The Module for Predicting Complications after Bone Marrow Transplantation in Children with Leukemia

Abstract

The paper presents the program module that estimates the probability of complications after the bone marrow transplantation. The shown approach is based on a logistic regression. The medical database containing the records of 155 patients aged from 5 to 21 is used. The program module contains a detailed graphical representation of the results in the form of a ROC curve, the calculated values of true and false predictions as well as likelihood ratio of the obtained values. The areas of reliable and unreliable predictions as well as the area where predictions are impossible are highlighted on the chart.

Keywords

• трансплантация костного мозга
• прогнозирование осложнений
• логистическая регрессия
• чувствительность
• специфичность
• истинный прогноз
• ложный прогноз
• отношение шансов
• отношение риска
• доверительный интервал
• программный модуль