Smart garment for trunk posture monitoring: A preliminary study

Abstract

Background: Poor postures of the spine have been considered in association with a number of spinal musculoskeletal disorders, including structural deformity of the spine and back pain. Improper posturing for the patients with spinal disorders may further deteriorate their pain and deformities. Therefore, posture training has been proposed and its rationale is to use the patient's own back muscles to keep the spine within the natural curvature. A posture training device may help to facilitate this therapeutic approach by providing continuous posture monitoring and feedback signals to the patient when "poor" posture is detected. In addition, the users of the device may learn good postural habits that could carry over into their whole life.

Methods: A smart garment with integrated accelerometers and gyroscopes, which can detect postural changes in terms of curvature variation of the spine in the sagittal and coronal planes, has been developed with intention to facilitate posture training. The smart garment was evaluated in laboratory tests and with 5 normal subjects during their daily activities.

Results: Laboratory tests verified that the accuracy of the system is < 1 degrees and < 1.5 degrees in static and dynamic tilting measurements respectively. The results showed that the smart garment could facilitate subjects to prevent prolonged poor postures of the spine, especially the posture of the lumbar spine in which at least 40% of the time in poor posture were reduced.

Conclusion: The smart garment has been developed to be a portable and user-friendly trunk posture monitoring system and it could be used for collection of the trunk posture information and provision of instant feedback to the user if necessary for posture training purpose. The current pilot study demonstrated that the posture of normal subjects could be monitored and trained via this smart garment. With further clinical investigations, this system could be considered in some flexible spinal deformities such as scoliosis and kyphosis.

Figure 1

Laboratory tests. The experimental setting for (a) static tilting and (b) dynamic tilting measurements.

Figure 2

A subject with the smart garment. A subject with the smart garment.

Figure 3

The calculation of the trunk posture change. The trunk postural changes were calculated with the data of the sensor modules and was defined as the change of inclination differences between angles α_(1,2,3) and β_(1,2,3), respectively, which formed by the tangent and vertical line in the (a) sagittal and (b) coronal planes.

Figure 4

The average trunk posture deviations in 4-day trunk posture monitoring. The average trunk posture deviation from the neutral standing position of the subjects in 4-Day trial. Three line-segments represent the relative tilting of the 3 sensor modules to the neutral position. Dotted lines represent the posture in neutral position. The inter-segmental angles between 2 line-segments represent the trunk posture angles. (NF: without feedback, F: with feedback).

Figure 5

The percentage of the time in out-of-target posture. The percentage of the time in out-of-target posture. (NF: without feedback, F10: Feedback with 10 degrees threshold, F5: Feedback with 5 degrees threshold).

Figure 6

The frequency distribution of the trunk postures. The frequency distribution of the angles of the thoracic and lumbar curves in the sagittal and coronal planes. (NF: without feedback, F: with feedback).