The convergence of artificial intelligence based data processing techniques with
information and communication technologies has enabled the development of a smart
diagnostic system that can be used to provide health care support in remote rural areas.
The paper focuses on the development of a smart diagnostic system that can predict the
physiological state of a patient given the past physiological data. The smart system has
been implemented on an Altera Cyclone EP1K6Q240C8 FPGA chip. Since patients’ data
randomly vary, therefore no crisp opinion can be made about the physiological state of a
patient knowing only the present data. The system employs a smart agent whose role is
to monitor and diagnose on a regular time basis and assist the health care professionals
in the process of therapeutic decisions. The paper provides an introduction to the notion
of smart agent based telemedicine. An extended example on the problem of monitoring
renal patients using Body Mass Index (BMI), blood glucose, urea, creatinine, systolic and
diastolic blood pressure has been presented in this paper. The system has been tested
with height, weight, blood glucose, urea, creatinine, systolic and diastolic blood pressure
data of patients where all data other than height have been taken at 10 days interval.
Applying the methodology, the chance of attainment of critical renal condition of a patient
before the patient actually reaches a critical state has been predicted with confidence.
1. INTRODUCTION
THE development of information and
communication technologies has opened up many
exciting possibilities for developing new services for
the mankind. Over the last few years, clinicians, health
service researchers and others have been investigating
the use of advanced telecommunications and
information technologies to improve health care. At
the intersection of many of these efforts lies“telediagnostics” – a combination of innovative and
mainstream technologies. Telediagnostics is the “use
of electronic information and communication
technologies to predict the physiological state of a
patient when distance separates the participants” [9].
Since the earliest documented telemedicine studies,
the focus of telemedicine has mainly been on using
videoconferencing to replace an in-person visit [1,2].
It has also been noted in teleradiology, that the storeand-
forward model is more practical because it
eliminates the need for scheduling [1]. Store-andforward
in the form of a simple e-mail for physician patient communication was first documented in 1994
[3,4].
___________________________________
Paper No45-A; Copyright © 2006 by the IETE. |
|
Conventional telemedicine systems using public
switched telephone network are already available to
enable a doctor to monitor a patient remotely for home
care or emergency applications [1,5-7]. But such type
of telemedicine systems cannot be used to provide
health care support in the remote rural areas where
public switched telephone network is not established.
Mobile systems have also been studied including one
mobile satellite communication system [1,5]. A
considerable delay is, however, incurred to get a
diagnostic decision about the patient from the doctor
at the remote site.
The current research focuses on the development
of an FPGA based artificial diagnostic system that can
predict the physiological state of a patient, given the
past physiological data of the patient. The scheme
under consideration can provide alarm to the relevant
personnel in advance before the patient reaches a
critical state, who would then contact the physician at
the remote site. The physician would then take up
necessary actions to provide medical support to the
patient. The remainder of the paper is organized as
follows:
|
|
