are often small, so the idea of integrating external knowledge into deep learning models for this cQA problem becomes more important. To this objective, we propose a neural network-based model which combines a Convolutional Neural Network (CNN) with features from other methods so that the deep learning model is enhanced with addtional knowledge sources. In our proposed model, the CNN component will learn the representation of two given questions, then combined additional features through a Multilayer Perceptron (MLP) to measure similarity between the two questions. We tested our proposed model on the SemEval 2016 task-3 data set and obtain better results in comparison with previous studies on the same task.

Modeling train timetables as images: A cost-sensitive deep learning framework for delay propagation pattern recognition

As a vital component of train operational control, train delay propagation pattern discovery is critically important for both railway controllers and passengers. In this study, we present a carefully designed deep learning model, called FCF-Net, that comprises fully connected neural networks (FCNN) and convolutional neural networks (CNN) for train delay propagation pattern recognition in railway systems. FCF-Net first uses a CNN component that handles train timetables as images to capture interactions of train events and an FCNN component to capture the influence of non-operational features separately; then it uses another FCNN component to combinedly learn the dependencies between operational and non-operational features. In addition, considering the imbalance of train delay data, a cost-sensitive technique that assigns different misclassification costs for different class was used to better deal with the imbalanced data. The main goal of the FCF-Net is to realize efficient and accurate train delay propagation pattern recognition by mining potential knowledge from train operation data. The predictive and computational performance of the model was tested and evaluated on data from two high-speed railway lines with different operational features in China. The results show that FCF-Net, once trained with sufficient data, outperforms conventional deep learning with common loss and other state-of-the-art deep learning models for train delay propagation pattern recognition, indicating its capability in knowledge discovery from train operation data. In addition, the computational results show that FCF-Net exhibits more efficient training process than existing state-of-the-art deep learning models.