Heart Rate and Accelerometer Data Fusion for Activity Assessment of Rescuers During Emergency Interventions

Abstract

The current state of the art in wearable electronics is the integration of very small devices into textile fabrics, the so-called ¿smart garment.¿ The ProeTEX project is one of many initiatives dedicated to the development of smart garments specifically designed for people who risk their lives in the line of duty such as fire fighters and Civil Protection rescuers. These garments have integrated multipurpose sensors that monitor their activities while in action. To this aim, we have developed an algorithm that combines both features extracted from the signal of a triaxial accelerometer and one ECG lead. Microprocessors integrated in the garments detect the signal magnitude area of inertial acceleration, step frequency, trunk inclination, heart rate (HR), and HR trend in real time. Given these inputs, a classifier assigns these signals to nine classes differentiating between certain physical activities (walking, running, moving on site), intensities (intense, mild, or at rest) and postures (lying down, standing up). Specific classes will be identified as dangerous to the rescuer during operation, such as, ¿subject motionless lying down¿ or ¿subject resting with abnormal HR.¿ Laboratory tests were carried out on seven healthy adult subjects with the collection of over 4.5 h of data. The results were very positive, achieving an overall classification accuracy of 88.8%.

Lifelogging Data Validation Model for Internet of Things Enabled Personalized Healthcare

“…As a major risk measure for chronic diseases, daily physical activity recognition and monitoring with wearable sensors have been investigated by a number of researchers [21][22][23][24][25][26][27][28] [29][30][31][32][33][34]. In [22][23], authors carry out a study on recognizing and classifying physical activity by analyzing signal features from 3D (triaxial) accelerometers on hip and wrist and GPS data with a hybrid classifier of custom decision tree and neural networks.…”

“…In IoT based personalized healthcare systems, physical activity data comes mostly from globally heterogeneous third party devices. The traditional classification methods [21][22][23][24][25][26][27][28] are infeasible to handle these scattered and heterogeneous physical activity data.…”

“…The products like Fitbit or Moves [29] occur on a 2D plane (Persons × TimeLine), which refer to scenarios that single device is used by increasing population over time. Similarly, physical activity recognition with sensor fusion [21][22][23][24][25][26][27][28] appears on a 2D plane (Devices × TimeLine) for classifying individual person's activities with historical health data. To distinct from the above two categories of studies, the target of LPAV-IoT model is a cube of rapid-growth lifelogging physical activity.…”

“…Among this data, physical activity is mostly wellobserved due to the maturity of microelectromechanical systems (MEMS) based accelerometer technology as well as easily and openly accessible Global Position System (GPS). Numerous research works [21][22][23][24][25][26][27][28] and commercial products [29][30][31] [33] have attempted to accurately monitor physical activity and access activity patterns and intensity level, by using either dedicated wearable sensors [29][30][31] or advanced machine learning algorithms [22][23][24][25]. But these studies mostly depend on performance optimization of single sensor or a combination of GPS and accelerometer by analyzing raw sensors' signals.…”

Validation of Smart Garments for Physiological and Activity-Related Monitoring of Humans in Harsh Environment

“…The RMS allows saving all the data for post-processing analysis. More details on the RM S and the algorith ms behind these mult i-parameters warn ing detections are reported in [8][9][10] and [12,13] respectively.…”