Performance of an accelerometer-based pulse presence detection approach compared to a reference sensor

Accelerometer sensors are ubiquitously available in consumer electronics such as smart phones and wearables. So far, mainly biomedical applications using accelerometers focus on providing contextual information like step counting, activity monitoring or motion artifact detection and suppression. Still, these sensors offer interesting opportunities for vital sign monitoring, even potentially for the demanding case around cardiac arrest. In this paper we show a basic feasibility study to compare the performance of an accelerometer (ACC) based pulse detection approach versus a commercially available device. For healthy subjects we found an excellent sensitivity of ACC-based pulse detection. The ACC performance was not influenced by changes in position, and ACC sensor placement was easy. The decision time for the commercial pulse detection device ranged from 10.0 – 25 s, while for our ACC-based approach it was 3.5 – 5.0 s using a trained classifier. From this preliminary study we conclude that ACC sensors might offer interesting opportunities for applications in emergency care for vital sign detection.

[1]  P. Moule,et al.  Checking the carotid pulse: diagnostic accuracy in students of the healthcare professions. , 2000, Resuscitation.

[2]  Christian Eickholt,et al.  Feasibility of pulse presence and pulse strength assessment during head-up tilt table testing using an accelerometer located at the carotid artery , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[3]  Stefan Fernsner,et al.  AS34 Detection of prehospital cardiac arrest by lays: Validation of a miniaturized sensor system in patients with cardiopulmonary bypass , 2011 .

[4]  F J Ochoa,et al.  Competence of health professionals to check the carotid pulse. , 1998, Resuscitation.

[5]  W. Dick,et al.  Checking the carotid pulse check: diagnostic accuracy of first responders in patients with and without a pulse. , 1996, Resuscitation.

[6]  J. Soar,et al.  European Resuscitation Council Guidelines for Resuscitation 2015: Section 3. Adult advanced life support. , 2015, Resuscitation.

[7]  Paulo Carvalho,et al.  Cardiac status assessment with a multi-signal device for improved home-based congestive heart failure management , 2011, 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[8]  T. Schauer,et al.  First-aid sensor system: New methods for single-point detection and analysis of vital parameters such as pulse and respiration , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[9]  Armin Bolz,et al.  Determining the cut-off of 1-point-vital signs detection based of a non-linear oscillating circuit (CPR|Check®) , 2012 .

[10]  Colin A Graham,et al.  Evaluation of a new method for the carotid pulse check in cardiopulmonary resuscitation. , 2002, Resuscitation.

[11]  Hannu Savolainen,et al.  Distal Pulse Palpation: Is It Reliable? , 1999, World Journal of Surgery.

[12]  James K. Russell,et al.  Towards an algorithm for automatic accelerometer-based pulse presence detection during cardiopulmonary resuscitation , 2016, 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[13]  Adrian Tarniceriu,et al.  Evaluation of accuracy and reliability of PulseOn optical heart rate monitoring device , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[14]  J. Tibballs,et al.  Reliability of pulse palpation by healthcare personnel to diagnose paediatric cardiac arrest. , 2008, Resuscitation.

[15]  C. Shearman,et al.  Peripheral pulse palpation: an unreliable physical sign. , 1992, Annals of the Royal College of Surgeons of England.