A Gamma-Ray Identification Algorithm Based on Fisher Linear Discriminant Analysis

In this work, we investigate the recognition of human emotional states from audiovisual signals. We extract prosodic, Mel-frequency Cepstral Coeffieient (MFCC), and formant frequency features to represent the audio characteristic of the emotional speech. A face detection scheme based on HSV color model is used to detect the face from the background. The facial expressions are represented by Gabor wavelet features. We perform feature selection by using the stepwise method based on Mahalanobis distance. The selected features are used to classify the emotional data into their corresponding classes. Different classification algorithms including Gaussian Mixture Model (GMM), K-nearest neighbours(K-NN), Neural Network (NN), and Fisher's Linear Discriminant Analysis (FLDA) are compared in this study. An adaptive multi-classifier scheme involving the analysis of individual class and combinations of different classes is proposed. Our recognition system is tested over a language independent database. The proposed FLDA-based multi-classifier scheme achieves the best overall and individual class recognition accuracy.

In this work, we investigate the recognition of human emotional states from audiovisual signals. We extract prosodic, Mel-frequency Cepstral Coeffieient (MFCC), and formant frequency features to represent the audio characteristic of the emotional speech. A face detection scheme based on HSV color model is used to detect the face from the background. The facial expressions are represented by Gabor wavelet features. We perform feature selection by using the stepwise method based on Mahalanobis distance. The selected features are used to classify the emotional data into their corresponding classes. Different classification algorithms including Gaussian Mixture Model (GMM), K-nearest neighbours(K-NN), Neural Network (NN), and Fisher's Linear Discriminant Analysis (FLDA) are compared in this study. An adaptive multi-classifier scheme involving the analysis of individual class and combinations of different classes is proposed. Our recognition system is tested over a language independent database. The proposed FLDA-based multi-classifier scheme achieves the best overall and individual class recognition accuracy.

A modified linear discriminant analysis technique for speaker verification, referred to as normalized discriminant analysis (NDA), is presented. Using this technique it is possible to design an efficient linear classifier with very limited training data and to generate normalized discriminant scores with comparable magnitudes for different classifiers. The NDA technique is applied to a classifier for speaker verification based on speaker specific information obtained when utterances are processed with speaker independent models. In experiments conducted on a network based telephone database, the NDA technique provides an equal-error rate of 6.13% while the classifier using Fisher linear discriminant analysis has an equal-error rate of 18.18%. Furthermore, when the NDA combined with HMM approach in a hybrid speaker verification system, the rate was reduced from 5.30% (HMM with cohort normalization) to 4.32%.

Nowadays, automatic tumor detection from brain images is extremely significant for many diagnostic as well as therapeutic purposes, due to the unpredictable shape and appearance of tumors. In medical image analysis, the automatic segmentation of tumors from brain using magnetic resonance imaging (MRI) data is the most critical issue. Existing research has some limitations, such as high processing time and lower accuracy, because of the time required for the training process. In this research, a new automatic segmentation process is introduced using machine learning and a swarm intelligence scheme. Here, a fuzzy logic with spiking neuron model (FL-SNM) is proposed for segmenting the brain tumor region in MR images. Initially, input images are preprocessed to remove Gaussian and Poisson noise using a modified Kuan filter (MKF). In the MKF, the optimal selection of the minimum MSE of image pixels is achieved using a random search algorithm (RSA), which improves the peak signal-to-noise ratio (PSNR). Then, the image is smoothed using an anisotropic diffusion filter (ADF) to reduce the over-filtering problem. Afterwards, to extract statistical texture features, Fisher's linear-discriminant analysis (FLDA) is used. Finally, extracted features are transferred to the FL-SNM process and this scheme effectively segments the tumor region. In FL-SNM, the consequent parameters such as weight and bias play an important role in segmenting the region. Therefore, optimizing the weight parameter values using a chicken behavior-based swarm intelligence (CSI) algorithm, is proposed. The proposed (FL-SNM) scheme attained better performance in terms of high accuracy (94.87%), sensitivity (92.07%), specificity (99.34%), precision rate (89.36%), recall rate (88.39%), F-measure (95.06%), G-mean (95.63%), and DSC rate (91.2%), compared to existing convolutional neural networks (CNNs) and hierarchical self-organizing maps (HSOMs).

Standard classification algorithms are generally designed to maximize the number of correct predictions (concordance). The criterion of maximizing the concordance may not be appropriate in certain applications. In practice, some applications may emphasize high sensitivity (e.g., clinical diagnostic tests) and others may emphasize high specificity (e.g., epidemiology screening studies). This paper considers effects of the decision threshold on sensitivity, specificity, and concordance for four classification methods: logistic regression, classification tree, Fisher's linear discriminant analysis, and a weighted k-nearest neighbor. We investigated the use of decision threshold adjustment to improve performance of either sensitivity or specificity of a classifier under specific conditions. We conducted a Monte Carlo simulation showing that as the decision threshold increases, the sensitivity decreases and the specificity increases; but, the concordance values in an interval around the maximum concordance are similar. For specified sensitivity and specificity levels, an optimal decision threshold might be determined in an interval around the maximum concordance that meets the specified requirement. Three example data sets were analyzed for illustrations.

<p indent="0mm">2019 novel coronaviruses (2019-nCoV) swept the world and are still at a high incidence stage. Highly sensitive and accurate nucleic acid molecular diagnostic techniques and methods have become a hot spot of global concern. 2019-nCoV is a kind of RNA virus. At present, there are many methods to detect virus RNA, including real-time fluorescence quantitative PCR (RT-PCR), colloidal gold detection, chemiluminescence detection, and so on. Colloidal gold detection and chemiluminescence detection need to use monoclonal antibodies against 2019-nCoV, but there are some problems in the preparation of 2019-nCoV monoclonal antibodies, such as long cycle, complex preparation and easy pollution. So the above two methods are not easy to be established for the detection of 2019-nCoV. The superiority of PCR technology is widely used in auxiliary medical clinical diagnosis, customs inspection and quarantine, the development of new agriculture, national defense military defense, and basic research of bioscience, and so on. However, the quantification of the target DNA fragment in RT-PCR is still relatively quantitative. Because this technique mainly depends on CT value and standard curve to quantify the target DNA in the experiment, which affects the amplification efficiency of the whole system, and the sensitivity and accuracy of the reaction system are greatly affected. Because the 2019-nCoV virus is highly contagious, the detection methodology is required to be highly sensitive, which is in line with the characteristics of digital PCR (droplet digital PCR system, referred to as ddPCR). ddPCR has the advantages of absolute quantification, high specificity, high sensitivity and strong interference ability, which can better combine molecular biology and medicine closely, and has a great advantage in the detection of 2019-nCoV. In this study, based on droplet digital PCR detection technology, the detection reagent and method were established using the principle of nested PCR. The Saliva of 333 2019-nCoV patients and their environmental samples were detected and compared with real-time PCR (RT-PCR). The linear range of the droplet digital PCR method established in this study was between 25–5 copies/µL, and minimum detection limit was 0.5 copies/µL, which was 2–3 orders of magnitude higher than that of the commercial real-time fluorescent quantitative PCR detection technology. In 333 samples, 7 were positive in 197 environmental samples, and 9 were positive in 136 human samples. The detection accuracy was 100%, and the environmental micro-virus residues that could not be detected by RT-PCR were detected. The digital PCR detection method established in this study was highly specific, and the minimum detection limit was far lower than that of RT-PCR. It is of great significance for sensitive and accurate detection of 2019-nCoV infection, early infection and environmental microvirus.

Structure–Activity Studies on Sulfamate Sweetners III: Structure–Taste Relationships for Heterosulfamates

The classificatory power afforded by the conjoint measurement of lever-pressing response rate and response duration was compared with a univariate analysis based exclusively on response rate. Operant behavior was measured following chlorpromazine, haloperidol, chlordiazepoxide, and pentobarbital, all of which produce similar dose-related decreases in rate of responding on simple fixed-ratio schedules of reinforcement. The multivariate technique of linear discriminant analysis indicates that both response rate and response duration provide useful and nonredundant information regarding the drug effect, separating the four drugs into three distinct classes: neuroleptic, chlordiazepoxide, and pentobarbital.

In this paper a new approach to localization of mobile nodes in wireless sensor networks is presented which utilizes the measurement of the received signal strength indicator by multiple receivers concentrated in a small area and applies the technique of linear discriminant analysis to the received data. The concept of gathering all receivers at a single spot provides great flexibility, while good spatial resolution is still achieved. This is demonstrated by a test network operating in the sub-GHz-range around 868 MHz.

A specific and sensitive radioimmunoassay for human choriogonadotropin (hCG) has been developed using rabbit antiserum to chemical analogs of beta subunit of human chorionic gonadotropin prepared by controlled reduction and S-alkylation of its disulfide linkages. The assay was highly specific for hCG as the binding of [125I]-hCG to the antibody was not affected by standard human lutropin, by human male serum, postpartum serum from women, serum from post-menopausal women and human menopausal gonadotropin (Pergonal). The assay was highly sensitive, the minimal detection limit in terms of highly purified hCG (L-129, 12500 IU/mg) being 1 ng/ml or 0.2 ng/tube (or 12 mlU/ml in terms of WHO 2nd international reference preparation of hCG). Using this assay we were unable to detect any immunoreactive hCG in human tissues like lung, liver and colon. The high specificity, sensitivity, accuracy and reproducibility of the assay make this a highly desirable radioimmunoassay for human choriogonadotropin.

Due to its high sensitivity to weak forces, the classical torsion pendulum was widely employed in the fields of precision gravitational measurement, which can be evaluated by testing of Newtonian inverse square law (ISL). In order to get much higher precision for the tests of ISL at short ranges, such as at millimeter and/or sub-millimeter level, the torsion pendulum cannot work precisely in free oscillation motion due to the electrostatic disturbance. This paper presents a method using proportional-integral-differential (PID) technology to make pendulum to keep quasi-static. A closed-loop system with PID controller is realized and calibration results show the system has the potential to get the precision of test of the ISL violation at the order of 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−16</sup> N·m at millimeter range.

Coefficients of variation for spine lengths are not constant over a range of mean values, but show an inverse relationship with sample means. The nearly hyperbolic relationship is evident in selection experiments, in seasonal samples from a single lake, and in widely dispersed geographic samples. The observed inverse relationship could indicate an interesting developmental property (e.g. that smaller, rudimentary spines are more variable than larger functional ones) or could reflect one of several possible statistical artifacts or properties.Detailed examination of Bosmina suggests that the strong inverse trend arises from a combination of biological and statistical properties; it is not an artifact of measurement error, although measurement error contributes to inflation of small values. Small spines are inherently more variable than longer spines, and the latter have more uniform component parts. Moreover, there are several alternative arrangements of components that sum to the same length and hence that achieve nearly equivalent fitness value. These alternative arrangements lower part‐whole correlations within the population.The two statistical properties that contribute to observed trends include a version of the central limit theorem and measurement error. The percentage variation of a character which is the sum of a number of randomly varying parts will be considerably less (roughly as the inverse square root of the number of parts) than the mean percentage variation of the parts themselves. Thus a statistical property of large numbers contributes to the observed inverse trend. In addition, coefficients of variation as fractions are especially sensitive to measurement error, so that precise measurements are recommended for small defensive spines.

Screen-printed Ga2O3 thin film derived from liquid metal employed in highly sensitive pH and non-enzymatic glucose recognition

Introduction. Vascular access recirculation (AR), which is often unacknowledged, remains an important cause of inadequate dialytic dose. The glucose infusion test (GIT) is a new method for detecting and quantifying AR. This paper reports on a polycentric evaluation of the new test and a comparison with the classical Urea-test (UT). Methods. GIT protocol comprises withdrawal from the arterial port (sample A), injection into the venous drip chamber of 1 g glucose in 4 seconds, withdrawal from the arterial port (sample B) continuously from 13 to 17 seconds. Glucose is determined on A and B by a reflectance photometer. If B = A then there is no recirculation. If B exceeds A by at least 20 mg/dl there is recirculation. AR quantification: AR% = (B-A) / 20. GIT was performed on 623 patients from eleven dialysis centers to screen the patients for AR. Subsequently, GIT and Urea-test (UT) were compared in 189 paired tests. The reproducibility of GIT and UT was studied in 28 paired tests performed in sequence. Results. The screening test by GIT was positive in 68 cases (11 %). The majority of positivities was found in central venous catheters (CVC, 27/50 cases, 54 %), whereas only 7 % of artero-venous fistulas (AVF) were positive. In the CVC group, Tesio catheters were more frequently positive compared to Dual Lumen Catheters (64 % vs. 29 %). The comparison GIT - UT showed that results matched in 162 tests (79 negative and 83 positive both by GIT and UT), showing that on the grounds of UT, GIT has high sensitivity and specificity. In 27 tests GIT was positive, but UT negative. This disagreement is due to the different minimal limit of detection, 1 % for GIT and 5% for UT. The reproducibility was greater with GIT than with UT with a lower D% (respectively -0.6 +/- 2.5 and -0.4 +/- 6.1 %, p<0.001) and a lower coefficient of variation (17 vs 33 %). Conclusions. The screening of 623 patients by GIT confirmed that AR in AVF is normally absent, whereas an un-expectedly high frequency of moderate AR in CVC was found. The GIT-UT comparison showed that the new test is simple and immediate, and gives results with higher accuracy, sensitivity and reproducibility than UT.

We briefly review the current trend in handling the neonatal baby weight, gestational age, and other related categorical and continuous variable data, through both robust and outlier-based methods. Graphs are generally presented as smooth curves which give a wrong visual impression and totally hide the actual variability of the data, robust methods are applied without considering that they may certainly be affected by asymmetrical contamination, and the outlier-based methods are incorrectly applied without ascertaining if the individual data arrays are normally distributed, free from discordant observations. Aim: To demonstrate our new procedure for handling of neonatal baby data from two case studies. Methodology: The univariate data were handled through the application of discordancy tests in the light of new precise and accurate critical values, and significance tests were applied to both original and discordant outlier-free data arrays. The bivariate data were also handled from identifying and separating discordant outliers and reporting the quality of the coefficients in terms of standard errors. Finally, the supervised multivariate technique of linear discriminant and canonical analysis is successfully applied. Results: The results of significance tests for some cases varied from the use of all data versus censored data. The identification and separation of bivariate discordant outliers drastically improved the quality of regression lines. Linear discriminant and canonical analysis provided high success values for laboratory-controlled experiments. Conclusion: The need of obtaining both central tendency and dispersion parameters in all medical data is established. The reporting of regression coefficients with individual errors is recommended.