Systematic prediction of mortality in trauma patients based on Arterial Blood Gas

Context- An effort to predict the final condition of patients is one of the purposes of many studies; since it enables the treatment system to provide the necessary facilities in the best possible time and prevent wasting time and energy as well as increasing patient mortality. Research purpose- This study was purposed to investigate the correlation between arterial blood gas (ABG) and patient mortality and design a system to predict the final patients condition. Method- In this study, a method has been proposed to identify dynamic systems to estimate the final condition of trauma patients and predict their death or survival probability during treatment or being confined in the medical center. The proposed method by using the information of patients arterial blood gases identifies a linear model indicating the correlation between these gases and the patients final condition. This method is based on system identification using ARX model simulated in MATLAB and its results are presented. Results- Data of 2802 patients (365 deaths and 2437 survivors) with an average age of 37.87 years old and GCS average of 9.27 including 470 female and 2332 male patients were studied. The designed structure was tested with 62.57% accuracy to be able to predict patient mortality. Therefore, it can be stated that the proposed method has a good accuracy in predicting the final patients condition based on dynamic analysis. Discussion and Conclusion- It is unavoidable mortality due to accidents and severe injuries. Also, it is important to predict the death probability based on data from the early hours of the onset of trauma in patients; since it takes time to collect the data of patients condition. Therefore, it is very important to find reliable methods to measure the patients condition and predict the mortality. The study of these methods has always been considered by physicians due to its high importance. This study has almost been able to meet physicians need by providing a method based on the study of dynamics and dynamic relationships discussing arterial and mortal blood gases.