The World Health Organization (WHO) states that 1 in 10 deaths in the world is due to accidents (trauma). Therefore, in the field of pre-hospital trauma care, adequate and timely treatment in the critical period may define the survival of a patient. Likewise, within the process of trauma patient care, there may be circumstances in which it is necessary to administer antibiotic therapy in order to avoid infections. Therefore, this work aims to contribute to the rational use of antibiotics in this type of patients, using data mining techniques to reduce the time required to select the most appropriate antibiotic for each case. To achieve this objective, a mobile application was developed to collect information on all patients treated in a public hospital and a private hospital in southern Ecuador during a period of 3 months. With this data, a decision support system based on cluster analysis was designed and classification techniques based on machine learning such as Random Forest and neural networks were tested. In the medical field, the preliminary results obtained with this research show the following: antibiotics of the beta-lactam family are the first choice, 56.35% as general monotherapy, 55.65% in the public hospital and 56.14% in the private hospital; the most used are cefalexin 14.4%, cefazolin 27.1% in both institutions. In the public hospital the frequency is higher than in the private hospital in these situations. It was found that "old" antibiotics or those of the first family of beta-lactam antibiotics are still used as the first choice, so that a guide should be made on the proper dosage and route of administration so as not to cause antibiotic resistance in the future. Regarding the decision support system, 85% accuracy was achieved in the classification of the type of antibiotic administered according to the type of injury suffered by the patient.
|Translated title of the contribution||Epidemiological Characteristics of the Use of Antibiotics in Surgical Trauma Patients Treated in Southern Ecuador: A Support System Based on Mobile Applications and Data Mining|
|Original language||Spanish (Ecuador)|
|Title of host publication||Tic y Sistemas Inteligentes como Herramientas de Soporte para el Manejo, Educación y Prevención del Trauma|
|State||Published - 1 Nov 2017|
- Data Mining
- Mobile Applications
- Decision Support