Poster | 6th Internet World Congress for Biomedical Sciences |
Jose Luis Ruiz Gonzalez(1), Francisco Taboada(2), Antonio M. López(3), Alberto Diez(4)
(1)Universidad de Oviedo - Oviedo. Spain
(2)Hospital Central de Asturias - Oviedo. Spain
(3)(4)University of Oviedo - Gijón. Spain
Contact address: |
Jose Luis Ruiz Gonzalez Universidad de Oviedo Oviedo Asturias 33006 Spain ruiz@ctv.es |
[Medical Electronics & Engineering] |
||||
[New Technology] |
Even though the clinical picture of the patients that are referred to an Intensive Care Unit can be easily diagnosed when it is established its much more difficult to recognize in its initial stages when the symptoms are imprecise. This usually means important delays in the treatment and, as a consequence, worse results. If we could make an early diagnosis of the patient by the simultaneous monitoring of different devices, the possibilities of surviving could increase.
The main goal of this work, financed by the Universidad de Oviedo inside the Programa de Interés Regional, is the development of a method of multiple monitoring, both bloody and bloodless, which allows an early diagnosis and an aggressive treatment from the beginning, and also allowing to make an analysis a posteriori of the stored data, in order to improve our knowledge about the symptoms previous to shock situations. To do so, we´re developing an integrated system which allows us to capture the data sent by different monitors, to show in a combined way the data we´ve read, and to store all the data in a centralized way. The system consists in a central computer, in which a database with all the information is stored, and one or more moveable computers, connected to the server by an Ethernet LAN. These moveable computers get the data from the monitors, show them in real time, and send them to the database.
The monitors send the data to the computer through a serial RS-232-C connection, which is already implemented in the monitors. The input at the reading computer is done through an RS-232 multiplex, with which we can read all the data through a single communications port at the PC.
The capture system has been designed in different layers, each one specialized in a single task, in order to allow an easy extension of the system, adding new monitoring devices and making the system as unaffected as possible by a possible communications failure. The lower layer is made by a whole of modules, each one specialized in the communication with a single monitor type. Above that layer, there is a coordination layer, which makes all the previous modules work properly integrated into the system and processes the information brought by them, storing it in a temporary buffer. Finally, at the upper level, there is a layer that gets the data from the previous layer and inserts it into the database. This layer works independently of the others, making the system unaffected by some net failures and allowing the system to capture from computers which are not plugged into the net, making it possible to send the data to the database later.
In the design of the database we´ve taken into account two main aspects. In first place, we want to store all the measured data, together with the patient´s data, the previous diagnosis and the appropriate treatment, so in the future we can have enough data to make proper analysis. Additionally, we want the database to be open enough to allow us to include new devices and variables without having to change it.
For the analysis of the information stored in the database, we´ve decided not to implement our own algorithms inside the system, but to develop an interface to export these data to external programs, such as SPSS and Excel. This way, all the possibilities of such programs can be exploited, making it an open system, not restricted to some algorithms fixed into the system
Finally, we´ve studied and worked alternative options to statistical treatment of data based in Artificial Intelligence (AI) techniques, such as genetic algorithms or artificial neural nets to make analysis that could not be made with classical techniques.
[Medical Electronics & Engineering] |
||||
[New Technology] |