Search

As an emerging technology for artificial intelligence‐aided medical diagnosis, deep learning combined with Raman spectroscopy has great potential. The technology still has some problems in the actual medical diagnosis research process. The differences in spectrometers, experimental conditions, and experimental operations can result in non‐uniform and universally applicable data standards, which in turn lead to low data utilization. At the same time, it is still necessary to retrain the models when building diagnostic models for different diseases, which is time‐consuming and laborious. In this paper, a more complete transfer learning model for multiple types of serum Raman spectra is established for the first time, and a decision fusion strategy is applied to this diagnostic model. The Raman spectral data of serum from hepatitis B patients/control group, serum from abnormal thyroid function patients/control group, and serum from glioma patients/control group were selected as the source domains, and the Raman spectral data of tissue from hepatitis C patients/control group, serum from esophageal cancer patients/control group, and tissue from cervical cancer and cervical inflammation (patients/control) group were selected as the target domains. Three deep neural network models, ResNet, GoogLeNet, and CNN‐LSTM were trained in the source domain data for disease diagnosis, and the trained models were transfer to the target domain. The model is fine‐tuned by freezing different layers and then combined with logistic regression algorithms to construct a decision fusion model, which further improves the model effect. The results show that the proposed method can effectively improve the accuracy of transfer learning models. At the same time, this experiment extends the application of transfer learning in Raman spectroscopy and demonstrates that unrelated and scale‐different Raman datasets are still intrinsically connected, which also lays the foundation for us to build more stable and data‐inclusive spectral transfer learning fusion models in the future. [ABSTRACT FROM AUTHOR]

Hadoop is a popular computing framework designed to deliver timely and cost-effective data processing on a large cluster of commodity machines. It relieves the burden of the programmers dealing with distributed programming, and an ecosystem of Big Data solutions has developed around it. However, Hadoop's job execution time can greatly depend on its runtime configurations and resource selections. Given the more than 100 job configuration settings provided by Hadoop, and diverse resource instance options in a cloud or virtualized computing environment, running Hadoop jobs still requires a substantial amount of expertise and experience. To address this challenge, we apply a deep neural network to predict Hadoop's job time based on historical execution data, and propose optimization methods to reduce job execution time and cost. The results show that our prediction method achieves almost 90% time prediction accuracy and clearly outperforms three other state-of-the-art regression-based prediction methods. Based on the time prediction, our proposed configuration search method and job scheduling algorithm successfully shorten the execution time of a single Hadoop job by more than a factor of 2 and reduce the time of processing a batch of Hadoop jobs by 40%∼65%. • We aim to optimize the performance and cost of running Hadoop MapReduce workload in a cloud environment with on-demand resources. • We propose a deep learning approach to predict the execution time of MapReduce jobs under any given resource allocation and Hadoop configuration setting. • We proposed a 2-step approximate search algorithm for finding the optimal resource allocation and execution configuration of individual job, and a 2D variable size bin packing algorithm for maximize overall resource utilization. • The experiment results show that our prediction method achieved almost 90% of time prediction accuracy and clearly out-performed three other state-of-art machine learning prediction methods. • Our optimization strategies successfully shorten the execution time of a single Hadoop job by more than a factor of 2 and reduce the time of processing a batch of Hadoop jobs by 40%∼65%. [ABSTRACT FROM AUTHOR]

Atmospheric effects are among the primary error sources affecting the accuracy of interferometric synthetic aperture radar (InSAR). The topography-dependent atmospheric effect is particularly noteworthy in reservoir areas for landslide monitoring utilizing InSAR, which must be effectively corrected to complete the InSAR high-accuracy measurement. This paper proposed a topography-dependent atmospheric correction method based on the Multi-Layer Perceptron (MLP) neural network model combined with topography and spatial data information. We used this proposed approach for the atmospheric correction of the interferometric pairs of Sentinel-1 images in the Baihetan dam. We contrasted the outcomes with those obtained using the generic atmospheric correction online service for InSAR (GACOS) correction and the traditional linear model correction. The results indicated that the MLP neural network model correction reduced the phase standard deviation of the Sentinel-1 interferogram by an average of 64% and nearly eliminated the phase-elevation correlation. Both comparisons outperformed the GACOS correction and the linear model correction. Through two real-world examples, we demonstrated how slopes with displacements, which were previously obscured by a significant topography-dependent atmospheric delay, could be successfully and clearly identified in the interferograms following the correction by the MLP neural network. The topography-dependent atmosphere can be better corrected using the MLP neural network model suggested in this paper. Unlike the previous model, this proposed approach could be adjusted to fit each interferogram, regardless of how much of the topography-dependent atmosphere was present. In order to improve the effectiveness of DInSAR and time-series InSAR solutions, it can be applied immediately to the interferogram to retrieve the effective displacement information that cannot be identified before the correction. [ABSTRACT FROM AUTHOR]

Distributed training is needed to shorten the training time of deep neural networks. However, the communication overhead often hurts performance efficiency, especially in a distributed computing environment with limited network bandwidth. Hence, gradient compression techniques have been proposed to reduce communication time. But, compression also has the risk of causing lower model accuracy and longer training time due to compression loss and compression time. As a result, compression may not consistently achieve desired results, and there are limited discussions on when and which compression should be used. To address this problem, we propose a performance-driven hybrid compression solution. We make three main contributions. (1) We describe a hybrid compression strategy that chooses the compression method for individual model gradients. (2) We build an offline performance estimator and an online loss monitor to ensure the compression decision can minimize training time without sacrificing mode accuracy. (3) Our implementation can be imported to existing deep learning frameworks and applicable to a wide range of compression methods. Up to 3.6x training performance speedup was observed compared to other state-of-the-art methods. • Our goal is to constrict a better compression strategy for training a DNN model in distributed computing environment. • Our proposed compression strategy is fine-grained, hybrid and performance-driven. • Our method is built based on an offline performance estimator, an online loss monitor, and a linear search time optimization algorithm. • Our approach is implemented as a lightweight library that can be imported to existing deep learning frameworks. • Our evaluations show that we achieved up to 3.6x training performance speedup compared to other compression methods. [ABSTRACT FROM AUTHOR]

In steerable filters, a filter of arbitrary orientation can be generated by a linear combination of a set of “basis filters.” Steerable properties dominate the design of the traditional filters, e.g., Gabor filters and endow features the capability of handling spatial transformations. However, such properties have not yet been well explored in the deep convolutional neural networks (DCNNs). In this paper, we develop a new deep model, namely, Gabor convolutional networks (GCNs or Gabor CNNs), with Gabor filters incorporated into DCNNs such that the robustness of learned features against the orientation and scale changes can be reinforced. By manipulating the basic element of DCNNs, i.e., the convolution operator, based on Gabor filters, GCNs can be easily implemented and are readily compatible with any popular deep learning architecture. We carry out extensive experiments to demonstrate the promising performance of our GCNs framework, and the results show its superiority in recognizing objects, especially when the scale and rotation changes take place frequently. Moreover, the proposed GCNs have much fewer network parameters to be learned and can effectively reduce the training complexity of the network, leading to a more compact deep learning model while still maintaining a high feature representation capacity. The source code can be found at https://github.com/bczhangbczhang. [ABSTRACT FROM AUTHOR]

As a highly prevalent and recurrent cancer, detecting the degree of differentiation in oral cancer is crucial. Current methods rely on biopsies in the presence of significant lesions, which are time-consuming. This study introduces an efficient and rapid approach for mass determination of oral cancer differentiation levels. By leveraging serum Raman spectroscopy combined with deep neural networks and extreme gradient boosting, we performed feature selection and interaction on oral cancer samples of varying differentiation levels, achieving the most accurate and reliable classification of oral cancer differentiation stages. [ABSTRACT FROM AUTHOR]

Recently, the deep learning schemes have been well investigated for improving the driving safety and efficiency in the transportation systems. In this paper, a probabilistic model named as CPGN (Collision Prediction model based on GA-optimized Neural Network) for decision-making in the rear-end collision avoidance system is proposed, targeting modeling the impact of important influential factors of collisions on the occurring probability of possible accidents in the Internet of Vehicles (IoV). The decision on how to serve the chauffeur is determined by a typical deep learning model, i.e., the BP neural network through evaluating the possible collision risk with V2I (Vehicle-to-Infrastructure) communication, V2V (Vehicle-to-Vehicle) communication and GPS infrastructure supporting. The proper structure of our BP neural network model is deeply learned with training data generated from VISSIM with multiple influential factors considered. In addition, since the selection of the connection coefficient array and thresholds of the neural network has great randomness, a local optimization issue is readily occurring during the modeling procedure. To overcome this problem and consider the ability to efficiently find out a global optimization, this paper chooses the genetic algorithm to optimize the coefficient array and thresholds of proposed neural network. For the purpose of enhancing the convergence speed of the proposed model, we further adjust the studying rate according to the relationship between the actual and predicated values of two adjacent iterations. Simulation results demonstrate that the proposed collision risk evaluation framework could offer rationale estimations to the possible collision risk in car-following scenarios for the next discrete monitoring interval. [ABSTRACT FROM AUTHOR]

In recent years, liver cancer has caused great harm to human health. This study aims to use serum Raman spectroscopy combined with deep learning algorithms to classify liver cancer patients and control groups. For improving the robustness of models, we added equal proportions of Gaussian white noise of 5, 10, 15, 20, 25 dBW to enhance the data, and compared the results of convolutional neural networks and long short-term memory networks which show that the convolutional neural network combined with 10-fold data enhancement was better. The accuracy was 96.95%, indicating the great potential of this experimental model in liver cancer detection. [ABSTRACT FROM AUTHOR]

• A deep learning method based on CNN and LSTM was used to recognise pig aggression. • This study directly process video episodes rather than individual frames. • CNN features were input into LSTM framework to extract spatial-temporal features. • The proposed method could recognise aggressive episodes with an accuracy of 97.2%. • A frame skipping approach can improve the accuracy to 98.4% and halve running time. Aggression is considered as a major animal welfare problem in commercial pig farming. The aim of this study is to develop a deep learning method based on convolutional neural network (CNN) and long short-term memory (LSTM) to recognise aggressive episodes of pigs. Compared to previous studies of pig behaviours based on deep learning, this study directly process video episodes rather than individual frames. In the experiment, nursery pigs (8/pen) were mixed for 3 days and then 8 h of video was recorded in each day. From these videos, 600 aggressive 2 s-episodes were manually selected and then augmented into 2400 episodes by using horizontal, vertical and diagonal mirroring. From the videos, 2400 non-aggressive 2 s-episodes were also manually selected. 80% of the data were randomly allocated as training set and the remaining 20% as validation set. Firstly, the CNN architecture VGG-16 was used to extract spatial features. These features were then input into LSTM framework to further extract temporal features. Through fully connected layer, the prediction function Softmax was finally used to determine if the current episode is aggression or non-aggression. Using the proposed method, aggressive episodes could be recognised with an accuracy of 97.2%. This result indicates that this method can be used to recognise aggressive episodes of pigs. Additionally, this paper further investigates the validity of this method under the conditions of skipping frames and reducing the episode length. The results show that a frame skipping approach whereby 30 fps is reduced into 15 fps within each 2 s-episode can improve the accuracy into 98.4% and halve the total running time. [ABSTRACT FROM AUTHOR]

Deep learning models, especially the convolutional neural networks (CNNs), are very active in hyperspectral remote sensing image classification. In order to better apply the CNN model to hyperspectral classification, we propose a CNN model based on Fletcher–Reeves algorithm (F–R CNN), which uses the Fletcher–Reeves (F–R) algorithm for gradient updating to optimize the convergence performance of the model in classification. In view of the fact that there are fewer optional training samples in practical applications, we further propose a method of increasing the number of samples by adding a certain degree of perturbed samples, which can also test the anti-interference ability of classification methods. Furthermore, we analyze the anti-interference and convergence performance of the proposed model in terms of different training sample data sets, different batch training sample numbers and iteration time. In this paper, we describe the experimental process in detail and comprehensively evaluate the proposed model based on the classification of CHRIS hyperspectral imagery covering coastal wetlands, and further evaluate it on a commonly used hyperspectral image benchmark dataset. The experimental results show that the accuracy of the two models after increasing training samples and adjusting the number of batch training samples is improved. When the number of batch training samples is continuously increased to 350, the classification accuracy of the proposed method can still be maintained above 80.7%, which is 2.9% higher than the traditional one. And its time consumption is less than that of the traditional one while ensuring classification accuracy. It can be concluded that the proposed method has anti-interference ability and outperforms the traditional CNN in terms of batch computing adaptability and convergence speed. [ABSTRACT FROM AUTHOR]

Rear-end collision prediction has gained an increasing attention for safety improvement in smart cities. It is urgent to design efficient warning strategies for rear-end collisions which is one of the main causes of traffic accidents. The existing researches have been conducted to predict the collisions. Learning-based methods are proposed to solve this complicated issue, which the traditional methods are difficult to solve. However, due to some limitations in terms of the feature extraction and prediction performance, back-propagation learning methods are facing challenge. In this paper, we proposed a novel Rear-end Collision Prediction Mechanism with deep learning method (RCPM), in which a convolutional neural network model is established. In RCPM, the dataset is smoothed and expanded based on genetic theory to alleviate the class imbalance problem. The preprocessed dataset is divided into training and testing sets as the input to train our convolutional neural network model. The experimental results show that compared with the Honda, Berkeley and multi-layer perception neural-network-based algorithms, RCPM effectively improves performance to predict rear-end collisions. [ABSTRACT FROM AUTHOR]