Causes and characteristics of the chest pain(in the chest-pain clinic)

BACKGROUND: Ischemic heart disease is the most important cause of the chest pain, and its frequency is increasing enormously. The purpose of this study is to find out the way of early detection and/or ruling out the cardiogenic chest pain by history taking. METHODS: From July 1996 to December 1999, 248 patients visited the chest-pain clinic and took the questionnaire about characteristics of the chest pain. And we found out the diagnosis that caused the chest pain. 46 patients of them were excluded because of the unreliable responses or uncertain diagnosis. So, we compared the characteristics of the chest pain with causes for 202 patients. RESULTS: The sex ratio of patients was 1.43:1(male:female). The average age was 41.8+/-14.0 for male and 47.3+/-14.8 for female. The causes of the chest pain were cardiogenic(23.2%), musculo-skeletal(19.3%), psychogenic (14.8%), gastrointestinal(12.4%), and pulmonary disease(6.9%). Patients with the past history of diabetes, hypertension, alcohol intake, or angina were more likely to have cardiac disease. Choking (O.R=2.19, C.I.=1.08-4.44), splitting(O.R=3.38, C.I.=1.24-9.21), or exploding pain (O.R=2.65, C.I=1.02-6.88) was more likely to be originated from cardiac disease. And patients with cardiogenic chest pain aggravated their symptoms by climbing the stairs(O.R=3.47, C.I= 1.52-7.90). But, pricking pain(O.R=0.18, C.I.= 0.04,-0.82) or chest pain associated with dyspepsia(O.R.=0.16, C.I.=0.04-0.69) was less likely to be originated from cardiac disease. CONCLUSION: For detection and/or ruling out the cardiogenic chest pain, we have to check out characteristics of the pain, but also factors that associated with the pain or aggravating the pain.

On the Use of Neural Networks for the Risk Factor Analysis of NIDDM / 대한의료정보학회지

There were many cases to apply artificial intelligence to medicine. Neural networks are nonparametric pattern recognition techniques that can be used to model complex relationships. In this paper, we present the analysis of the risk factors of the noninsulin-dependent diabetes mellitus using the artificial neural network and the logistic regression model. First, we developed five prediction models using artificial neural networks and a logistic regression model with the data of Yonchon study of diabetes mellitus. Next, we measured each area under the ROC(Receiver-Operating Characteristic) plots for the performance, and results re followings; multilayer perceptron with seventeen variables(MLP17) was 0.7608, multilayer perceptron with seven variables(MLP7) was 0.7664, radial basis function network with seventeen variables(RBF17) was 0.7919, radial basis function network with seven variables(RBF7) was 0.7715 and logistic regression model(REG7) was 0.8343. All of the variables used are seventeen, and seven variables for neural networks(MLP7 and RBF7) were selected by logistic regression model. The order of higher risk variables in the neural networks(slope) did not completely agree with that in the logistic regression model(odds ratio). However, all of the four higher risk variables that were significant in the statistic model(0.05) also had large slopes(0.3) in the neural network model. And our neural network model also display the influence of another variables in development of the noninsulin-dependent diabetes mellitus.