Breath detection using a fuzzy neural network and sensor fusion

Abstract-Patient monitoring is an important part ofhealthcare workflow in hospitals. In the general ward, nurses typically check on patients three times a day by manually measuring and taking note of vital signs every eight hours. Sometimes this results in misdetection of adverse physiological events, which in turn may lead to severe and irreversible conditions. Although conventional bedside monitors could be used in general wards, they are expensive and impractical for ambulatory patient monitoring (bulky size and wires restricting patients' mobility).Based on our Sensium ® system-on-chip Toumaz has recently developed wearable, low cost technologies for healthcare ambulatory monitoring. The SensiumVitals ® is an end to end system that incorporates a new and alternative concept in wireless monitoring -the digital patch. This wireless, unobtrusive, lightweight and disposable body-worn device is capable of continuously acquiring and processing vital signs information from patients (temperature, hear rate and respiration rate) in real-time. To achieve this, the digital patch runs a number of embedded algorithms, and then wirelessly transmits every two minutes the data to bridges and servers for interpretation and display. This enables continuous monitoring of patients and early detection of physiological adverse events; thus alerting clinical staff about potential patient deterioration that may cause irreversible damage and in some cases fatal consequences if untreated.Unfortunately, one major issue of healthcare ambulatory monitoring technologies is the negative impact of body movement on the quality of the physiological data. Motion artifacts (MAs) often distort the ECG signals and compromise the reliability of vital signs computation. Furthermore, MA and heart contaminants are responsible for inaccurate results when processing respiration rates from impedance pneumography signals.In this paper we briefly describe of our system and embedded algorithms, including how we dealt with different disturbances affecting the quality of the signals. Finally, we present preliminary results of the evaluation of our SensiumVitals ® in terms of accuracy and performance of our algorithms when compared with a validated/clinical vital signs monitor.

show abstract

The use of fuzzy techniques in the study of non-invasive mechanical ventilation therapies

Nebot

Valdés²

Proceedings of 6th International Fuzzy Systems Conference

View full text Add to dashboard Cite

Breath detection using a fuzzy neural network and sensor fusion

Cited by 6 publications

References 4 publications

Autonomous patient monitoring with a pressure sensor array

Autonomous patient monitoring with a pressure sensor array

Challenges and Trade-Offs Involved in Designing Embedded Algorithms for a Low-Power Wearable Wireless Monitor

The use of fuzzy techniques in the study of non-invasive mechanical ventilation therapies

Contact Info

Product

Resources

About