Search

Metric multidimensional scaling (MDS) is a widely used multivariate method with applications in almost all scientific disciplines. Eigenvalues obtained in the analysis are usually reported in order to calculate the overall goodness-of-fit of the distance matrix. In this paper, we refine MDS goodness-of-fit calculations, proposing additional point and pairwise goodness-of-fit statistics that can be used to filter poorly represented observations in MDS maps. The proposed statistics are especially relevant for large data sets that contain outliers, with typically many poorly fitted observations, and are helpful for improving MDS output and emphasizing the most important features of the dataset. Several goodness-of-fit statistics are considered, and both Euclidean and non-Euclidean distance matrices are considered. Some examples with data from demographic, genetic and geographic studies are shown. [ABSTRACT FROM AUTHOR]

Content-based medical image retrieval (CBMIR) is the emerging research area in the field of medical image retrieval and analysis. The proposed CBMIR system includes two phases: feature extraction and feature matching. In the current decades, the feature matching approaches are developed well, but still the feature extraction algorithms are facing difficulties such as poor result in illuminance, lighting variations, etc. For addressing these concerns, Manhattan-distance based histogram of oriented gradients (M-HOG) was proposed in this research paper by assuming the different contributions of the training and testing samples from (Dicom medical image dataset for computed tomography (CT) and magnetic resonance imaging (MRI) images). At first, the medical images were enhanced by using Gaussian filter. The respective pre-processed medical image features were extracted by applying M-HOG. Using a Euclidean distance measure, the similarity between the testing and training medical images was calculated and matched. In the experimental section, the proposed system improved the retrieval accuracy in CBMIR system up to 5–15% related to the existing methodologies. [ABSTRACT FROM AUTHOR]