The odor fingerprint of Pollygonati Rhizoma samples with different mildewing degrees was analyzed and the relationship between the odor variation and the mildewing degree was explored. A fast discriminant model was established according to the response intensity of electronic nose. The α-FOX3000 electronic nose was applied to analyze the odor fingerprint of Pollygonati Rhizoma samples with different mildewing degrees and the radar map was used to analyze the main contributors among the volatile organic compounds. The feature data were processed and analyzed by partial least squares discriminant analysis(PLS-DA), K-nearest neighbor(KNN), sequential minimal optimization(SMO), random forest(RF) and naive Bayes(NB), respectively. According to the radar map of the electronic nose, the response values of three sensors, namely T70/2, T30/1, and P10/2, increased with the mildewing, indicating that the Pollygonati Rhizoma produced alkanes and aromatic compounds after the mildewing. According to PLS-DA model, Pollygonati Rhizoma samples of three mildewing degrees could be well distinguished in three areas. Afterwards, the variable importance analysis of the sensors was carried out and then five sensors that contributed a lot to the classification were screened out: T70/2, T30/1, PA/2, P10/1 and P40/1. The classification accuracy of all the four models(KNN, SMO, RF, and NB) was above 90%, and KNN was most accurate(accuracy: 97.2%). Different volatile organic compounds were produced after the mildewing of Pollygonati Rhizoma, and they could be detected by electronic nose, which laid a foundation for the establishment of a rapid discrimination model for mildewed Pollygonati Rhizoma. This paper shed lights on further research on change pattern and quick detection of volatile organic compounds in moldy Chinese herbal medicines.
Electronic Nose ; Odorants/analysis* ; Volatile Organic Compounds/analysis* ; Bayes Theorem ; Drugs, Chinese Herbal/analysis* ; Discriminant Analysis

This study explored the feasibility of mineral element content and ratios of nitrogen isotopes to discriminate the cultivation mode of Dendrobium nobile in order to provide theoretical support for the discrimination of the cultivation mode of D. nobile. The content of 11 mineral elements(N, K, Ca, P, Mg, Na, Fe, Cu, Zn, Mn, and B) and nitrogen isotope ratios in D. nobile and its substrate samples in three cultivation methods(greenhouse cultivation, tree-attached cultivation, and stone-attached cultivation) were determined. According to the analysis of variance, principal component analysis, and stepwise discriminant analysis, the samples of different cultivation types were classified. The results showed that the nitrogen isotope ratios and the content of elements except for Zn were significantly different among different cultivation types of D. nobile(P<0.05). The results of correlation analysis showed that the nitrogen isotope ratios, mineral element content, and effective component content in D. nobile were correlated with the nitrogen isotope ratio and mineral element content in the corresponding substrate samples to varying degrees. Principal component analysis can preliminarily classify the samples of D. nobile, but some samples overlapped. Through stepwise discriminant analysis, six indicators, including δ~(15)N, K, Cu, P, Na, and Ca, were screened out, which could be used to establish the discriminant model of D. nobile cultivation methods, and the overall correct discrimination rates after back-substitution test, cross-check, and external validation were all 100%. Therefore, nitrogen isotope ratios and mineral element fingerprints combined with multivariate statistical analysis could effectively discriminate the cultivation types of D. nobile. The results of this study provide a new method for the identification of the cultivation type and production area of D. nobile and an experimental basis for the quality evaluation and quality control of D. nobile.
Dendrobium ; Minerals ; Discriminant Analysis ; Multivariate Analysis ; Nitrogen Isotopes

Coding with high-frequency stimuli could alleviate the visual fatigue of users generated by the brain-computer interface (BCI) based on steady-state visual evoked potential (SSVEP). It would improve the comfort and safety of the system and has promising applications. However, most of the current advanced SSVEP decoding algorithms were compared and verified on low-frequency SSVEP datasets, and their recognition performance on high-frequency SSVEPs was still unknown. To address the aforementioned issue, electroencephalogram (EEG) data from 20 subjects were collected utilizing a high-frequency SSVEP paradigm. Then, the state-of-the-art SSVEP algorithms were compared, including 2 canonical correlation analysis algorithms, 3 task-related component analysis algorithms, and 1 task discriminant component analysis algorithm. The results indicated that they all could effectively decode high-frequency SSVEPs. Besides, there were differences in the classification performance and algorithms' speed under different conditions. This paper provides a basis for the selection of algorithms for high-frequency SSVEP-BCI, demonstrating its potential utility in developing user-friendly BCI.
Humans ; Brain-Computer Interfaces ; Evoked Potentials, Visual ; Algorithms ; Discriminant Analysis ; Electroencephalography

Error self-detection based on error-related potentials (ErrP) is promising to improve the practicability of brain-computer interface systems. But the single trial recognition of ErrP is still a challenge that hinters the development of this technology. To assess the performance of different algorithms on decoding ErrP, this paper test four kinds of linear discriminant analysis algorithms, two kinds of support vector machines, logistic regression, and discriminative canonical pattern matching (DCPM) on two open accessed datasets. All algorithms were evaluated by their classification accuracies and their generalization ability on different sizes of training sets. The study results show that DCPM has the best performance. This study shows a comprehensive comparison of different algorithms on ErrP classification, which could give guidance for the selection of ErrP algorithm.
Algorithms ; Brain ; Brain-Computer Interfaces ; Discriminant Analysis ; Electroencephalography ; Support Vector Machine

Vision is an important way for human beings to interact with the outside world and obtain information. In order to research human visual behavior under different conditions, this paper uses a Gaussian mixture-hidden Markov model (GMM-HMM) to model the scanpath, and proposes a new model optimization method, time-shifting segmentation (TSS). The TSS method can highlight the characteristics of the time dimension in the scanpath, improve the pattern recognition results, and enhance the stability of the model. In this paper, a linear discriminant analysis (LDA) method is used for multi-dimensional feature pattern recognition to evaluates the rationality and the accuracy of the proposed model. Four sets of comparative trials were carried out for the model evaluation. The first group applied the GMM-HMM to model the scanpath, and the average accuracy of the classification could reach 0.507, which is greater than the opportunity probability of three classification (0.333). The second set of trial applied TSS method, and the mean accuracy of classification was raised to 0.610. The third group combined GMM-HMM with TSS method, and the mean accuracy of classification reached 0.602, which was more stable than the second model. Finally, comparing the model analysis results with the saccade amplitude (SA) characteristics analysis results, the modeling analysis method is much better than the basic information analysis method. Via analyzing the characteristics of three types of tasks, the results show that the free viewing task have higher specificity value and a higher sensitivity to the cued object search task. In summary, the application of GMM-HMM model has a good performance in scanpath pattern recognition, and the introduction of TSS method can enhance the difference of scanpath characteristics. Especially for the recognition of the scanpath of search-type tasks, the model has better advantages. And it also provides a new solution for a single state eye movement sequence.
Algorithms ; Discriminant Analysis ; Eye Movements ; Humans ; Markov Chains ; Normal Distribution ; Probability

OBJECTIVE: To observe the changes of serum metabolites in patients with multiple myeloma (MM) by metabonomics, and explore the potential biomarkers for diagnosis, prognosis, and progression of MM. METHODS: Serum samples were collected from 26 patients with MM and 50 healthy controls. The data detected by liquid chromatography-mass spectrometry (LC-MS) was input into SIMCA-14.0 software for multivariate statistical analysis. Principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), and orthogonal partial least squares discriminant analysis (OPLS-DA) were used to analyze the changes of metabolites. RESULTS: The metabolic change of uric acid and trans-vaccenic acid in serum samples of MM patients was 9.39 times and 2.77 times of these in healthy people, respectively, which were significantly higher than those of healthy people, and the difference was statistically significant(P<0.01). CONCLUSION Uric acid and trans-vaccenic acid are expected to be important metabolic indicators for the diagnosis, prognosis, and efficacy evaluation of MM, thus providing some clues for the pathogenesis of MM.
Biomarkers ; Chromatography, Liquid ; Discriminant Analysis ; Humans ; Mass Spectrometry ; Metabolomics ; Multiple Myeloma

Objective To construct a discriminant analysis model based on the differential expression of multiple microRNAs （miRNAs） in two kinds of blood samples （peripheral blood and menstrual blood） and three non-blood samples （saliva, semen and vaginal secretion）, to form an identification solution for peripheral blood and menstrual blood. Methods Six kinds of miRNA （miR-451a, miR-144-3p, miR-144-5p, miR-214-3p, miR-203-3p and miR-205-5p） were selected from literature, the samples of five kinds of body fluids commonly seen in forensic practice （peripheral blood, menstrual blood, saliva, semen, vaginal secretion） were collected, then the samples were divided into training set and testing set and detected by SYBR Green real-time qPCR. A discriminant analysis model was set up based on the expression data of training set and the expression data of testing set was used to examine the accuracy of the model. Results A discriminant analysis statistical model that could distinguish blood samples from non-blood samples and distinguish peripheral blood samples from menstrual blood samples at the same time was successfully constructed. The identification accuracy of the model was over 99%. Conclusion This study provides a scientific and accurate identification strategy for forensic fluid identification of peripheral blood and menstrual blood samples and could be used in forensic practice.
Body Fluids ; Discriminant Analysis ; Female ; Forensic Genetics ; MicroRNAs/genetics* ; Semen

Objective To estimate sex based on patella measurements of Sichuan Han population by computed tomography three-dimensional volume reconstruction technique, and to explore the application value of patella in sex estimation. Methods CT three-dimensional volume reconstruction images of patella of 250 individuals were collected, the four measurement indicators including patellar length, patellar width, patellar thickness, and patellar volume were measured. The t-test was used to determine measurement indicators with sex differences. Fisher discriminant analysis was used to establish the sex discriminant function and the prediction accuracy was calculated by leave-one-out cross validation. Results The sex differences of the four measurement indicators had a statistical significance （P<0.05）. The accuracy rate of the univariate discriminant function established by the patellar length was the highest （82.0%）. The accuracy rates of the all indicators discriminant function and the stepwise discriminant function were 80.4% and 81.6%, respectively. Conclusion It is feasible and accurate to estimate sex of Sichuan Han population by patella measurements with CT three-dimensional volume reconstruction technique. The method may be used as an alternative for sex estimation of Sichuan Han population when other bones with higher accuracy are not available.
Discriminant Analysis ; Female ; Forensic Anthropology ; Humans ; Imaging, Three-Dimensional ; Male ; Patella/diagnostic imaging* ; Sex Determination by Skeleton ; Tomography, X-Ray Computed

Objective To obtain the protein expression profile of rat liver tissue after death by the 2100 bioanalyzer combined with protein chip, and infer the relationship between protein expression profile and postmortem interval. Methods Rats were killed by abdominal anesthesia and placed at 16 ℃. Water-soluble proteins in liver tissues were extracted at 14 time points after death. The expression profile data of proteins with relative molecular weight of 14 000-230 000 were obtained using protein chip, and principal component analysis （PCA）, partial least squares-discriminant analysis （PLS-DA） and Fisher discriminant were used to analyze the data. Results According to the changes of protein expression profile, the postmortem interval was divided into group A （0 d）, group B （1-9 d）, group C （12-30 d） according to the result of PLS-DA. The prediction accuracy of the training set and test set of the model were all 100.0%, and the internal cross-validation of the training set was 100.0% according to Fisher discriminant. The Fisher discriminant model at each time point of group B and C was established to narrow the time window of postmortem interval estimation. The prediction accuracy of the training set and test set were all 100.0%, and the internal cross-validation accuracy of the training set was 100.0% in group B. The prediction accuracy of the training set and test set were respectively 95.2% and 78.6% in group C, and the internal cross-validation of the training set was 88.1%. Conclusion Protein chip detection technology can quickly and easily obtain the expression profile of water-soluble proteins of rat liver tissue with a relative molecular weight of 14 000-230 000 at different time points after death. PLS-DA and Fisher discriminant models are established to classify and predict the postmortem interval, in order to provide new ideas and methods for postmortem interval estimation.
Animals ; Autopsy ; Discriminant Analysis ; Least-Squares Analysis ; Postmortem Changes ; Protein Array Analysis ; Rats ; Technology

Objective To screen serum biomarkers after skeletal muscle contusion in rats based on gas chromatography-mass spectrometry （GC-MS） metabolomics technology, and support vector machine （SVM） regression model was established to estimate skeletal muscle contusion time. Methods The 60 healthy SD rats were randomly divided into experimental group （n=50）, control group （n=5） and validation group （n=5）. The rats in the experimental group and the validation group were used to establish the model of skeletal muscle contusion through free fall method, the rats in experimental group were executed at 0 h, 2 h, 4 h, 8 h, 12 h, 24 h, 48 h, 96 h, 144 h and 240 h, respectively, and the rats in validation group were executed at 192 h, while the rats in the control group were executed after three days' regular feeding. The skeletal muscles were stained with hematoxylin-eosin （HE）. The serum metabolite spectrum was detected by GC-MS, and orthogonal partial least square-discriminant analysis （OPLS-DA） pattern recognition method was used to discriminate the data and select biomarkers. The SVM regression model was established to estimate the contusion time. Results The 31 biomarkers were initially screened by metabolomics method and 6 biomarkers were further selected. There was no regularity in the changes of the relative content of the 6 biomarkers with the contusion time and the SVM regression model can be successfully established according to the data of 6 biomarkers and the 31 biomarkers. Compared with the injury time ［（55.344±7.485） h］ estimated from the SVM regression model based on the data of 6 biomarkers, the injury time ［（195.781±1.629） h］ estimated from the SVM regression model based on the data of 31 biomarkers was closer to the actual value. Conclusion The SVM regression model based on metabolites data can be used for the contusion time estimation of skeletal muscles.
Animals ; Biomarkers ; Contusions ; Discriminant Analysis ; Metabolomics ; Muscle, Skeletal/injuries* ; Rats ; Rats, Sprague-Dawley