scholarly journals Spatio-temporal prediction of the COVID-19 pandemic in US counties: modeling with a deep LSTM neural network

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Behnam Nikparvar ◽  
Md. Mokhlesur Rahman ◽  
Faizeh Hatami ◽  
Jean-Claude Thill
Keyword(s):  
Neural Network ◽  
Short Term Memory ◽  
Predictive Performance ◽  
Small Data ◽  
Multiple Time ◽  
Ensemble Model ◽  
Temporal Prediction ◽  
Us Counties ◽  
Spatio Temporal ◽  
Similar Accuracy

AbstractPrediction of complex epidemiological systems such as COVID-19 is challenging on many grounds. Commonly used compartmental models struggle to handle an epidemiological process that evolves rapidly and is spatially heterogeneous. On the other hand, machine learning methods are limited at the beginning of the pandemics due to small data size for training. We propose a deep learning approach to predict future COVID-19 infection cases and deaths 1 to 4 weeks ahead at the fine granularity of US counties. The multi-variate Long Short-term Memory (LSTM) recurrent neural network is trained on multiple time series samples at the same time, including a mobility series. Results show that adding mobility as a variable and using multiple samples to train the network improve predictive performance both in terms of bias and of variance of the forecasts. We also show that the predicted results have similar accuracy and spatial patterns with a standard ensemble model used as benchmark. The model is attractive in many respects, including the fine geographic granularity of predictions and great predictive performance several weeks ahead. Furthermore, data requirement and computational intensity are reduced by substituting a single model to multiple models folded in an ensemble model.

scholarly journals CLPred: a sequence-based protein crystallization predictor using BLSTM neural network

2020 ◽  
Vol 36 (Supplement_2) ◽  
pp. i709-i717
Author(s):  
Wenjing Xuan ◽  
Ning Liu ◽  
Neng Huang ◽  
Yaohang Li ◽  
Jianxin Wang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Long Range ◽  
Short Term Memory ◽  
Protein Crystallization ◽  
Predictive Performance ◽  
Attrition Rate ◽  
Multiple Time ◽  
Interaction Patterns ◽  
Range Interaction

Abstract Motivation Determining the structures of proteins is a critical step to understand their biological functions. Crystallography-based X-ray diffraction technique is the main method for experimental protein structure determination. However, the underlying crystallization process, which needs multiple time-consuming and costly experimental steps, has a high attrition rate. To overcome this issue, a series of in silico methods have been developed with the primary aim of selecting the protein sequences that are promising to be crystallized. However, the predictive performance of the current methods is modest. Results We propose a deep learning model, so-called CLPred, which uses a bidirectional recurrent neural network with long short-term memory (BLSTM) to capture the long-range interaction patterns between k-mers amino acids to predict protein crystallizability. Using sequence only information, CLPred outperforms the existing deep-learning predictors and a vast majority of sequence-based diffraction-quality crystals predictors on three independent test sets. The results highlight the effectiveness of BLSTM in capturing non-local, long-range inter-peptide interaction patterns to distinguish proteins that can result in diffraction-quality crystals from those that cannot. CLPred has been steadily improved over the previous window-based neural networks, which is able to predict crystallization propensity with high accuracy. CLPred can also be improved significantly if it incorporates additional features from pre-extracted evolutional, structural and physicochemical characteristics. The correctness of CLPred predictions is further validated by the case studies of Sox transcription factor family member proteins and Zika virus non-structural proteins. Availability and implementation https://github.com/xuanwenjing/CLPred.

scholarly journals FetNet: a recurrent convolutional network for occlusion identification in fetoscopic videos

2020 ◽  
Vol 15 (5) ◽  
pp. 791-801 ◽  
Author(s):  
Sophia Bano ◽  
Francisco Vasconcelos ◽  
Emmanuel Vander Poorten ◽  
Tom Vercauteren ◽  
Sebastien Ourselin ◽  
...  
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Laser Photocoagulation ◽  
Short Term Memory ◽  
Total Duration ◽  
Frame Rate ◽  
Superior Performance ◽  
Computer Assisted ◽  
Convolutional Network ◽  
Spatio Temporal

Abstract Purpose Fetoscopic laser photocoagulation is a minimally invasive surgery for the treatment of twin-to-twin transfusion syndrome (TTTS). By using a lens/fibre-optic scope, inserted into the amniotic cavity, the abnormal placental vascular anastomoses are identified and ablated to regulate blood flow to both fetuses. Limited field-of-view, occlusions due to fetus presence and low visibility make it difficult to identify all vascular anastomoses. Automatic computer-assisted techniques may provide better understanding of the anatomical structure during surgery for risk-free laser photocoagulation and may facilitate in improving mosaics from fetoscopic videos. Methods We propose FetNet, a combined convolutional neural network (CNN) and long short-term memory (LSTM) recurrent neural network architecture for the spatio-temporal identification of fetoscopic events. We adapt an existing CNN architecture for spatial feature extraction and integrated it with the LSTM network for end-to-end spatio-temporal inference. We introduce differential learning rates during the model training to effectively utilising the pre-trained CNN weights. This may support computer-assisted interventions (CAI) during fetoscopic laser photocoagulation. Results We perform quantitative evaluation of our method using 7 in vivo fetoscopic videos captured from different human TTTS cases. The total duration of these videos was 5551 s (138,780 frames). To test the robustness of the proposed approach, we perform 7-fold cross-validation where each video is treated as a hold-out or test set and training is performed using the remaining videos. Conclusion FetNet achieved superior performance compared to the existing CNN-based methods and provided improved inference because of the spatio-temporal information modelling. Online testing of FetNet, using a Tesla V100-DGXS-32GB GPU, achieved a frame rate of 114 fps. These results show that our method could potentially provide a real-time solution for CAI and automating occlusion and photocoagulation identification during fetoscopic procedures.

MDTP

2021 ◽  
Vol 14 (8) ◽  
pp. 1289-1297
Author(s):  
Ziquan Fang ◽  
Lu Pan ◽  
Lu Chen ◽  
Yuntao Du ◽  
Yunjun Gao
Keyword(s):  
Neural Network ◽  
Short Term Memory ◽  
Temporal Dynamics ◽  
Real Life ◽  
Feature Modeling ◽  
Traffic Prediction ◽  
Interactive System ◽  
Trajectory Data ◽  
Spatio Temporal ◽  
Prediction Approach

Traffic prediction has drawn increasing attention for its ubiquitous real-life applications in traffic management, urban computing, public safety, and so on. Recently, the availability of massive trajectory data and the success of deep learning motivate a plethora of deep traffic prediction studies. However, the existing neural-network-based approaches tend to ignore the correlations between multiple types of moving objects located in the same spatio-temporal traffic area, which is suboptimal for traffic prediction analytics. In this paper, we propose a multi-source deep traffic prediction framework over spatio-temporal trajectory data, termed as MDTP. The framework includes two phases: spatio-temporal feature modeling and multi-source bridging. We present an enhanced graph convolutional network (GCN) model combined with long short-term memory network (LSTM) to capture the spatial dependencies and temporal dynamics of traffic in the feature modeling phase. In the multi-source bridging phase, we propose two methods, Sum and Concat, to connect the learned features from different trajectory data sources. Extensive experiments on two real-life datasets show that MDTP i) has superior efficiency, compared with classical time-series methods, machine learning methods, and state-of-the-art neural-network-based approaches; ii) offers a significant performance improvement over the single-source traffic prediction approach; and iii) performs traffic predictions in seconds even on tens of millions of trajectory data. we develop MDTP + , a user-friendly interactive system to demonstrate traffic prediction analysis.

A Neural Network for Modeling Multicategorical Parcel Use Change

2013 ◽  
pp. 1297-1308
Author(s):  
Kang Shou Lu ◽  
John Morgan ◽  
Jeffery Allen
Keyword(s):  
Neural Network ◽  
Land Use ◽  
Multinomial Logistic Regression ◽  
Slight Variation ◽  
Ann Model ◽  
Temporal Prediction ◽  
Percentage Points ◽  
Spatio Temporal ◽  
Coastal County ◽  
Artificial Neural Network Ann

This paper presents an artificial neural network (ANN) for modeling multicategorical land use changes. Compared to conventional statistical models and cellular automata models, ANNs have both the architecture appropriate for addressing complex problems and the power for spatio-temporal prediction. The model consists of two layers with multiple input and output units. Bayesian regularization was used for network training in order to select an optimal model that avoids over-fitting problem. When trained and applied to predict changes in parcel use in a coastal county from 1990 to 2008, the ANN model performed well as measured by high prediction accuracy (82.0-98.5%) and high Kappa coefficient (81.4-97.5%) with only slight variation across five different land use categories. ANN also outperformed the benchmark multinomial logistic regression by average 17.5 percentage points in categorical accuracy and by 9.2 percentage points in overall accuracy. The authors used the ANN model to predict future parcel use change from 2007 to 2030.

A Neural Network for Modeling Multicategorical Parcel Use Change

2012 ◽  
pp. 131-142
Author(s):  
Kang Shou Lu ◽  
John Morgan ◽  
Jeffery Allen
Keyword(s):  
Neural Network ◽  
Land Use ◽  
Multinomial Logistic Regression ◽  
Slight Variation ◽  
Ann Model ◽  
Temporal Prediction ◽  
Percentage Points ◽  
Spatio Temporal ◽  
Coastal County ◽  
Artificial Neural Network Ann

This paper presents an artificial neural network (ANN) for modeling multicategorical land use changes. Compared to conventional statistical models and cellular automata models, ANNs have both the architecture appropriate for addressing complex problems and the power for spatio-temporal prediction. The model consists of two layers with multiple input and output units. Bayesian regularization was used for network training in order to select an optimal model that avoids over-fitting problem. When trained and applied to predict changes in parcel use in a coastal county from 1990 to 2008, the ANN model performed well as measured by high prediction accuracy (82.0-98.5%) and high Kappa coefficient (81.4-97.5%) with only slight variation across five different land use categories. ANN also outperformed the benchmark multinomial logistic regression by average 17.5 percentage points in categorical accuracy and by 9.2 percentage points in overall accuracy. The authors used the ANN model to predict future parcel use change from 2007 to 2030.

A Neural Network for Modeling Multicategorical Parcel Use Change

2011 ◽  
Vol 2 (3) ◽  
pp. 20-31 ◽  
Author(s):  
Kang Shou Lu ◽  
John Morgan ◽  
Jeffery Allen
Keyword(s):  
Neural Network ◽  
Land Use ◽  
Multinomial Logistic Regression ◽  
Land Use Changes ◽  
Slight Variation ◽  
Ann Model ◽  
Temporal Prediction ◽  
Percentage Points ◽  
Spatio Temporal ◽  
Artificial Neural Network Ann

This paper presents an artificial neural network (ANN) for modeling multicategorical land use changes. Compared to conventional statistical models and cellular automata models, ANNs have both the architecture appropriate for addressing complex problems and the power for spatio-temporal prediction. The model consists of two layers with multiple input and output units. Bayesian regularization was used for network training in order to select an optimal model that avoids over-fitting problem. When trained and applied to predict changes in parcel use in a coastal county from 1990 to 2008, the ANN model performed well as measured by high prediction accuracy (82.0-98.5%) and high Kappa coefficient (81.4-97.5%) with only slight variation across five different land use categories. ANN also outperformed the benchmark multinomial logistic regression by average 17.5 percentage points in categorical accuracy and by 9.2 percentage points in overall accuracy. The authors used the ANN model to predict future parcel use change from 2007 to 2030.

scholarly journals Convolutional LSTM Architecture for Precipitation Nowcasting Using Satellite Data

2021 ◽  
Vol 8 (1) ◽  
pp. 33
Author(s):  
Carlos Javier Gamboa-Villafruela ◽  
José Carlos Fernández-Alvarez ◽  
Maykel Márquez-Mijares ◽  
Albenis Pérez-Alarcón ◽  
Alfo José Batista-Leyva
Keyword(s):  
Neural Network ◽  
Satellite Data ◽  
Training Data ◽  
Time Interval ◽  
Precipitation Data ◽  
Short Term ◽  
Data Set ◽  
Temporal Prediction ◽  
Spatio Temporal ◽  
Convolutional Lstm

The short-term prediction of precipitation is a difficult spatio-temporal task due to the non-uniform characterization of meteorological structures over time. Currently, neural networks such as convolutional LSTM have shown ability for the spatio-temporal prediction of complex problems. In this research, we propose an LSTM convolutional neural network (CNN-LSTM) architecture for immediate prediction of various short-term precipitation events using satellite data. The CNN-LSTM is trained with NASA Global Precipitation Measurement (GPM) precipitation data sets, each at 30-min intervals. The trained neural network model is used to predict the sixteenth precipitation data of the corresponding fifteen precipitation sequence and up to a time interval of 180 min. The results show that the increase in the number of layers, as well as in the amount of data in the training data set, improves the quality of the forecast.

scholarly journals An Incremental Class-Learning Approach with Acoustic Novelty Detection for Acoustic Event Recognition

Sensors ◽  
2021 ◽  
Vol 21 (19) ◽  
pp. 6622
Author(s):  
Barış Bayram ◽  
Gökhan İnce
Keyword(s):  
Neural Network ◽  
Acoustic Signal ◽  
Large Scale ◽  
Short Term Memory ◽  
Novelty Detection ◽  
Event Recognition ◽  
Acoustic Event ◽  
Audio Features ◽  
Spatio Temporal ◽  
Self Learning

Acoustic scene analysis (ASA) relies on the dynamic sensing and understanding of stationary and non-stationary sounds from various events, background noises and human actions with objects. However, the spatio-temporal nature of the sound signals may not be stationary, and novel events may exist that eventually deteriorate the performance of the analysis. In this study, a self-learning-based ASA for acoustic event recognition (AER) is presented to detect and incrementally learn novel acoustic events by tackling catastrophic forgetting. The proposed ASA framework comprises six elements: (1) raw acoustic signal pre-processing, (2) low-level and deep audio feature extraction, (3) acoustic novelty detection (AND), (4) acoustic signal augmentations, (5) incremental class-learning (ICL) (of the audio features of the novel events) and (6) AER. The self-learning on different types of audio features extracted from the acoustic signals of various events occurs without human supervision. For the extraction of deep audio representations, in addition to visual geometry group (VGG) and residual neural network (ResNet), time-delay neural network (TDNN) and TDNN based long short-term memory (TDNN–LSTM) networks are pre-trained using a large-scale audio dataset, Google AudioSet. The performances of ICL with AND using Mel-spectrograms, and deep features with TDNNs, VGG, and ResNet from the Mel-spectrograms are validated on benchmark audio datasets such as ESC-10, ESC-50, UrbanSound8K (US8K), and an audio dataset collected by the authors in a real domestic environment.

Fast Video Facial Expression Recognition by a Deeply Tensor-Compressed LSTM Neural Network for Mobile Devices

2021 ◽  
Vol 2 (4) ◽  
pp. 1-26
Author(s):  
Peining Zhen ◽  
Hai-Bao Chen ◽  
Yuan Cheng ◽  
Zhigang Ji ◽  
Bin Liu ◽  
...  
Keyword(s):  
Neural Network ◽  
Facial Expression ◽  
Mobile Devices ◽  
Mobile Device ◽  
Facial Expression Recognition ◽  
Short Term Memory ◽  
Expression Recognition ◽  
Feature Maps ◽  
Temporal Model ◽  
Spatio Temporal

Mobile devices usually suffer from limited computation and storage resources, which seriously hinders them from deep neural network applications. In this article, we introduce a deeply tensor-compressed long short-term memory (LSTM) neural network for fast video-based facial expression recognition on mobile devices. First, a spatio-temporal facial expression recognition LSTM model is built by extracting time-series feature maps from facial clips. The LSTM-based spatio-temporal model is further deeply compressed by means of quantization and tensorization for mobile device implementation. Based on datasets of Extended Cohn-Kanade (CK+), MMI, and Acted Facial Expression in Wild 7.0, experimental results show that the proposed method achieves 97.96%, 97.33%, and 55.60% classification accuracy and significantly compresses the size of network model up to 221× with reduced training time per epoch by 60%. Our work is further implemented on the RK3399Pro mobile device with a Neural Process Engine. The latency of the feature extractor and LSTM predictor can be reduced 30.20× and 6.62× , respectively, on board with the leveraged compression methods. Furthermore, the spatio-temporal model costs only 57.19 MB of DRAM and 5.67W of power when running on the board.

Sign in / Sign up

Export Citation Format

Share Document