Construction of the prognostic model for assessing the risk of medullary thyroid cancer metastases using binary logistic regression

Background. Medullary thyroid cancer is a topical disease that often accompanied by metastases. The danger of this pathology requires timely and adequate surgery. Objective to assess the prognostic value informativeness some clinical indicators with selection most optimal reliable potential factors in development mathematical equation for calculating personal probability detecting metastases medullary absence instrumental signs preoperative stage. Materials methods. Patients carcinoma without locoregional participated study. To verify metastatic deposits, pathomorphological study was performed using TNM classification (UICC). StatPlus Pro v.7.6, EpiTools MedCalc statistical calculators were used processing results. Data Mining technologies degree impact predictors data mining add-on MS Office. diagnostic test, ROC analysis corresponding characteristic curve constructed calculation area under it (according DeLong method). For operational characteristics tests, 95 % confidence interval calculated according Wilson method. results considered statistically significant at p < 0.05. Results. Basal calcitonin, patient sex, multifocality, total tumor size have been shown be priori risk assessment. These can not only as unfavourable prognosis, but also individual determination surgery scope. Conclusions. method binary logistic regression latent metastasis showed lower sensitivity (0.77 vs 0.89) higher specificity (0.90 0.64) contrast monofactorial prognosis based on calcitonin levels. Further improvement model additional erroneous test applied training sample. proposed due calculations Excel allows you easily quickly obtain information, so an tool when choosing surgical treatment.