Prediction of Hospital Charges for the Cancer Patients with Data Mining Techniques / 대한의료정보학회지

OBJECTIVE: Predictions of hospital charges for cancer patients are very important, because they provide a basis for allocating medical resources in the hospital and for establishing national medical policies. But previous studies to predict hospital charges were mainly based on statistical analysis, which has used only a small aspect among huge medical data so that the prediction power was limited. Thus we developed four data mining models, including two artificial neural network (ANN) models and two classification and regression tree (CART) models, to predict both the total amount of hospital charges and the amount paid by the insurance of cancer patients and compared their efficacies. METHODS:The data was generated from400,625 medical records of 1,605 cancer patients who had been hospitalized toKyungHeeUniversityHospital fromMarch 1, 2003 to February 29, 2004. Clementine 8.1 programwas used to build four data mining prediction models, two for the total amount and two for the amount paid by insurance. The variables included all of the data fields of standard medical record form of Korea. The neural network model used feed-forward back propagation method, which had 2 hidden layers. For decision tree model, RELIEFF method was used and the maximum tree depth was set to 30.We divided the dataset into 67%of training dataset and 33%of test dataset, using stratified sampling. Linear correlation coefficient and gain chart were compared. RESULTS: The ANN models showed better linear correlation coefficient than the CART models in predicting both the total amount (0.824 vs. 0.791) and the amount paid by insurance (0.838 vs. 0.699). The estimated accuracy of ANN model was more than 98%to predict both total amount and amount paid by insurance. The CART model for total amount showed that the relative importance of the variables were duration of admission(0.073), number of consultation(0.061), and treatment group 16(0.06). The CART model for the amount paid by insurance showed that the relative importance of the cariables were duration of admission (0.09), number of ICUadmission (0.063), and number of consultations (0.062). The percent gain of ANN model shows better %gain than CART to predict total amount but to predict amount paid by insurance, ANN showed similar pattern to CART CONCLUSION: The ANNmodels showed better prediction accuracy than CART models. However, the CART models, which serve different information from ANN model, can be used to allocate limited medical resources effectively and efficiently. For the purpose of establishing medical policies and strategies, using those models together is warranted.

Anti-neutrophil Cytoplasmic Antibodies (ANCA) in Patients with Chronic Inflammatory Bowel Disease / 대한내과학회지

OBJECTIVE: Anti-neutrophil cytoplasmic antibody (ANCA), known as a useful diagnostic marker in patients with ulcerative colitis (UC), are specific for granule proteins of granulocytes and monocytes and induce distinct fluorescence patterns. To evaluate the significance of ANCA in chronic inflammatory bowel disease (IHD), the presence of ANCA in chronic IBD was studied using indirect immunofluorescent test (IIF), METHODS: Between March, 1994 and September 1995, 51 patients with chronic inflammatory bowel disease were subjected in this study. We had analysed the correlation between duration, disease activity, location by colonoscopy and radiologic examinations, steroid treatment. RESULTS: 1) Among 34 patients with ulcerative colitis (UC), ANCA was demonstrated in 23 patients (67.6%). Among 19 patients with other chronic IBD (4 Crohn's disease, 6 Behcet's colitis, 7 intestinal tuberculosis and 2 radiation colitis) 2 patients (10.5%) had ANCA. The positivity of ANCA in patients with UC was significantly higher than in patients with other chronic IBD. 2) In patients with UC, c-ANCA was positive in 2 (5.9%) patients and p-ANCA was positive in 21 (61.8%) patients. In patients with other chronic IBD, ANCA was positive in one patient with Behcet's colitis or one patient with intestinal tuberculosis but negative in all patients with Crohn's disease or radiation colitis. 3) The mean duration of disease in ANCA positive patients was 42.4 +/- 39.4 months and the mean duration of disease in ANCA negative patients was 44.9 +/- 36.8 months, but there was no significant difference. 4) The number of patients in clinically mild, moderate and severe group were 23 (37.6%), 6 (83.2%) and 5 (14.7%) respectively. Among these groups the number of ANCA positive patients were 15 (65.2%), 5 (83.2%) and 3 (60%) respectively, but there was no significant difference. 5) The number of patients with proctitis, left side colitis and pancolitis were 9 (26.5%), 14 (41.2%) and 11 (32.4%) respectively, Among these groups the number of ANCA positive patients were 4 (44.4%), 10 (71.4%) and 9 (81.8%) respectively, but there was no significant difference. 6) Among 13 patients with steroid treatment 9 patients (69.2%) were ANCA positive. Among 21 patients without steroid treatment 16 patients (76.2%) were ANCA positive, but there was no significant difference. CONCLUSION: Although there was no correlation between ANCA and duration, disease activity, location or steroid treatment in UC patients, ANCA could be a diagnostic marker of UC in chronic IBD patients.