Author/Authors :
Raeesi، Ahmad نويسنده Health Information Technology, Zahedan University of Medical Sciences, Zahedan, Iran. , , Ebrahimi، Saeed نويسنده Department of Mechanical Engineering of Yazd University , , Irfan Nia، Leila نويسنده Health Information Technology, Zahedan University of Medical Sciences, Zahedan, Iran , , Arji، Goli نويسنده Instructor of Health Information Technology, Zabol University of Medical Sciences, Zabol, Iran , , Askani، Moslem نويسنده Health Information Technology, Zahedan University of Medical Sciences, Zahedan, Iran. ,
Abstract :
Introduction: The aim of this study was to determine the performance of data mining techniques for predicting the causes of traumatic brain injuries in Khatamolanbya hospital, Zahdan city.
Method: In this cross–sectional, the study population included all patients who died of brain injury. Data were collected by the use of a researcher- made check list, provided under the direct observation of authorities in this area and analyzed by the data mining software of Clementine 12.0.
Results: According to the results of this algorithm, C5.0 decision tree algorithm has an accuracy of 81.4 percent, the highest precision; then, the algorithm is C & R(The Classification and Regression) with 77.8 percent.
Conclusion: Overall, it can be concluded from the decision tree algorithm that age is one of the leading causes of traumatic brain injuries . The results showed that all the cases involving traumatic lesions of the brain lead to the patient’s death.. Although in some algorithms, some of the variables are important, they cannot be used alone as the main variable to be taken into account for the death of the patient.
