An Automatic and early detection of the deterioration of patients in Intensive and Intermediate Care Units

• Main Authors: Javier Aldo Balladini, Pablo Bruno, Rafael Zurita, Cristina Orlandi
• Format: Article
• Language: English
• Published: Postgraduate Office, School of Computer Science, Universidad Nacional de La Plata 2018-12-01
• Series: Journal of Computer Science and Technology
• Subjects: intensive care unit; clinical decision support system; medical rules processing; big data; embedded system
• Online Access: http://journal.info.unlp.edu.ar/JCST/article/view/1139

In the Intensive and Intermediate Care Units of healthcare centres, many sensors are connected to patients to measure high frequency physiological data. In order to analyse the state of a patient, the medical staff requires both appropriately presented and easily accessed information. As most medical devices do not support the extraction of digital data in known formats, medical staff need to fill out forms manually. The traditional methodology is prone to human errors due to the large volume of information, with variable origins and complexity. The automatic and real-time detection of changes in parameters, based on known medical rules, will make possible to avoid these errors and, in addition, to detect deterioration early. In this article, we propose and discuss a high-level system architecture, an embedded system that extracts the electrocardiogram signal from an analog output of a medical monitor, and a real-time Big Data infrastructure that integrate Free Software products. We believe that the experimental results, obtained with a simple prototype of the system, demonstrate the viability of the techniques and technologies used, leaving solid foundations for the construction of a reliable system for medical use, able to scale and support an increasing number of patients and captured data.