Fault detection and classification with feature representation based on deep residual convolutional neural network

This paper proposes a novel deep learning (DL) based fast full-wave broad-band solver to realize electromagnetic forward (EMF) modelling for 2D dielectric target. This DL-based EMF solver is based on the complex-valued deep residual convolutional neural network (DRCNN), which employs the residual block and deep convolutional operation to improve its generality and performance. While the input of DRCNN combines the incident EM field at the arbitrary frequency within the extremely-broad frequency band with the contrasts distribution of the region-of-interest (ROI), the corresponding output is the total EM field illuminated by the input incident EM field. The training data is created based on only the simple synthetic dataset, while the incident EM fields are produced by the sources surrounding ROI. Compared with traditional approaches, EMF modelling has been realized with high accuracy and the greatly-reduced computation time. Unlike most of reported DL-based methods, the proposed method can work on the extremely-broad frequency band. Numerical examples based on class-specific 2D dielectric objects demonstrate the validity of this broad-band EMF solver, which acts as the potential candidate for modelling EMF process in real time.

Under the impact of global climate changes and human activities, harmful algae blooms (HABs) have become a growing concern due to negative impacts on water related industries, such as tourism, fishing and safe water supply. Many jurisdictions have introduced specific water quality regulations to protect public health and safety. Therefore reliable and cost effective methods of quantifying the type and concentration of algae cells has become critical for ensuring successful water management. In this work we present an innovative system to automatically classify multiple types of algae by combining standard morphological features with their multi-wavelength signals. To accomplish this we use a custom-designed microscopy imaging system which is configured to image water samples at two fluorescent wavelengths and seven absorption wavelengths using discrete-wavelength high-powered light emitting diodes (LEDs). We investigate the effectiveness of automatic classification using a deep residual convolutional neural network and achieve a classification accuracy of 96% in an experiment conducted with six different algae types. This high level of accuracy was achieved using a deep residual convolutional neural network that learns the optimal combination of spectral and morphological features. These findings illustrate the possibility of leveraging a unique fingerprint of algae cell (i.e. spectral wavelengths and morphological features) to automatically distinguish different algae types. Our work herein demonstrates that, when coupled with multi-band fluorescence microscopy, machine learning algorithms can potentially be used as a robust and cost-effective tool for identifying and enumerating algae cells.

For people in developing countries, cassava is a major source of calories and carbohydrates. However, Cassava Mosaic Disease (CMD) has become a major cause of concern among farmers in sub-Saharan Africa countries, which rely on cassava for both business and local consumption. The article proposes a novel deep residual convolution neural network (DRNN) for CMD detection in cassava leaf images. With the aid of distinct block processing, we can counterbalance the imbalanced image dataset of the cassava diseases and increase the number of images available for training and testing. Moreover, we adjust low contrast using Gamma correction and decorrelation stretching to enhance the color separation of an image with significant band-to-band correlation. Experimental results demonstrate that using a balanced dataset of images increases the accuracy of classification. The proposed DRNN model outperforms the plain convolutional neural network (PCNN) by a significant margin of 9.25% on the Cassava Disease Dataset from Kaggle.

Due to the favorable denoising performance of the discriminative learning model, a large number of discriminative learning models have been designed to remove noise. However, most discriminative learning models can only deal with the noise that already exists in the training data. In this paper, a multi-scale trainable deep residual convolutional neural network (DCMSNet) based on dilated convolution is proposed. DCMSNet consists of a chain of dilated convolution layers, convolution layers, normalized multi-scale convolution blocks (BNMCBlock), multi-scale convolution blocks (MCBlock) and multi-scale dropout convolution blocks (MCDBlock). The use of dilated convolution can avoid over-parameterization problems caused by too deep networks. In order to capture more feature information during feature extraction and image reconstruction, we designed novel BNMCBlock and MCBlock. In order to reduce the degree of coupling between image features and improve the generalization ability of the network, we also designed a MCDBlock. Meanwhile, residual learning, batch normalization and dropout are utilized to speed up the training process and boost the denoising performance. Unlike existing denoising models, DCMSNet is able to remove different degrees of noise. Compared with state-of-the-art image denoising methods, DCMSNet has achieved relatively competitive results.

Heartbeat classification using deep residual convolutional neural network from 2-lead electrocardiogram

DeepcomplexMRI: Exploiting deep residual network for fast parallel MR imaging with complex convolution

Contact-map of a protein sequence dictates the global topology of structural fold. Accurate prediction of the contact-map is thus essential to protein 3D structure prediction, which is particularly useful for the protein sequences that do not have close homology templates in the Protein Data Bank. We developed a new method, ResPRE, to predict residue-level protein contacts using inverse covariance matrix (or precision matrix) of multiple sequence alignments (MSAs) through deep residual convolutional neural network training. The approach was tested on a set of 158 non-homologous proteins collected from the CASP experiments and achieved an average accuracy of 50.6% in the top-L long-range contact prediction with L being the sequence length, which is 11.7% higher than the best of other state-of-the-art approaches ranging from coevolution coupling analysis to deep neural network training. Detailed data analyses show that the major advantage of ResPRE lies at the utilization of precision matrix that helps rule out transitional noises of contact-maps compared with the previously used covariance matrix. Meanwhile, the residual network with parallel shortcut layer connections increases the learning ability of deep neural network training. It was also found that appropriate collection of MSAs can further improve the accuracy of final contact-map predictions. The standalone package and online server of ResPRE are made freely available, which should bring important impact on protein structure and function modeling studies in particular for the distant- and non-homology protein targets. https://zhanglab.ccmb.med.umich.edu/ResPRE and https://github.com/leeyang/ResPRE. Supplementary data are available at Bioinformatics online.

Human activity recognition (HAR) is a highly prized application in the pattern recognition and the computer vision fields. Up till now, deep neural networks have acquired big attention in computer studies and image processing fields, and have generated significant results. In this paper, we propose a deep temporal residual system for daily living activity recognition that aims to enhance spatiotemporal feature representation in order to improve the HAR system performance. To this end, we adopt a deep residual convolutional neural network (RCN) to retain discriminative visual features relayed to appearance and long short-term memory neural network to capture the long-term temporal evolution of actions. The latter was considered to implement time dependencies occurring when carrying out the activity to enhance features extracted from the RCN network by adding time information to address the dynamic activity recognition problem as a sequence labeling job. The deep temporal residual model for human activity recognition system is performed on two benchmark publicly available datasets: MSRDailyActivity3D and CAD-60. the proposed system achieves very competitive results when compared to others from the state of the art.

Alzheimer’s disease (AD) is a neurodegenerative illness of unclear pathogenic origin that is characterized in its early stages by a steady deterioration in memory and cognitive impairment. Mild cognitive impairment (MCI) is a preclinical stage of AD. If a patient is diagnosed with early MCI, he or she can take preventive actions before permanent brain damage occurs. Improper diagnosis of MCI may cause the patient to miss the best time for treatment and incur heavy cost. Cost-sensitivity refers to the significant difference in cost to the patient between a misdiagnosis of MCI as normal control (NC) or AD and a misdiagnosis of NC or AD as MCI. In order to give consideration to both accuracy and cost, a computer-aided diagnosis algorithm based on cost-sensitive, attention mechanism and deep residual convolutional neural network (CSAResnet) was proposed for AD early diagnosis from MRI images. MRI data were obtained from the open-access AD Neuroimaging Initiative (ADNI) database. The experimental results show that the CSAResnet algorithm can balance the reduction of the total misclassification cost and the improvement of the accuracy to distinguish AD and MCI patients from NC subjects. It can address multiple classes of cost-sensitive problems.

A grid fault diagnosis framework based on adaptive integrated decomposition and cross-modal attention fusion

Idiopathic Parkinsons disease and atypical parkinsonian syndromes have similar symptoms at early disease stages, which makes the early differential diagnosis difficult. Positron emission tomography with 18F-FDG shows the ability to assess early neuronal dysfunction of neurodegenerative diseases and is well established for clinical use. In the past decades, machine learning methods have been widely used for the differential diagnosis of parkinsonism based on metabolic patterns. Unlike these conventional machine learning methods relying on hand-crafted features, the deep convolutional neural networks, which have achieved significant success in medical applications recently, have the advantage of learning salient feature representations automatically and effectively. This advantage may offer more appropriate invisible features extracted from data for the enhancement of the diagnosis accuracy. Therefore, this paper develops a 3D deep convolutional neural network on 18F-FDG PET images for the automated early diagnosis. Furthermore, we depicted in saliency maps the decision mechanism of the deep learning method to assist the physiological interpretation of deep learning performance. The proposed method was evaluated on a dataset with 920 patients. In addition to improving the accuracy in the differential diagnosis of parkinsonism compared to state-of-the-art approaches, the deep learning methods also discovered saliency features in a number of critical regions (e.g., midbrain), which are widely accepted as characteristic pathological regions for movement disorders but were ignored in the conventional analysis of FDG PET images.

AimsHeart failure (HF) is a leading cause of death. Early intervention is the key to reduce HF-related morbidity and mortality. This study assesses the utility of electrocardiograms (ECGs) in HF risk prediction.Methods and resultsData from the baseline visits (1987–89) of the Atherosclerosis Risk in Communities (ARIC) study was used. Incident hospitalized HF events were ascertained by ICD codes. Participants with good quality baseline ECGs were included. Participants with prevalent HF were excluded. ECG-artificial intelligence (AI) model to predict HF was created as a deep residual convolutional neural network (CNN) utilizing standard 12-lead ECG. The area under the receiver operating characteristic curve (AUC) was used to evaluate prediction models including (CNN), light gradient boosting machines (LGBM), and Cox proportional hazards regression. A total of 14 613 (45% male, 73% of white, mean age ± standard deviation of 54 ± 5) participants were eligible. A total of 803 (5.5%) participants developed HF within 10 years from baseline. Convolutional neural network utilizing solely ECG achieved an AUC of 0.756 (0.717–0.795) on the hold-out test data. ARIC and Framingham Heart Study (FHS) HF risk calculators yielded AUC of 0.802 (0.750–0.850) and 0.780 (0.740–0.830). The highest AUC of 0.818 (0.778–0.859) was obtained when ECG-AI model output, age, gender, race, body mass index, smoking status, prevalent coronary heart disease, diabetes mellitus, systolic blood pressure, and heart rate were used as predictors of HF within LGBM. The ECG-AI model output was the most important predictor of HF.ConclusionsECG-AI model based solely on information extracted from ECG independently predicts HF with accuracy comparable to existing FHS and ARIC risk calculators.

Confocal laser scanning microscopy (CLSM) is a non-destructive optical measurement system of high precision, applicable to the construction of three-dimensional (3-D) topographies of biological cells and engineered materials at the micro- and sub-micro scales. Compressive sensing (CS) has recently been applied in microscope systems to reduce the amount of sampled data required for the reconstruction of images; however, the iterative nature of the CS recovery algorithm imposes high computational complexity. This article presents an end-to-end non-iterative deep residual convolutional neural network (CNN) applicable to CLSM systems for CS-based reconstruction. In experiments and numerical simulations, the proposed scheme outperformed the existing CS recovery algorithms in terms of reconstructed image quality and computation time. The proposed algorithm also enabled the reconstruction of images using samples obtained in different regions of an image at various sampling rates to overcome nonuniform information density. The reconstruction performance of the model in terms of robustness and efficiency was validated using real-world CLSM data obtained via random scanning patterns.

With the development of the Internet, the captcha technology has also been widely used. Captcha technology is used to distinguish between humans and machines, namely Completely Automated Public Turing test to tell Computers and Humans Apart. In this paper, an end-to-end deep CNN-RNN network model is constructed by studying the captcha recognition technology, which realizes the recognition of 4-character text captcha. The CNN-RNN model first constructs a deep residual convolutional neural network based on the residual network structure to accurately extract the input captcha picture features. Then, through the constructed variant RNN network, that is, the two-layer GRU network, the deep internal features of the captcha are extracted, and finally, the output sequence is the 4-character captcha. The experiments results show that the end-to-end deep CNN-RNN network model has a good performance on different captcha datasets, achieving 99% accuracy. And experiment on the few samples dataset which only has 4000 training samples also shows an accuracy of 72.9 % and a certain generalization ability.

Numerical weather prediction has traditionally been based on the models that discretize the dynamical and physical equations of the atmosphere. Recently, however, the rise of deep learning has created increased interest in purely data‐driven medium‐range weather forecasting with first studies exploring the feasibility of such an approach. To accelerate progress in this area, the WeatherBench benchmark challenge was defined. Here, we train a deep residual convolutional neural network (Resnet) to predict geopotential, temperature and precipitation at 5.625° resolution up to 5 days ahead. To avoid overfitting and improve forecast skill, we pretrain the model using historical climate model output before fine‐tuning on reanalysis data. The resulting forecasts outperform previous submissions to WeatherBench and are comparable in skill to a physical baseline at similar resolution. We also analyze how the neural network creates its predictions and find that, for the case studies analyzed, the model has learned physically reasonable correlations. Finally, we perform scaling experiments to estimate the potential skill of data‐driven approaches at higher resolutions.