Calibrating a novel multi-sensor physical activity measurement system

Abstract

Advancing the field of physical activity (PA) monitoring requires the development of innovative multi-sensor measurement systems that are feasible in the free-living environment. The use of novel analytical techniques to combine and process these multiple sensor signals is equally important. This paper describes a novel multi-sensor 'integrated PA measurement system' (IMS), the lab-based methodology used to calibrate the IMS, techniques used to predict multiple variables from the sensor signals, and proposes design changes to improve the feasibility of deploying the IMS in the free-living environment. The IMS consists of hip and wrist acceleration sensors, two piezoelectric respiration sensors on the torso, and an ultraviolet radiation sensor to obtain contextual information (indoors versus outdoors) of PA. During lab-based calibration of the IMS, data were collected on participants performing a PA routine consisting of seven different ambulatory and free-living activities while wearing a portable metabolic unit (criterion measure) and the IMS. Data analyses on the first 50 adult participants are presented. These analyses were used to determine if the IMS can be used to predict the variables of interest. Finally, physical modifications for the IMS that could enhance the feasibility of free-living use are proposed and refinement of the prediction techniques is discussed.

Figure 1

The IMS and its various components

Figure 2

Individual responses from the various sensors of the IMS and corresponding metabolic equivalent (MET) values for 6 different activities. (A) Acceleration response from the hip accelerometer. (B) Acceleration response from the wrist accelerometer. (C) Respiratory sensor response. (D) Light sensor response to 5 different testing conditions.

Figure 2

Individual responses from the various sensors of the IMS and corresponding metabolic equivalent (MET) values for 6 different activities. (A) Acceleration response from the hip accelerometer. (B) Acceleration response from the wrist accelerometer. (C) Respiratory sensor response. (D) Light sensor response to 5 different testing conditions.

Figure 2

Individual responses from the various sensors of the IMS and corresponding metabolic equivalent (MET) values for 6 different activities. (A) Acceleration response from the hip accelerometer. (B) Acceleration response from the wrist accelerometer. (C) Respiratory sensor response. (D) Light sensor response to 5 different testing conditions.

Figure 2

Individual responses from the various sensors of the IMS and corresponding metabolic equivalent (MET) values for 6 different activities. (A) Acceleration response from the hip accelerometer. (B) Acceleration response from the wrist accelerometer. (C) Respiratory sensor response. (D) Light sensor response to 5 different testing conditions.

Figure 3

Comparison between mean estimated and measured respiratory variables obtained using the one-belt abdominal respiratory sensor IMS (A) Breathing frequency using the correlation criteria and (B) ventilation volumes.

Figure 3

Comparison between mean estimated and measured respiratory variables obtained using the one-belt abdominal respiratory sensor IMS (A) Breathing frequency using the correlation criteria and (B) ventilation volumes.

Figure 4

Comparison between mean predicted METS from the one-belt IMS system and actual METs obtained from the portable metabolic unit.

Figure 5

Activity type recognition accuracy of two pattern recognition techniques and four prediction models using the one-belt abdominal respiratory sensor IMS. Values are mean and SD.

Figure 6

Individual activity type prediction performance of the SVM with Gaussian kernel pattern recognition technique using the one-belt abdominal respiratory sensor IMS.