A bstract-This paper presents a statistical interpretation of the chest radiograph for the detection of pulmonary tuberculosis (PTB). Each region of interest was represented by a vector of wavelet texture measures which is then multiplied by the orthogonal matrix Q. The first two elements of the transformed vectors were shown to have a bivariate normal distribution. Misclassification probabilities were estimated using probability ellipsoids and discriminant functions. The most important result of this study recommends the detection of pulmonary tuberculosis by constructing discriminant function using maximum column sum energy texture measures where the misclassification probabilities were less than 0.15. In the validation exercise, the proposed discriminant procedure yielded 94% correct classification rate.
This study attempted to discriminate three types of lung disease pair-wise, namely lobar pneumonia (PNEU), pulmonary tuberculosis (PTB) and lung cancer (LC) using chest radiograph. A modified principal component method applied to wavelet texture measures yielded feature vectors for the pair-wise statistical discrimination procedure. The combination of mean of energy and maximum value texture measures gave good pairwise discrimination rate and classification rate for PNEU-PTB, PNEU-LC and PTB-LC. The result if this study is very promising and further work are needed for verification and validation study with larger sample size.
This paper attempts to estimate the transition probabilities of credit ratings for a number of companies whose ratings have a dependence structure. Binary codes are used to represent the index of a company together with its ratings in the present and next quarters. We initially fit the data on the vector of binary codes with a multivariate power-normal distribution. We next compute the multivariate conditional distribution for the binary codes of rating in the next quarter when the index of the company and binary codes of the company in the present quarter are given. From the conditional distribution, we compute the transition probabilities of the company's credit ratings in two consecutive quarters. The resulting transition probabilities tally fairly well with the maximum likelihood estimates for the time-independent transition probabilities.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.