Techniques to Produce and Evaluate Realistic Multivariate Synthetic Data

Background: Proper data modeling in biomedical research requires sufficient data for exploration and reproducibility purposes. A limited sample size can inhibit objective performance evaluation. Objective: We are developing a synthetic population (SP) generation technique to address the limited sample size condition. We show how to estimate a multivariate empirical probability density function (pdf) by converting the task to multiple one-dimensional (1D) pdf estimations. Methods: Kernel density estimation (KDE) in 1D was used to construct univariate maps that converted the input variables (X) to normally distributed variables (Y). Principal component analysis (PCA) was used to transform the variables in Y to the uncoupled representation (T), where the univariate pdfs were assumed normal with specified variances. A standard random number generator was used to create synthetic variables with specified variances in T. Applying the inverse PCA transform to the synthetic variables in T produced the SP in Y. Applying the inverse maps produced the respective SP in X. Multiple tests were developed to compare univariate and multivariate pdfs and covariance matrices between the input (sample) and synthetic samples. Three datasets were investigated (n = 667) each with 10 input variables. Results: For all three datasets, both the univariate (in X, Y, and T) and multivariate (in X, Y, and T) tests showed that the univariate and multivariate pdfs from synthetic samples were statistically similar to their pdfs from the respective samples. Application of several tests for multivariate normality indicated that the SPs in Y were approximately normal. Covariance matrix comparisons (in X and Y) also indicated the same similarity. Conclusions: The work demonstrates how to generate multivariate synthetic data that matches the real input data by converting the input into multiple 1D problems. The work also shows that it is possible to convert a multivariate input pdf to a form that approximates a multivariate normal, although the technique is not dependent upon this finding. Further studies are required to evaluate the generalizability of the approach.

Classification and Change Detection of Tirupati Urban Area using Erosion and Dilation Based PCA Transform

In remote sensing, the identification of the land use and land cover (LULC) changes in the global and local region are developed by classification and detection algorithms. This classification system can be developed to meet the needs of state agencies, and Federal for an up-to-date analyze of LULC throughout the entire selected of region area. The multispectral images have multiple low-resolution bands due to lack of sensory acquisition problem, haze-covered on earth objects and atmospheric distributions. So difficult to analyze the full information, the user wrongly interprets the information. Image processing applications can be done for compress and enhance the details of land surface details. The Principal Component Analysis and Morphological operations are implemented for compressing and feature extract the color and earth object values with good accuracy level. Change Detection between the time difference of the proposed enhanced images for land objects classes was computed. The most extensive land cover change category identification of the Tirupati urban Agricultural and forest area for the last 14 years. The change analyzed by using the image differencemethod for obtaining the changing level of the forest and urban development areas between two-timeintervals.

Statistical feature embedding for heart sound classification

Cardiovascular Disease (CVD) is considered as one of the principal causes of death in the world. Over recent years, this field of study has attracted researchers’ attention to investigate heart sounds’ patterns for disease diagnostics. In this study, an approach is proposed for normal/abnormal heart sound classification on the Physionet challenge 2016 dataset. For the first time, a fixed length feature vector; called i-vector; is extracted from each heart sound using Mel Frequency Cepstral Coefficient (MFCC) features. Afterwards, Principal Component Analysis (PCA) transform and Variational Autoencoder (VAE) are applied on the i-vector to achieve dimension reduction. Eventually, the reduced size vector is fed to Gaussian Mixture Models (GMMs) and Support Vector Machine (SVM) for classification purpose. Experimental results demonstrate the proposed method could achieve a performance improvement of 16% based on Modified Accuracy (MAcc) compared with the baseline system on the Physionet2016 dataset.

Study on Alteration Information Extraction by Using Assimilation of Multi-Sensor Spectral Data

Some alterations of similar spectral reflectances cannot be distinguished accurately for their lower spectral resolution when the traditional methods, such as, band ratio and principal component analysis are used to extract alteration information from Landsat ETM multi-spectral data. In this paper, the band1~band7 of MODIS whose wave lengths are among 10~500nm, together with ETM’s multi-spectral bands, whose spatial resolutions are 30m, are chosen in the execution of data assimilation. After the third order wavelet transformation, the low-frequency component of ETM data are replaced by the MODIS data subsequently, then the inverse wavelet transform is in progress. The result of data assimilation consists of not only ETM’s spatial information but also MODIS’ spectral information. At last, four bands of assimilation results are selected to process PCA transform, as a result, two types of alteration in the study area are extracted accurately according to their components.