Search

Timely and accurate identification of fault types at the early stage of minor faults is significant for cutting off fault evolution. In order to have a clear understanding of the pipeline robot’s own situation in the pipeline, this paper proposes a fault diagnosis system for pipeline robots based on sound signal recognition. This can effectively reduce the probability of serious faults such as shutdown and loss of control in the pipeline without affecting the safe operation of the pipeline robot, which is a key issue to improve the reliability of the pipeline robot. The system consists of a combination of three parts: hardware, software, and algorithm. On the one hand, Raspberry Pi is the core module, while on the other hand, it is also responsible for the data transmission between the various modules, including storing the original sound signals collected by the sensors and transmitting the diagnosis results to the upper computer software interface. The proposed system is validated on the dataset collected by the data experimentation platform. The experimental results show that the proposed fault prediction method obtains advanced results on this dataset, verifying the effectiveness and stability of the proposed fault diagnosis system for pipeline robots based on sound signal recognition.

Big data has the traits such as “the curse of dimensionality,” high storage cost, and heavy computation burden. Self-representation-based feature extraction methods cannot effectively deal with the image-level structural noise in the data, so how to character a better relationship of reconstruction representation is very important. Recently, sparse representation with smoothed matrix multivariate elliptical distribution (SMED) using structural information to handle low-rank error images caused by illumination or occlusion has been proposed. Based on SMED, we present a new method named SMEDP for feature extraction. SMEDP firstly utilizes SMED to automatically construct an adjacency graph and then obtains an optimal projection matrix by maximizing the ratio of the local scatter matrix and the total scatter matrix in the PCA subspace. Experiments on the COIL-20 object database, ORL face database, and CMU PIE face database prove that SMEDP works well and can achieve considerable visual and recognition performance than the relevant methods.

BackgroundThe historical data of rare disease is very scarce in reality, so how to perform drug repositioning for the rare disease is a great challenge. Most existing methods of drug repositioning for the rare disease usually neglect father–son information, so it is extremely difficult to predict drugs for the rare disease.MethodIn this paper, we focus on father–son information mining for the rare disease. We propose GRU-Cooperation-Attention-Network (GCAN) to predict drugs for the rare disease. We construct two heterogeneous networks for information enhancement, one network contains the father-nodes of the rare disease and the other network contains the son-nodes information. To bridge two heterogeneous networks, we set a mapping to connect them. What’s more, we use the biased random walk mechanism to collect the information smoothly from two heterogeneous networks, and employ a cooperation attention mechanism to enhance repositioning ability of the network.ResultComparing with traditional methods, GCAN makes full use of father–son information. The experimental results on real drug data from hospitals show that GCAN outperforms state-of-the-art machine learning methods for drug repositioning.ConclusionThe performance of GCAN for drug repositioning is mainly limited by the insufficient scale and poor quality of the data. In future research work, we will focus on how to utilize more data such as drug molecule information and protein molecule information for the drug repositioning of the rare disease.

High-throughput CRISPR-Cas9 knockout screens using a tiling-sgRNA design permit in situ evaluation of protein domain function. Here, to facilitate de novo identification of essential protein domains from such screens, we propose ProTiler, a computational method for the robust mapping of CRISPR knockout hyper-sensitive (CKHS) regions, which refer to the protein regions associated with a strong sgRNA dropout effect in the screens. Applied to a published CRISPR tiling screen dataset, ProTiler identifies 175 CKHS regions in 83 proteins. Of these CKHS regions, more than 80% overlap with annotated Pfam domains, including all of the 15 known drug targets in the dataset. ProTiler also reveals unannotated essential domains, including the N-terminus of the SWI/SNF subunit SMARCB1, which is validated experimentally. Surprisingly, the CKHS regions are negatively correlated with phosphorylation and acetylation sites, suggesting that protein domains and post-translational modification sites have distinct sensitivities to CRISPR-Cas9 mediated amino acids loss., Tiling-sgRNA designs allow the in situ evaluation of protein domain functions. Here the authors present ProTiler - a computational method to predict CRISPR knockout hyper-sensitive regions, revealing previously unannotated domains.

As a new method to detect vital signs, Ultra-wideband (UWB) radar could continuously monitor human respiratory signs without contact. Aimed at addressing the problem of large interference and weak acquisition signal in radar echo signals from complex scenes, this paper adopts a UWB radar echo signal processing method that combines strong physical sign information extraction at P time and Variational Mode Decomposition (VMD) to carry out theoretical derivation. Using this novel processing scheme, respiration and heartbeat signals can be quickly reconstructed according to the selection of the appropriate intrinsic mode functions (IMFs), and the real-time detection accuracy of human respiratory signs is greatly improved. Based on an experimental platform, the data collected by the UWB radar module were first verified against the measured values obtained at the actual scene. The results of a validation test proved that our UWB radar echo signal processing method effectively eliminated the respiratory clutter signal and realized the accurate measurement of respiratory and heartbeat signals, which would prove the existence of life and further improve the quality of respiration and heartbeat signal and the robustness of detection.

Background The altered concentrations of amino acids were found in the bone marrow or blood of leukemia patients. Metabolomics technology combining mathematical model of biomarkers could be used for assisting the diagnosis of pediatric acute leukemia (AL). Methods The concentrations of 17 amino acids was measured by targeted liquid chromatograph–tandem mass spectrometry in periphery blood collected using dried blood spots. After evaluation, the mathematical models were further evaluated by prospective clinical validation cohort for AL diagnosis. Results The concentrations of 13 in 17 amino acids were statistically different between the periphery blood dried serum dots measured by targeted LC–MS/MS. The receiver operating characteristic analysis for the models of amino acid panel showed that the area under curve for AL diagnosis were 0.848, 0.834 and 0.856 by SVM, RF and XGBoost. The Kappa values in further prospectively evaluated clinical cohort were 0.697, 0.703 and 0.789 (p > 0.05) respectively, and the accuracies for the models were 84.86%, 85.20% and 89.46% respectively with further clinical validation. Conclusions The established mathematical model is a faster, cheaper and more convenient way than conventional methods, and no significant difference on the effect of diagnosis comparing with conventional methods. The mathematical model can be clinically useful for assisting pediatric AL diagnosis. Electronic supplementary material The online version of this article (10.1186/s12967-019-1783-9) contains supplementary material, which is available to authorized users.

Real-time monitoring of non-point source (NPS) pollution is challenging owing to the minute-scale change in runoff flow and concentration under rainfall condition. In this study, we proposed a real-time measurement method for total nitrogen (TN) by combining the timeliness of sensor detection and the accuracy of intelligent algorithms, based on the physical and chemical relationships between TN and sensor-measured indexes. Extra tree regression was selected as the TN inversion algorithm, which has high precision, high computational efficiency, and better ability in over-fitting control. The results show that: (1) the real-time inversion algorithm of TN can achieve the monitoring frequency at the minute scale (5 min); (2) the method performs well (R

To assess if a templated algorithm can improve the diagnostic performance of MRI for characterization of T2 isointense and hypointense renal masses.In this retrospective study, 60 renal masses with histopathologic diagnoses that were also confirmed as T2 iso- or hypointense on MRI were identified (mean ± standard deviation, range: 3.9 ± 2.5, 1.0-13.7 cm). Two semi-quantitative diagnostic algorithms were created based on MRI features of renal masses reported in the literature. Three body-MRI trained radiologists provided clinical diagnoses based on their experience and separately provided semiquantitative data for each components of the two algorithms. The algorithms were applied separately by a radiology trainee without additional interpretive input. Logistic regression was used to compare the accuracy of the three methods in distinguishing malignant versus benign lesions and in diagnosing the exact histopathology. Inter-reader agreement for each method was calculated using Fleiss' kappa statistics.The accuracy of the two algorithms and clinical experience were similar (70%, 69%, and 64%, respectively, p = 0.22-0.32), with fair to moderate inter-reader agreement (Fleiss's kappa: r = 0.375, r = 0.308, r = 0.375, respectively, all p 0.0001). The accuracy of the two algorithms and clinical experience in diagnosing specific histopathology were also no different from each other (34%, 29%, and 32%, respectively, p = 0.49-0.74), with fair to moderate inter-reader agreement (Fleiss's kappa: r = 0.20, r = 0.28, r = 0.375, respectively, all p 0.0001).Semi-quantitative templated algorithms based on MRI features of renal masses did not improve the ability to diagnose T2 iso- and hypointense renal masses when compared to unassisted interpretation by body MR trained subspecialists.

To design and develop artificial intelligence (AI) hydrocephalus (HYC) imaging diagnostic model using a transfer learning algorithm and evaluate its application in the diagnosis of HYC by non-contrast material-enhanced head computed tomographic (CT) images. A training and validation dataset of non-contrast material-enhanced head CT examinations that comprised of 1000 patients with HYC and 1000 normal people with no HYC accumulating to 28,500 images. Images were pre-processed, and the feature variables were labeled. The feature variables were extracted by the neural network for transfer learning. AI algorithm performance was tested on a separate dataset containing 250 examinations of HYC and 250 of normal. Resident, attending and consultant in the department of radiology were also tested with the test sets, their results were compared with the AI model. Final model performance for HYC showed 93.6% sensitivity (95% confidence interval: 77%, 97%) and 94.4% specificity (95% confidence interval: 79%, 98%), with area under the characteristic curve of 0.93. Accuracy rate of model, resident, attending, and consultant were 94.0%, 93.4%, 95.6%, and 97.0%. AI can effectively identify the characteristics of HYC from CT images of the brain and automatically analyze the images. In the future, AI can provide auxiliary diagnosis of image results and reduce the burden on junior doctors.

Motivation RNA secondary structure prediction is widely used to understand RNA function. Recently, there has been a shift away from the classical minimum free energy methods to partition function-based methods that account for folding ensembles and can therefore estimate structure and base pair probabilities. However, the classical partition function algorithm scales cubically with sequence length, and is therefore prohibitively slow for long sequences. This slowness is even more severe than cubic-time free energy minimization due to a substantially larger constant factor in runtime. Results Inspired by the success of our recent LinearFold algorithm that predicts the approximate minimum free energy structure in linear time, we design a similar linear-time heuristic algorithm, LinearPartition, to approximate the partition function and base-pairing probabilities, which is shown to be orders of magnitude faster than Vienna RNAfold and CONTRAfold (e.g. 2.5 days versus 1.3 min on a sequence with length 32 753 nt). More interestingly, the resulting base-pairing probabilities are even better correlated with the ground-truth structures. LinearPartition also leads to a small accuracy improvement when used for downstream structure prediction on families with the longest length sequences (16S and 23S rRNAs), as well as a substantial improvement on long-distance base pairs (500+ nt apart). Availability and implementation Code: http://github.com/LinearFold/LinearPartition; Server: http://linearfold.org/partition. Supplementary information Supplementary data are available at Bioinformatics online.

Background The number of patients with neurological diseases is currently increasing annually, which presents tremendous challenges for both patients and doctors. With the advent of advanced information technology, digital medical care is gradually changing the medical ecology. Numerous people are exploring new ways to receive a consultation, track their diseases, and receive rehabilitation training in more convenient and efficient ways. In this paper, we explore the use of facial expression recognition via artificial intelligence to diagnose a typical neurological system disease, Parkinson disease (PD). Objective This study proposes methods to diagnose PD through facial expression recognition. Methods We collected videos of facial expressions of people with PD and matched controls. We used relative coordinates and positional jitter to extract facial expression features (facial expression amplitude and shaking of small facial muscle groups) from the key points returned by Face++. Algorithms from traditional machine learning and advanced deep learning were utilized to diagnose PD. Results The experimental results showed our models can achieve outstanding facial expression recognition ability for PD diagnosis. Applying a long short-term model neural network to the positions of the key features, precision and F1 values of 86% and 75%, respectively, can be reached. Further, utilizing a support vector machine algorithm for the facial expression amplitude features and shaking of the small facial muscle groups, an F1 value of 99% can be achieved. Conclusions This study contributes to the digital diagnosis of PD based on facial expression recognition. The disease diagnosis model was validated through our experiment. The results can help doctors understand the real-time dynamics of the disease and even conduct remote diagnosis.

BACKGROUND In current approaches to vision screening in the community, a simple and efficient process is needed to identify individuals who should be referred to tertiary eye care centres for vision loss related to eye diseases. The emergence of deep learning technology offers new opportunities to revolutionise this clinical referral pathway. We aimed to assess the performance of a newly developed deep learning algorithm for detection of disease-related visual impairment. METHODS In this proof-of-concept study, using retinal fundus images from 15 175 eyes with complete data related to best-corrected visual acuity or pinhole visual acuity from the Singapore Epidemiology of Eye Diseases Study, we first developed a single-modality deep learning algorithm based on retinal photographs alone for detection of any disease-related visual impairment (defined as eyes from patients with major eye diseases and best-corrected visual acuity of

Multi-organ segmentation is a challenging task due to the label imbalance and structural differences between different organs. In this work, we propose an efficient cascaded V-Net model to improve the performance of multi-organ segmentation by establishing dense Block Level Skip Connections ( BLSC ) across cascaded V-Net. Our model can take full advantage of features from the first stage network and make the cascaded structure more efficient. We also combine stacked small and large kernels with an inception-like structure to help our model to learn more patterns, which produces superior results for multi-organ segmentation. In addition, some small organs are commonly occluded by large organs and have unclear boundaries with other surrounding tissues, which makes them hard to be segmented. We therefore first locate the small organs through a multi-class network and crop them randomly with the surrounding region, then segment them with a single-class network. We evaluated our model on SegTHOR 2019 challenge unseen testing set and Multi-Atlas Labeling Beyond the Cranial Vault challenge validation set. Our model has achieved an average dice score gain of 1.62 percents and 3.90 percents compared to traditional cascaded networks on these two datasets, respectively. For hard-to-segment small organs, such as the esophagus in SegTHOR 2019 challenge, our technique has achieved a gain of 5.63 percents on dice score, and four organs in Multi-Atlas Labeling Beyond the Cranial Vault challenge have achieved a gain of 5.27 percents on average dice score.

The varieties of cabbage seeds directly affect the yield and quality of cabbage, in order to rapidly and nondestructively identify the varieties of cabbage seeds, near infrared spectra technique were applied in this study and reflectance spectrum of the cabbage seeds was obtained. Firstly, to excavate the effective information in the spectral data and improve signal to noise ratio, the raw spectra was pre-processed with the method of standard normal variate (SNV) and multiplicative scatter correction (MSC). Secondly, principal component analysis (PCA) was used to analyze the clustering of cabbage samples, then the characteristic differentia of three cabbage varieties was obtained through qualitative analysis. Six Effective wavelengths were selected by successive projections algorithm (SPA). Finally, the full spectra variable, the first three principal components (PCs) using PCA and selected effective wavelengths using SPA were respectively set as inputs of the partial least squares discriminant analysis (PLS-DA) and least-squares support vector machine (LS-SVM) models for the classification of cabbage seeds. As can be seen from the two dimensional plot drawn with the scores of PC1 and PC2 (the first two principle components), PC1 and PC2 had a good clustering effect for different kinds of cabbage seeds. LS-SVM models performed better than PLS-DA models, the correct rates of discrimination were 100% achieved with LS-SVM models. PLS-DA and LS-SVM models built based on the selected wavelengths performed better than the models built based on the first three principal components, moreover, the SPA-LS-SVM model obtained the best results among all models, with 100% discrimination accuracy for both the calibration set and the prediction set. The overall results show that SPA can extract wavelengths, and the LS-SVM model combined with SPA can obtain optimal classification results. So the present paper could offer an alternate approach for the rapid discrimination of cabbage seeds variety.

Purpose: To remove blood vessel shadows from optical coherence tomography (OCT) images of the optic nerve head (ONH).\udMethods: Volume scans consisting of 97 horizontal B-scans were acquired through the center of the ONH using a commercial OCT device for both eyes of 13 subjects. A custom generative adversarial network (named DeshadowGAN) was designed and trained with 2328 B-scans in order to remove blood vessel shadows in unseen B-scans. Image quality was assessed qualitatively (for artifacts) and quantitatively using the intralayer contrast—a measure of shadow visibility ranging from 0 (shadow-free) to 1 (strong shadow). This was computed in the retinal nerve fiber layer (RNFL), the inner plexiform layer (IPL), the photoreceptor (PR) layer, and the retinal pigment epithelium (RPE) layer. The performance of DeshadowGAN was also compared with that of compensation, the standard for shadow removal.\udResults: DeshadowGAN decreased the intralayer contrast in all tissue layers. On average, the intralayer contrast decreased by 33.7 ± 6.81%, 28.8 ± 10.4%, 35.9 ± 13.0%, and 43.0 ± 19.5% for the RNFL, IPL, PR layer, and RPE layer, respectively, indicating successful shadow removal across all depths. Output images were also free from artifacts commonly observed with compensation.\udConclusions: DeshadowGAN significantly corrected blood vessel shadows in OCT images of the ONH. Our algorithm may be considered as a preprocessing step to improve the performance of a wide range of algorithms including those currently being used for OCT segmentation, denoising, and classification.\udTranslational Relevance: DeshadowGAN could be integrated to existing OCT devices to improve the diagnosis and prognosis of ocular pathologies.

In this paper, a time periodic and two–group reaction–diffusion epidemic model with distributed delay is proposed and investigated. We firstly introduce the basic reproduction number $R_0$ for the model via the next generation operator method. We then establish the threshold dynamics of the model in terms of $R_0$, that is, the disease is uniformly persistent if $R_0 \gt 1$, while the disease goes to extinction if $R_0 \lt 1$. Finally, we study the global dynamics for the model in a special case when all the coefficients are independent of spatio–temporal variables.

The biological age (BA) equation is a prediction model that utilizes an algorithm to combine various biological markers of ageing. Different from traditional concepts, the BA equation does not emphasize the importance of a golden index but focuses on using indices of vital organs to represent the senescence of whole body. This model has been used to assess the ageing process in a more precise way and may predict possible diseases better as compared with the chronological age (CA). The principal component analysis (PCA) is applied as one of the common and frequently used methods in the construction of the BA formula. Compared with other methods, PCA has its own study procedures and features. Herein we summarize the up-to-date knowledge about the BA formula construction and discuss the influential factors, so as to give an overview of BA estimate by PCA, including composition of samples, choices of test items, and selection of ageing biomarkers. We also discussed the advantages and disadvantages of PCA with reference to the construction mechanism, accuracy, and practicability of several common methods in the construction of the BA formula.

Visible near spectra tecnnology was adopted to detect soil total nitrogen content. 394 soil samples were collected from Wencheng, Zhejiang province to be used for calibration model (n=263) and independent prediction set (n=131). Raw spectra and wavelength-reduced spectra with five different pretreatment methods (SG smoothing, SNV, MSC, 1st-D and 2nd-D) were compared to determine the optimal wavelength range and pretreatment method for analysis. The results with 5 different pretreatment methods were not improved compared to that both of full spectra PLS model and wavelength reduction spectra model. Spectral variable selection is an important strategy in spectrum modeling analysis, because it tends to parsimonious data representation and can lead to multivariate models with better performance. In order to simply calibration models, the wavelength variables selected by two different variable selection methods (i. e. regression coefficient analysis (RCA) and successive projections algorithm (SPA) were proposed to be the inputs of calibration methods of PLS, MLR and LS-SVM models separately. These calibration models were also compared to select the best model to predict soil TN. In total, 9 different models were built ahd the best results indicated that PLS, MLR and LS-SVM obtained the highest precision with determination coefficient of prediction R2(pre) =0. 81, RMSEP=0. 0031 and RPD=2. 26 based on wavelength variables selected by RCA (0. 0002) and SPA as inputs of models. SPA-MLR model and other three models based on 7 sensitive variables selected by RC using 0. 0002 regression coefficient threshold value obtained the best result with R2(pre), RMSEP and RPD as 0. 81, 0. 0031 and 2. 26. This prediction accuracy is classied to be very good. For all the models, it could be concluded that RCA and SPA could be very useful ways to selected sensitive wavelengths, and the selected wavelengths were effective to estimate soil TN. It is recommended to adopt SPA variable selection or RCA variable selection method with both linear and nonlinear calibration models for measurement of the soil TN using Vis-NIR spectroscopy technology, and wavelengths selection could be very useful to reduce collinearity and redundancies of spectra.

In the present study, hyperspectral imaging combined with chemometrics was successfully proposed to identify different varieties of black bean. The varieties of black bean were defined based on the three different colors of the bean core. The hy-perspectral images in the spectral range of 380-1,030 nm of black bean were acquired using the developed hyperspectral imaging system, and the reflectance spectra were extracted from the region of interest (ROD) in the images. The average spectrum of a ROI of the sample in the images was used to represent the spectrum of the sample and build classification models. In total, 180 spectra of 180 samples were extracted. The wavelengths from 440 to 943 nm were used for analysis after the removal of the spec- tral region with absolute noises, and 440-943 nm spectra were preprocessed by multiplicative scatter correction (MSC). Five classification methods, including partial least squares discriminant analysis (PLS-DA), soft independent modeling of class analogy (SIMCA), K-nearest neighbor algorithm (KNN), support vector machine (SVM) and extreme learning machine (ELM), were used to build discriminant models using the preprocessed full spectra, the feature information extracted by principal component analysis (PCA) and the feature information extracted by wavelet transform (WT) from the preprocessed spectra, respectively. Among all the classification models using the preprocessed full spectra, ELM models obtained the best performance; among all the classification models using the feature information extracted from the preprocessed spectra by PCA, ELM model also obtained the best classification accuracy; and among all the classification models using the feature information extracted from the preprocessed spectra by WT, ELM models obtained the best classification performance with 100% accuracy in both the calibration set and the prediction set. Among all classification models, WT-ELM model obtained the best classification accuracy. The overall results indicated that it was feasible to identify black bean varieties nondestructively by using hyperspectral imaging, and WT could effectively extract feature information from spectra and ELM algorithm was effective to build high performance classification models.

Hyperspectral imaging technology was developed to identify different brand famous green tea based on PCA information and image information fusion. First 512 spectral images of six brands of famous green tea in the 380 approximately 1 023 nm wavelength range were collected and principal component analysis (PCA) was performed with the goal of selecting two characteristic bands (545 and 611 nm) that could potentially be used for classification system. Then, 12 gray level co-occurrence matrix (GLCM) features (i. e., mean, covariance, homogeneity, energy, contrast, correlation, entropy, inverse gap, contrast, difference from the second-order and autocorrelation) based on the statistical moment were extracted from each characteristic band image. Finally, integration of the 12 texture features and three PCA spectral characteristics for each green tea sample were extracted as the input of LS-SVM. Experimental results showed that discriminating rate was 100% in the prediction set. The receiver operating characteristic curve (ROC) assessment methods were used to evaluate the LS-SVM classification algorithm. Overall results sufficiently demonstrate that hyperspectral imaging technology can be used to perform classification of green tea.

Visible and short wave infrared spectroscopy (Vis/SW-NIRS) was investigated in the present study for measurement of soil organic matter (OM) and available potassium (K). Four types of pretreatments including smoothing, SNV, MSC and SG smoothing+first derivative were adopted to eliminate the system noises and external disturbances. Then partial least squares regression (PLSR) and least squares-support vector machine (LS-SVM) models were implemented for calibration models. The LS-SVM model was built by using characteristic wavelength based on successive projections algorithm (SPA). Simultaneously, the performance of LSSVM models was compared with PLSR models. The results indicated that LS-SVM models using characteristic wavelength as inputs based on SPA outperformed PLSR models. The optimal SPA-LS-SVM models were achieved, and the correlation coefficient (r), and RMSEP were 0. 860 2 and 2. 98 for OM and 0. 730 5 and 15. 78 for K, respectively. The results indicated that visible and short wave near infrared spectroscopy (Vis/SW-NIRS) (325 approximately 1 075 nm) combined with LS-SVM based on SPA could be utilized as a precision method for the determination of soil properties.