• Publikationen
Show publication details

Trimpop, John; Schenk, Hannes; Bieber, Gerald; Lämmel, Friedrich; Burggraf, Paul

Smartwatch based Respiratory Rate and Breathing Pattern Recognition in an End-consumer Environment

2017

iWOAR 2017

International Workshop on Sensor-based Activity Recognition (iWOAR) <4, 2017, Rostock, Germany>

Smartwatches as wearables became part of social life and practically and technically offer the possibility to collect medical body parameters next to usual fitness data. In this paper, we present an evaluation of the respiratory rate detection of the &gesund system. &gesund is a health assistance system, which automatically records detailed long-term health data with end-consumer smartwatches. The &gesund core is based on technology exclusively licensed from the Fraunhofer Institute of applied research. In our study, we compare the &gesund algorithms for respiration parameter detection in low-amplitude activities against data recorded from actual sleep laboratory patients. The results show accuracies of up to 89%. We are confident that wearable technologies will be used for medical health assistance in the near future.

Show publication details

Trimpop, John; Haescher, Marian; Bieber, Gerald; Matthies, Denys J.C.; Lämmel, Friedrich; Burggraf, Paul

The Digital Health Companion: Personalized Health Support on Smartwatches via Recognition of Activity- and Vital-Data

2015

Proceedings of the International Summer School on Visual Computing 2015

International Summer School on Visual Computing <1, 2015, Rostock, Germany>

It has been shown that in various fields of social life, people tend to seek opportunities to measure their daily activities, bodily behaviors, and health related parameters. These kinds of activity tracking should be accomplished comfortably, unobtrusively and implicitly. Tracking behavior can be important for certain user groups, such as the growing population of elderlies. These people have a substantially higher risk of falling down, as they often live alone and thus have a greater need for other supporting services, as emergencies quickly occur. We would like to support these people, while providing a comfortable emergency detection and a monitoring of physical activities. Moreover, we believe such tracking applications to be beneficial for any user group, since we can perceive the trend of quantified self: knowing about one's own body characteristics, which is expressed in body movement. Simultaneously, we also perceive that a strong desire for a comprehensive monitoring of vital and health data is emerging. In this paper we describe the concept and implementation of the Digital Health Companion, a smart health support system that combines research developments of activity, vital data, and anomaly recognition with the functionality of contemporary smartwatches. The system's health monitoring includes an emergency detection and allows for the prevention of health risks in the short and long term through the recognition of body movement patterns.