DEVELOPING A FRAMEWORK FOR ANALYZING DOOR TO BALLOON TIME PROCESSES BASED ON SELF- ORGANIZING MAP

Abstract

Self-organizing networks is a kind of neural network with unsupervised that is very powerful in analyzing complex spaces. Door-to-balloon time is one of the most important criteria for quality evaluation of cardiac patients cure process. Considering that process faster doing of restoring the blood to vessel in patients of severe heart stroke is with reduction of mortality rate, so reduction in door-to-balloon time, the treatment process analysis, quality management and continuous improvement of it are very important. In this research the door-to-balloon time has been investigated by using self-organizing map. In order to this, getting electrocardiogram (ECG), patient transportation to emergency department (ED) and catheterization laboratory are observed and sub processes are extracted by direct observation, interviewing emergency staff and experts of Tehran Heart Center Cathlab. Using door- to- balloon time of Tehran heart center during 2010 to 2012 it has been shown that 6.5% of patients who referred with acute coronary syndrome are diagnosed STEMI and are treated by angioplasty. The mean of door-to-balloon time of this center is obtained 137.8 minutes with standard deviation of 161.338. self-organizing network for analyzing process and developing a framework is checked using different levels of process clustering and after identifying processes are presented in order to visualization and analyzing door-to-balloon time workflow. Moreover the method of self-organizing map can provide approximately from simultaneous effect of different therapy levels, obtained results feedback and effective levels identification. Analyzing door- to- balloon time processes based on self-organizing map instead of just paying attention to the average time of treatment and developing a framework for extracting some data about door- to- balloon time are such innovations of this research.