Authors:
L. V. R. Prasada Raju
1
;
Madhubabu Anumukonda
1
and
Shubhajit Roy Chowdhury
2
Affiliations:
1
Center of VLSI and Embedded Systems Technologies, International Institute of Information Technology, Hyderabad and India
;
2
School of Computing and Electrical Engineering, Indian Institute of Technology (IIT), Mandi and India
Keyword(s):
Fail-safe Systems, Fault Detection, Fault Tolerant Systems, Patient Monitoring, Biomedical Engineering, Wearable Computers.
Abstract:
Non-invasive Biomedical signals sensing and measured vital signs monitoring is critically significant for patients monitored in intensive care units(ICU) at hospital environments. Typically, nurses do periodic measures & record the primary vital signs and attentive to all types of alarms generated by the monitoring instrument.
At critical times, the problem of false & insignificant alarms may deviate the health assessment of a patient, which lead to unnecessary additional medication to a patient or a patient death and this suggests improving the device design strategy for patient safety by reduction of insignificant alarms.
This paper presents, the aptness usage of safety related architectures in design of fault tolerant multimodal human health monitoring systems with safe de-gradable mechanism for monitoring authentic measurements of health parameters. This design approach investigations and evaluations is to improve the safety features in a non-invasive medical diagnostic system,
by addressing the current challenges like, effective removal of single point of failures (SPOF) and reducing the fault alarms. An experimental, Cardiac health monitoring system (CHMS) has been developed using configured 2oo2- (two out of two logic) safety architecture having two non-redundant diverse channels along with a fault tolerant de-gradable safe function. This approach evaluation, confirms reduction of SPOF and perform the functional safe operation by checking the minimum two working diverse channels for triggering the safe output. The CHMS uses sensors PCG & ECG, and concurrently computes vital heartrate parameter in each diverse channel. The Pearson’s coefficient of correlation method is used to correlate the measured heartrate values from these individual sensors for high degree of relationship (rAB> +0.5). The resultant output is used in the safe function and confirms near reduction of fault alarms along with near uninterrupted heart rate monitoring with respect to 5 people being studied.
(More)