Stochastic Neighbor Embedding Algorithm Research Articles

Identification and Validation of a Hypoxia-Immune-Based Prognostic mRNA Signature for Oral Squamous Cell Carcinoma.

Background Oral squamous cell carcinoma (OSCC) is a commonly encountered head and neck malignancy. Increasing evidence shows that there are abnormal immune response and chronic cell hypoxia in the development of OSCC. However, there is a lack of a reliable hypoxia-immune-based gene signature that may serve to accurately prognosticate OSCC. Methods The mRNA expression data of OSCC patients were extracted from the TCGA and GEO databases. Hypoxia status was identified using the t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm. Both ESTIMATE and single-sample gene-set enrichment analysis (ssGSEA) were used for further evaluation of immune status. The DEGs in different hypoxia and immune status were determined, and univariate Cox regression was used to identify significantly prognostic genes. A machine learning method, least absolute shrinkage and selection operator (LASSO) Cox regression analysis, allowed us to construct prognostic gene signature to predict the overall survival (OS) of OSCC patients. Results A total of 773 DEGs were identified between hypoxia high and low groups. According to immune cell infiltration, patients were divided into immune high, medium, and low groups and immune-associated DEGs were identified. A total of 193 overlapped DEGs in both immune and hypoxia status were identified. With the univariate and LASSO Cox regression model, eight signature mRNAs (FAM122C, RNF157, RANBP17, SOWAHA, KIAA1211, RIPPLY2, INSL3, and DNAH1) were selected for further calculation of their respective risk scores. The risk score showed a significant association with age and perineural and lymphovascular invasion. In the GEO validation cohort, a better OS was observed in patients from the low-risk group in comparison with those in the high-risk group. High-risk patients also demonstrated different immune infiltration characteristics from the low-risk group and the low-risk group showed potentially better immunotherapy efficacy in contrast to high-risk ones. Conclusion The hypoxia-immune-based gene signature has prognostic potential in OSCC.

Longitudinal clustering analysis and prediction of Parkinson's disease progression using radiomics and hybrid machine learning.

We employed machine learning approaches to (I) determine distinct progression trajectories in Parkinson's disease (PD) (unsupervised clustering task), and (II) predict progression trajectories (supervised prediction task), from early (years 0 and 1) data, making use of clinical and imaging features. We studied PD-subjects derived from longitudinal datasets (years 0, 1, 2 & 4; Parkinson's Progressive Marker Initiative). We extracted and analyzed 981 features, including motor, non-motor, and radiomics features extracted for each region-of-interest (ROIs: left/right caudate and putamen) using our standardized standardized environment for radiomics analysis (SERA) radiomics software. Segmentation of ROIs on dopamine transposer - single photon emission computed tomography (DAT SPECT) images were performed via magnetic resonance images (MRI). After performing cross-sectional clustering on 885 subjects (original dataset) to identify disease subtypes, we identified optimal longitudinal trajectories using hybrid machine learning systems (HMLS), including principal component analysis (PCA) + K-Means algorithms (KMA) followed by Bayesian information criterion (BIC), Calinski-Harabatz criterion (CHC), and elbow criterion (EC). Subsequently, prediction of the identified trajectories from early year data was performed using multiple HMLSs including 16 Dimension Reduction Algorithms (DRA) and 10 classification algorithms. We identified 3 distinct progression trajectories. Hotelling's t squared test (HTST) showed that the identified trajectories were distinct. The trajectories included those with (I, II) disease escalation (2 trajectories, 27% and 38% of patients) and (III) stable disease (1 trajectory, 35% of patients). For trajectory prediction from early year data, HMLSs including the stochastic neighbor embedding algorithm (SNEA, as a DRA) as well as locally linear embedding algorithm (LLEA, as a DRA), linked with the new probabilistic neural network classifier (NPNNC, as a classifier), resulted in accuracies of 78.4% and 79.2% respectively, while other HMLSs such as SNEA + Lib_SVM (library for support vector machines) and t_SNE (t-distributed stochastic neighbor embedding) + NPNNC resulted in 76.5% and 76.1% respectively. This study moves beyond cross-sectional PD subtyping to clustering of longitudinal disease trajectories. We conclude that combining medical information with SPECT-based radiomics features, and optimal utilization of HMLSs, can identify distinct disease trajectories in PD patients, and enable effective prediction of disease trajectories from early year data.

Pediatric severe traumatic brain injury mortality prediction determined with machine learning-based modeling

IntroductionSevere traumatic brain injury (sTBI) is a leading cause of mortality in children. As clinical prognostication is important in guiding optimal care and decision making, our goal was to create a highly discriminative sTBI outcome prediction model for mortality. MethodsMachine learning and advanced analytics were applied to the patient admission variables obtained from a comprehensive pediatric sTBI database. Demographic and clinical data, head CT imaging abnormalities and blood biochemical data from 196 children and adolescents admitted to a tertiary pediatric intensive care unit (PICU) with sTBI were integrated using feature ranking by way of a forest of randomized decision trees, and a model was generated from a reduced number of admission variables with maximal ability to discriminate outcome. ResultsIn total, 36 admission variables were analyzed using feature ranking with variable weighting to determine their predictive importance for mortality following sTBI. Reduction analysis utilizing Borata feature selection resulted in a parsimonious six-variable model with a mortality classification accuracy of 82%. The final admission variables that predicted mortality were: partial thromboplastin time (22%); motor Glasgow Coma Scale (21%); serum glucose (16%); fixed pupil(s) (16%); platelet count (13%) and creatinine (12%). Using only these six admission variables, a t-distributed stochastic nearest neighbor embedding algorithm plot demonstrated visual separation of sTBI patients that lived or died, with high mortality predictive ability of this model on the validation dataset (AUC = 0.90) which was confirmed with a conventional area-under-the-curve statistical approach on the total dataset (AUC = 0.91; P < 0.001). ConclusionsMachine learning-based modeling identified the most clinically important prognostic factors resulting in a pragmatic, high performing prognostic tool for pediatric sTBI with excellent discriminative ability to predict mortality risk with 82% classification accuracy (AUC = 0.90). After external multicenter validation, our prognostic model might help to guide treatment decisions, aggressiveness of therapy and prepare family members and caregivers for timely end-of-life discussions and decision making. Level of evidenceIII; Prognostic.

Spatiotemporal and kinematic characteristics augmentation using Dual-GAN for ankle instability detection.

Obtaining massive amounts of training data is often crucial for computer-assisted diagnosis using deep learning. Unfortunately, patient data is often small due to varied constraints. We develop a new approach to extract significant features from a small clinical gait analysis dataset to improve computer-assisted diagnosis of Chronic Ankle Instability (CAI) patients. In this paper, we present an approach for augmenting spatiotemporal and kinematic characteristics using the Dual Generative Adversarial Networks (Dual-GAN) to train a series of modified Long Short-Term Memory (LSTM) detection models making the training process more data-efficient. Namely, we use LSTM-, LSTM-Fully Convolutional Networks (FCN)-, and Convolutional LSTM-based detection models to identify the patients with CAI. The Dual-GAN enables the synthesized data to approximate the real data distribution visualized by the t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm. Then we trained the proposed detection models using real data collected from a controlled laboratory study and mixed data from real and synthesized gait features. The detection models were tested in real data to validate the positive role in data augmentation as well as to demonstrate the capability and effectiveness of the modified LSTM algorithm for CAI detection using spatiotemporal and kinematic characteristics in walking. Dual-GAN generated efficient spatiotemporal and kinematic characteristics to augment the training set promoting the performance of CAI detection and the modified LSTM algorithm yielded an enhanced classification outcome to identify those CAI patients from a group of control subjects based on gait analysis data than any previous reports.

Galectin-9/TIM-3 as a Key Regulator of Immune Response in Gliomas With Chromosome 1p/19q Codeletion.

Gliomas with chromosome 1p/19q codeletion were considered a specific tumor entity. This study was designed to reveal the biological function alterations tightly associated with 1p/19q codeletion in gliomas. Clinicopathological and RNA sequencing data from glioma patients were obtained from The Cancer Genome Atlas and Chinese Glioma Genome Atlas databases. Gene set variation analysis was performed to explore the differences in biological functions between glioma subgroups stratified by 1p/19q codeletion status. The abundance of immune cells in each sample was detected using the CIBERSORT analytical tool. Single-cell sequencing data from public databases were analyzed using the t-distributed stochastic neighbor embedding (t-SNE) algorithm, and the findings were verified by in vitro and in vivo experiments and patient samples.We found that the activation of immune and inflammatory responses was tightly associated with 1p/19q codeletion in gliomas. As the most important transcriptional regulator of Galectin-9 in gliomas, the expression level of CCAAT enhancer-binding protein alpha in samples with 1p/19q codeletion was significantly decreased, which led to the downregulation of the immune checkpoints Galectin-9 and TIM-3. These results were validated in three independent datasets. The t-SNE analysis showed that the loss of chromosome 19q was the main reason for the promotion of the antitumor immune response. IHC protein staining, in vitro and in vivo experiments verified the results of bioinformatics analysis. In gliomas, 1p/19q codeletion can promote the antitumor immune response by downregulating the expression levels of the immune checkpoint TIM-3 and its ligand Galectin-9.

Potential PD-L1 expressing cytotoxic T-cell immunopathology in Alzheimer disease.

Growing evidence suggested that adaptive immunity involved Alzheimer's disease (AD). Peripheral blood mononuclear cell (PBMC) subpopulations and expression of immune checkpoints such as programmed death-ligand 1(PD-L1) and programmed cell death protein-1 (PD-1) could play roles in Alzheimer disease. The interaction of PD-L1/PD-1 lead PD-1 expressing T cells to attenuate T cell-activating signals. Recent studies had demonstrated the blockade of PD-L1/PD-1 axis modified brain pathology and restored cognitive performance in tauopathy murine model. However similar human experiment suggested controversial results in AD patients. To observe the changes of PBMC subpopulations and PD-L1/PD-1 expression in AD and healthy controls, we utilized flow cytometry, and applied t-distributed stochastic neighbor embedding (t-SNE) algorithm as a new approach to analyze results. AD patients (n=16) with different disease severity (CDR= 1 - 2) and healthy controls (n=16) were enrolled. Expression levels of PD-L1/PD-1 in AD patients and healthy controls were derived by flow cytometric analysis parameters. Moreover, a dimensionality reduction algorithm for visualized t-SNE, was implemented to analyze the PD-L1/PD-1 expressing cell clusters between AD patients and healthy controls. We found no significant PD-1 expression intensity difference between cytotoxic T lymphocytes (CTL) of AD patients and healthy controls. In AD patients, subpopulation of CTL expressing PD-L1 was elevated by 85.53%. Further, PD-L1 expressing CTL clusters between AD patients and healthy controls were compared. We observed that densities of PD-L1high CTL clusters are significantly higher than the healthy controls. Upregulation of PD-L1 and PD-L1 expressing CTL were increased in peripheral blood of AD patients, linking to newly found strong correlation between the level of PD-L1 positive CTL and markers which indicated negative immune climate. Abnormal adaptive immune response of CTL might associate with concurrent viral infection, intracellular bacterial infection, and underlying cancer in AD patients. Our results suggested a novel strategy to prevent AD progression with regulation of PD-L1 expressing CTL, and notation of underlying intracellular infection and malignancy for which could relate to dementia progression.

Modeling Car-Following Behavior on Freeways Considering Driving Style

To build more accurate and realistic freeway car-following models, driving characteristics specific to freeway car following should be considered. This study, therefore, analyzed three car-following models calibrated for different driving styles. A total of 5,900 freeway car-following events were extracted from 161,055 km of driving data collected in the Shanghai Naturalistic Driving Study (SH-NDS) database. Based on the fuzzy inference system built in this study, these car-following events were categorized as representing one of two styles: nonaggressive or aggressive. The two driving styles were visualized by using the t-distributed stochastic neighbor embedding (t-SNE) algorithm. The Gipps, Wiedemann, and intelligent driver model (IDM) car-following models were calibrated and validated for each driving style group. Using a genetic algorithm to analyze the calibrated parameters of the investigated car-following models, it was found that the model parameter values were related to driving style. When their performances were evaluated, results showed that the IDM performed best. The nonaggressive IDM and the aggressive IDM were used to simulate the car-following scenarios based on the same leading vehicle trajectories. The t-test and the F-test results showed that regarding both time gap and spacing gap, the differences of mean and variance are significant between aggressive and nonaggressive styles in nearly all of the simulated car-following scenarios. The mean spacing gap (nonaggressive: 38 m; aggressive: 30 m) and time gap (nonaggressive: 1.7 s; aggressive: 1.4 s) obtained from modeled car-following scenarios could be used directly in simulation software to show the characteristics of different driving styles.

Ratiometric 3D DNA Machine Combined with Machine Learning Algorithm for Ultrasensitive and High-Precision Screening of Early Urinary Diseases.

Urinary extracellular vesicles (uEVs) have received considerable attention as a potential biomarker source for the diagnosis of urinary diseases. Present studies mainly focus on the discovery of biomarkers based on high-throughput proteomics, while limited efforts have been paid to applying the uEVs' biomarkers for the diagnosis of early urinary disease. Herein, we demonstrate a diagnosis protocol to realize ultrasensitive detection of uEVs and accurate classification of early urinary diseases, by combing a ratiometric three-dimensional (3D) DNA machine with machine learning (ML). The ratiometric 3D DNA machine platform is constructed by conjugating a padlock probe (PLP) containing cytosine-rich (C-rich) sequences, anchor strands, and nucleic-acid-stabilized silver nanoclusters (DNAAgNCs) onto the magnetic nanoparticles (MNPs). The competitive binding of uEVs with the aptamer releases the walker strand, thus the ratiometric 3D DNA machine was activated to undergo an accurate amplification reaction and produce a ratiometric fluorescence signal. The present biosensor offers a detection limit of 9.9 × 103 particles mL-1 with a linear range of 104-108 particles mL-1 for uEVs. Two ML algorithms, K-nearest neighbor (KNN) and support vector machine (SVM), were subsequently applied for analyzing the correlation between the sensing signals of uEV multibiomarkers and the clinical state. The disease diagnostic accuracy of optimal biomarker combination reaches 95% and 100% by analyzing with KNN and SVM, and the disease type classification exhibits an accuracy of 94.7% and 89.5%, respectively. Moreover, the protocol results in 100% accurate visual identification of clinical uEV samples from individuals with disease or as healthy control by a t-distributed stochastic neighbor embedding (tSNE) algorithm.

INTELLIGENT HEALTH STATUS DETECTION METHOD FOR LOCOMOTIVE FUEL CELL BASED ON DATA-DRIVEN TECHNIQUES

Main drawbacks of fuel cell systems, namely, high cost, poor reliability, and short lifespan, limit the large-scale commercial application of fuel cell systems. The health status detection of fuel cell systems for locomotives is of great significance to the safe and stable operation of locomotives. To identify the failure modes of the fuel cell system accurately and quickly, this study proposed an intelligent health status detection method for locomotive fuel cells based on data-driven techniques. In this study, the actual test data of a 150-kW fuel cell system for locomotives was analyzed. The t-distributed stochastic neighbor embedding (t-SNE) algorithm was combined with the general regression neural network (GRNN) to intelligently detect the health status of the fuel cell system for locomotives. Specifically, t-SNE was used to process the high-dimensionality and strong coupling raw data of health status, enabling the dimensional reduction of the raw data to reflect essential features. Then, GRNN was used to identify the feature data to achieve the fast and accurate detection of the health status of the fuel cell system. Results show that the proposed method can effectively detect four health conditions, namely, normal state, high inlet coolant temperature, low air pressure, and low spray pump pressure, with a diagnostic accuracy of 98.75%. This study is applicable to the analysis of the actual measurement data of high-power level fuel cell systems and provides a reference for the health status detection of fuel cell systems for locomotives. Keywords: fuel cell system for locomotive; data-driven; general regression neural network; t-distributed stochastic neighbor embedding; health status detection

Recognizing blood pressure patterns in sedated critically ill patients on mechanical ventilation by spectral clustering.

BackgroundBlood pressure is a critical therapeutic goal in intensive care unit (ICU). One important factor influencing blood pressure are analgesia and sedation. Analgesic and sedative drugs are commonly used in critically ill patients. These drugs affect blood pressure by reducing the tension of the venous system, the cardiac preload, and cardiac output and inhibiting cardiac functions. Consequently, vasoactive agents are commonly used to increase blood pressure. The indications for the usage of vasoactive agents are unequivocal. However, opinions on when to stop raising blood pressure vary. This study explored the relationship between blood pressure and sedation.MethodsPatients in the Multiparameter Intelligent Monitoring in Intensive Care-III (MIMIC) database who had received mechanical ventilation, had been administered sedative analgesics during their ICU stay, and met the inclusion criteria were included in this study. The mean arterial pressure (MAP) tendency patterns were identified using spectral clustering and visualized using the t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm. The 28-day mortality rates of patients with different MAP patterns during their first 24 hours in the ICU and their sedation levels were calculated in the crosstab.ResultsFourteen thousand seven hundred and eighty-five patients from the MIMIC-III database were included in this study. Three MAP patterns were identified by spectral clustering. The median MAP of the low, moderate, and high MAP groups was 71.2, 80.4, and 97.6 mmHg, respectively. The 28-day mortality rate of patients in the moderate MAP group (13.0%) was lower than that of patients in the low (16.6%) and high (15.6%) MAP groups. No difference was found in the 28-day mortality rate between the low and high MAP groups. Dynamic changes in blood pressure at different sedation depths were also examined. Notably, compared with light and moderate sedation level, patients in the deep sedation group, especially those in the high MAP group (48.5%), had a higher 28-day mortality rate (36.5%).ConclusionsLow MAP in the first 24 hours in ICU indicates a high possibility of poor prognosis for critically ill patients on mechanical ventilation. For patients under deep sedation, maintaining a high mean arterial pressure also indicates poor prognosis. A personalized MAP target should be determined according to the severity of illness and level of sedation for each patient.

Discovery and construction of prognostic model for clear cell renal cell carcinoma based on single-cell and bulk transcriptome analysis.

BackgroundClear cell renal cell carcinoma (ccRCC) is the most common malignant kidney tumor in adults. Single-cell transcriptome sequencing can provide accurate gene expression data of individual cells. Integrated single-cell and bulk transcriptome data from ccRCC samples provide comprehensive information, which allows the discovery of new understandings of ccRCC and the construction of a novel prognostic model for ccRCC patients.MethodsSingle-cell transcriptome sequencing data was preprocessed by using the Seurat package in R software. Principal component analysis (PCA) and the t-distributed stochastic neighbor embedding (t-SNE) algorithm were used to perform cluster classification. Two subtypes of cancer cells were identified, pseudotime trajectory analysis and gene ontology (GO) analysis were conducted with the monocle and clusterProfiler packages. Two novel cancer cell biomarkers were identified according to the single-cell sequencing and were confirmed by The Cancer Genome Atlas (TCGA) data. T cell-related marker genes according to single-cell sequencing were screened by a combination of Kaplan-Meier (KM) analysis, univariate Cox analysis, least absolute shrinkage and selection operator (Lasso) regression and multivariate Cox analysis of TCGA data. Four survival predicting genes were screened out to develop a risk score model. A nomogram consisting of the risk score and clinical information was constructed to predict the prognosis for ccRCC patients.ResultsA total of 5,933 cells were included in the study after quality control. Fifteen cell clusters were classified by PCA and t-SNE algorithm. Two clusters of cancer cells with distinct differentiation status were identified. Besides, GO analysis revealed that biological processes were different between the two subgroups. Egl-9 family hypoxia-inducible factor 3 (EGLN3) and nucleolar protein 3 (NOL3) were specifically expressed in cancer cell clusters, bulk RNA sequencing data from TCGA confirmed their high expression in ccRCC tissues. GTSE1, CENPF, SMC2 and H2AFV were screened out and applied to the construction of risk score model. A nomogram was generated to predict prognosis of ccRCC by combing the risk score and clinical parameters.ConclusionsWe integrated single-cell and bulk transcriptome data from ccRCC in this study. Two subtypes of ccRCC cells with different biological characteristics and two potential biomarkers of ccRCC were discovered. A novel prognostic model was constructed for clinical application.

Approaching Optimal Nonlinear Dimensionality Reduction by a Spiking Neural Network

This work deals with the presentation of a spiking neural network as a means for efficiently solving the reduction of dimensionality of data in a nonlinear manner. The underneath neural model, which can be integrated as neuromorphic hardware, becomes suitable for intelligent processing in edge computing within Internet of Things systems. In this sense, to achieve a meaningful performance with a low complexity one-layer spiking neural network, the training phase uses the metaheuristic Artificial Bee Colony algorithm with an objective function from the principals in the machine learning science, namely, the modified Stochastic Neighbor Embedding algorithm. To demonstrate this fact, complex benchmark data were used and the results were compared with those generated by a reference network with continuous-sigmoid neurons. The goal of this work is to demonstrate via numerical experiments another method for training spiking neural networks, where the used optimizer comes from metaheuristics. Therefore, the key issue is defining the objective function, which can relate optimally the information at both sides of the spiking neural network. Certainly, machine learning techniques have advanced in defining efficient loss functions that can become suitable objective function candidates in the metaheuristic training phase. The practicality of these ideas is shown in this article. We use MSE values for evaluating the relative quality of the results and also co-ranking matrices.

Computational pathology reveals unique spatial patterns of immune response in H&E images from COVID-19 autopsies: preliminary findings.

Purpose: We used computerized image analysis and machine learning approaches to characterize spatial arrangement features of the immune response from digitized autopsied H&E tissue images of the lung in coronavirus disease 2019 (COVID-19) patients. Additionally, we applied our approach to tease out potential morphometric differences from autopsies of patients who succumbed to COVID-19 versus H1N1. Approach: H&E lung whole slide images from autopsy specimens of nine COVID-19 and two H1N1 patients were computationally interrogated. 606 image patches ( per patient) of were extracted from the 11 autopsied patient studies. A watershed-based segmentation approach in conjunction with a machine learning classifier was employed to identify two types of nuclei families: lymphocytes and non-lymphocytes (i.e., other nucleated cells such as pneumocytes, macrophages, and neutrophils). Based off the proximity of the individual nuclei, clusters for each nuclei family were constructed. For each of the resulting clusters, a series of quantitative measurements relating to architecture and density of nuclei clusters were calculated. A receiver operating characteristics-based feature selection method, violin plots, and the t-distributed stochastic neighbor embedding algorithm were employed to study differences in immune patterns. Results: In COVID-19, the immune response consistently showed multiple small-size lymphocyte clusters, suggesting that lymphocyte response is rather modest, possibly due to lymphocytopenia. In H1N1, we found larger lymphocyte clusters that were proximal to large clusters of non-lymphocytes, a possible reflection of increased prevalence of macrophages and other immune cells. Conclusion: Our study shows the potential of computational pathology to uncover immune response features that may not be obvious by routine histopathology visual inspection.

A soft-sensor model of VCM rectification concentration based on an improved WOA-RBFNN

Due to the complex nonlinear relationship among many variables in the rectification process of vinyl chloride monomer (VCM), there is a problem that its concentration is difficult to measure in real time. A method based on the tabu whale optimization algorithm for optimizing the radial basis function neural network (RBFNN) to model the concentration of the VCM rectification process is proposed. Firstly, the t-distributed stochastic neighbor embedding algorithm is used to compress high-dimensional data into low-dimensional space to obtain new data, the input of the soft-sensor model, and to maximize the retention of information about the input data to minimize the impact on the data redundancy model. Secondly, since the whale optimization algorithm (WOA) is prone to produce local optimality, a tabu search algorithm is introduced to help it jump out of the local optimality. Finally, the improved WOA is used to optimize the parameters of the RBFNN model, and the model is applied to the comparison experiment in the process of vinyl chloride rectification. According to the simulation results, the method improves the prediction accuracy of the model and has better practicability.

Identification of distribution network topology parameters based on multidimensional operation data

The connection relationship of distribution network topology is of great significance for the maintenance and fault diagnosis of distribution network, and scheduled power outage optimization. At present, the verification of topological documents mainly relies on on-site inspection, which consumes a lot of manpower and material resources and is inefficient. Therefore, an efficient method for topology verification of low-voltage substation areas is required. Given this background, a model for error correction and user access phase identification of low-voltage stations based on multi-dimensional voltage data collected by smart meters is presented in this paper, which can provide a certain reference for topology identification and line troubleshooting of low-voltage substations. First, the t-Distributed Stochastic Neighbor Embedding (t-SNE) algorithm and the Principal Component Analysis (PCA) performs dimensionality reduction on the original load data to solve the problem of redundancy caused by the high dimension of the original voltage data set. Second, the Local Outlier Factor (LOF) algorithm is used to identify abnormal samples in the voltage data set. Then, the spectral clustering method is used to cluster the dimensionality-reduced load data to realize the phase identification of single-phase users in the low-voltage station area. Finally, the real data of a certain area in Haining, Zhejiang Province of China are used as simulation cases for demonstrating. The results of the case studies show that the model proposed in this paper is feasible and effective.

RNA sequencing reveals the expression profiles of circRNA and identifies a four-circRNA signature acts as a prognostic marker in esophageal squamous cell carcinoma

BackgroundCircRNAs with tissue-specific expression and stable structure may be good tumor prognostic markers. However, the expression of circRNAs in esophageal squamous cell carcinoma (ESCC) remain unknown. We aim to identify prognostic circRNAs and construct a circRNA-related signature in ESCC.MethodsRNA sequencing was used to test the circRNA expression profiles of 73 paired ESCC tumor and normal tissues after RNase R enrichment. Bioinformatics methods, such as principal component analysis (PCA), t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm, unsupervised clustering and hierarchical clustering were performed to analyze the circRNA expression characteristics. Univariate cox regression analysis, random survival forests-variable hunting (RSFVH), Kaplan–Meier analysis, multivariable Cox regression and ROC (receiver operating characteristic) curve analysis were used to screen the prognostic circRNA signature. Real-time quantitative PCR (qPCR) and fluorescence in situ hybridization(FISH) in 125 ESCC tissues were performed.ResultsCompared with normal tissues, there were 11651 differentially expressed circRNAs in cancer tissues. A total of 1202 circRNAs associated with ESCC prognosis (P < 0.05) were identified. Through bioinformatics analysis, we screened a circRNA signature including four circRNAs (hsa_circ_0000005, hsa_circ_0007541, hsa_circ_0008199, hsa_circ_0077536) which can classify the ESCC patients into two groups with significantly different survival (log rank P < 0.001), and found its predictive performance was better than that of the TNM stage(0.84 vs. 0.66; 0.65 vs. 0.62). Through qPCR and FISH experiment, we validated the existence of the screened circRNAs and the predictive power of the circRNA signature.ConclusionThe prognostic four-circRNA signature could be a new prognostic biomarker for ESCC, which has high clinical application value.

Exhausted T cells in systemic lupus erythematosus patients in long-standing remission.

The mechanisms that drive systemic lupus erythematosus (SLE) patients to achieve remission are unknown; one possible explanation might be T cell exhaustion. The aim of the present study was to measure CD4+ and CD8+ T cell exhaustion in SLE patients in prolonged remission (PR-SLE) and compared them with patients with active SLE (Act-SLE) and healthy subjects. We included 15PR-SLE patients, 15Act-SLE and 29healthy subjects. T cell exhaustion was determined by flow cytometry according to the expression of programmed cell death 1 (PD)-1, T cell immunoglobulin and mucin 3 (Tim-3), natural killer cell receptor (2B4), eomesodermin (EOMES) and T-box transcription factor TBX21 (T-bet) in CD4+ and CD8+ T cells. Dimensionality reduction using the T-distributed stochastic neighbor-embedding algorithm and clustering analysis was used for the identification of relevant populations. Percentages of CD3+ , CD4+ and CD8+ T cells were similar among groups. We identified five subpopulations of CD8+ and seven of CD4+ cells. The CD4+ T-bet+ CD45RO+ cells identified in the unsupervised analysis were significantly increased in PR-SLE versus Act-SLE [median=0·20, interquartile range (IQR)=1·74-30·50 versus 1·68, IQR=0·4-2·83; P<0·01]. CD4+ EOMES+ cells were also increased in PR-SLE versus Act-SLE (5·24, IQR=3·38-14·70 versus 1·39, IQR=0·48-2·87; P<0·001). CD8+ EOMES+ cells were increased in PR-SLE versus Act-SLE (37·6, IQR=24·9-53·2 versus 8·13, IQR=2·33-20·5; P<0·001). Exhausted and activated T cells presented an increased frequency of PD-1, CD57 and EOMES in SLE patients versus healthy subjects. Some subpopulations of T cells expressing markers associated with exhaustion are increased in patients in remission, supporting T cell exhaustion as a tolerance mechanism in SLE. Exhaustion of specific populations of T cells might represent a potential therapeutic tool that will contribute to the goal of achieving sustained remission in these patients.

Social Touch Recognition Based on Support Vector Machine and T-Distributed Stochastic Neighbour Embedding as Pre-processing

Today, one of important field is social touch gesture recognition for touch modality, which can lead to highly efficient and realistic human–robot interaction. Touch is an important interaction modality in social interaction, for instance touch can communicate emotions and can intensify emotions communicated by other modalities. In this paper, the touch gesture recognition is performed using a dataset that previously measured for numerous subjects that perform various social gestures. This dataset is dubbed as the corpus of social touch (CoST), where touches were performed on a mannequin arm. The T-Distributed Stochastic Neighbor Embedding (T-SNE) algorithm is used to reduce the dimensions of the input data. The T-SNE algorithm was used as a preprocessing stage before classification operations. The output of the T-SNE is fed to the support vector machine (SVM). The performance of the proposed systems was evaluated using leave-one-subject-out cross-validation method. The range of recognition results 31.6% to 81.4%, Mean = 61.7% and Standard Deviation = 10.05%. The proposed method can recognize gestures in nearly real time after acquiring a minimum number of frames (629 ms). which is comparable with the results of Jung et al.

Video-based person-dependent and person-independent facial emotion recognition

Facial emotion recognition is a challenging problem that has attracted the attention of researchers in the last decade. In this paper, we present a system for facial emotion recognition in video sequences. Then, we evaluate the system for a person-dependent and person-independent cases. Depending on the purpose of the designed system, the importance of training a personalized model versus a non-personalized one differs. In this paper, first, we compute 60 geometric features for video frames of two datasets, namely RML and SAVEE databases. In the next step, k-means clustering is applied to the geometric features to select k most discriminant frames for each video clip. Then, we employ various classifiers like linear support vector machine (SVM) and Gaussian SVM to find the best representative k. Finally, five pre-trained convolutional neural networks, namely VGG-16, VGG-19, ResNet-50, AlexNet, and GoogleNet, were used evaluating two scenarios: person-dependent and person-independent emotion recognition. Additionally, the effect of geometric features in keyframe selection for a person-dependent and person-independent scenarios is studied based on different regions of the face. Also, the extracted features by CNNs are visualized using the t-distributed stochastic neighbor embedding algorithm to study the discriminative ability in these scenarios. Experiments show that person-dependent systems result in higher accuracy and suitable to be used in personalized systems.

Unsupervised Machine Learning-Based Clustering of Nanosized Fluorescent Extracellular Vesicles.

Extracellular vesicles (EV) are biological nanoparticles that play an important role in cell-to-cell communication. The phenotypic profile of EV populations is a promising reporter of disease, with direct clinical diagnostic relevance. Yet, robust methods for quantifying the biomarker content of EV have been critically lacking, and require a single-particle approach due to their inherent heterogeneous nature. Here, multicolor single-molecule burst analysis microscopy is used to detect multiple biomarkers present on single EV. The authors classify the recorded signals and apply the machine learning-based t-distributed stochastic neighbor embedding algorithm to cluster the resulting multidimensional data. As a proof of principle, the authors use the method to assess both the purity and the inflammatory status of EV, and compare cell culture and plasma-derived EV isolated via different purification methods. This methodology is then applied to identify intercellular adhesion molecule-1 specific EV subgroups released by inflamed endothelial cells, and to prove that apolipoprotein-a1 is an excellent marker to identify the typical lipoprotein contamination in plasma. This methodology can be widely applied on standard confocal microscopes, thereby allowing both standardized quality assessment of patient plasma EV preparations, and diagnostic profiling of multiple EV biomarkers in health and disease.