Maximum probability of existence of cancer in human bodies is normally diagnosed very late, so that it is highly cumbersome for physicians to cure. Reliability in predicting cancer at initial stage is always needed, s...Maximum probability of existence of cancer in human bodies is normally diagnosed very late, so that it is highly cumbersome for physicians to cure. Reliability in predicting cancer at initial stage is always needed, so that curing and medical recovery is possible. In this paper, an investigation was made to diagnose the presence of blood cancer using Markov Chain Monte Carlo （MCMC） trace model, which is most efficient on a wide range of complex Bayesian statistical models. The analysis was carried out using version 18 of SPSS AMOS software. Totally, 19 components were considered from the blood samples of 750 patients. Various factors such as class, age, lymphatics, block of affarc, block of lymph c, block of lymph s, bypass, extravasate, regeneration of, early uptake in, lym nodes dimin, lyre nodes enlar, change in lym, defect in node, changes in node, changes in strue special forms, dislocation, exclusion of node, number of nodes in blood cancer are analyzed. The maximum likelihood estimators of the parameters were derived and assessed their performance through a Monte Carlo simulation study. The convergence in prior distribution and posterior distribution takes irregular position in the diagrams and thus blood cancer is diagnosed through this model.展开更多
文摘Maximum probability of existence of cancer in human bodies is normally diagnosed very late, so that it is highly cumbersome for physicians to cure. Reliability in predicting cancer at initial stage is always needed, so that curing and medical recovery is possible. In this paper, an investigation was made to diagnose the presence of blood cancer using Markov Chain Monte Carlo （MCMC） trace model, which is most efficient on a wide range of complex Bayesian statistical models. The analysis was carried out using version 18 of SPSS AMOS software. Totally, 19 components were considered from the blood samples of 750 patients. Various factors such as class, age, lymphatics, block of affarc, block of lymph c, block of lymph s, bypass, extravasate, regeneration of, early uptake in, lym nodes dimin, lyre nodes enlar, change in lym, defect in node, changes in node, changes in strue special forms, dislocation, exclusion of node, number of nodes in blood cancer are analyzed. The maximum likelihood estimators of the parameters were derived and assessed their performance through a Monte Carlo simulation study. The convergence in prior distribution and posterior distribution takes irregular position in the diagrams and thus blood cancer is diagnosed through this model.
