File
Furone R1,2, Vagnini A2, Zanotti G1, Temporiti F1, Gatti R1,3
1Humanitas Research Hospital, Physiotherapy Unit, Rozzano (Milan), Italy, 2BTS S.p.A., Garbagnate Milanese (Milan), Italy, 3Humanitas University, Pieve Emanuele (Milan), Italy
Background: Disorders of balance are one of the possible causes of fall in the elderly. The measure of Center of Mass (CoM) displacement is relevant in assessing risk of fall. Different techniques have been used to assess CoM, but they present disadvantages in terms of cost, time consuming and post processing. Wearable inertial sensors has been proposed to overcome these issues but they have never been compared to the gold standard , the optoelectronic system.
Purpose: The aim of the study was to analyse the agreement of CoM displacement measured with an user- friendly inertial sensor, characterized by a sensor fusion technology with four integrated inertial platforms including an accelerometer, magnetometer and gyroscope (G-sensor, BTS, Italy), and with an optoelectronic system (SmartDX, BTS, Italy).
Methods: Thirty young healthy volunteers (22±2.5 years old, 15 male, 15 female) were enrolled. They did a single session acquisition in the Movement Analysis and Rehabilitation Laboratory of Humanitas Research Hospital. Subjects were in a standardized standing position with a fixed 10 cm intramelleolar distance. In order to collect different conditions of CoM displacement, subjects performed four tasks, one minute each, in bipodalic stance, opened and closed eyes, and in monopodalic stance, on the right and left leg. The inertial sensor, with a retroreflective marker over it, was positioned to the sacrum at S2 level and fixed with medical tape. The two system were synchronized by a trigger box and the signals were sampled at 100 Hz. Total, medio-lateral (ML) and anterior-posterior (AP) length of trace, ML and AP speed, the confidence ellipse of CoM , the root mean squared (RMS) and range in ML and AP direction were computed. For the statistical analysis, the four tasks were overall considered. The Intraclass Correlation Coefficients (ICCs) and Bland-Altman plot were used to assess systems' agreement.
Results: ICCs revealed an excellent correlation for ML speed (r=0.93), AP speed (r=0.95), total length (r=0.91), good correlation for ML length (r=0.88), AP length (r=0.89), ML RMS (r=0.77), moderate correlation for AP RMS (r=0.65) , confidence ellipse (r=0.748) and ML range (r=0.699), poor correlation for AP range (r=0.47) (always p 0.001). The Bland-Altman plot revealed a good accuracy between the two systems. The Bland-Altman plot showed the smallest mean difference for AP speed (-0.0003 m, 95% limit of agreement: -0.0032 to 0.0025, 6 outliers) and the greatest mean difference for confidence Ellipse (0.37 m2, 95% limit of agreement: -0.1463 to 0.2204, 5 outliers).
Conclusion(s): The inertial sensor can offer an accurate method to study CoM displacement in the variables of AP-ML speeds, AP-ML lengths, total length, AP-ML RMS and confidence ellipse. It can be applied in several clinical conditions in which is important to assess these variables of CoM displacement.
Implications: The inertial sensor provides a wearable, low-cost, no time-consuming CoM's evaluation. The inertial sensors can be used in different clinical environments and it can support functional evaluation of balance.
Keywords: center of mass, inertial sensor, balance
Funding acknowledgements: No souce of funding or support was received.
Purpose: The aim of the study was to analyse the agreement of CoM displacement measured with an user- friendly inertial sensor, characterized by a sensor fusion technology with four integrated inertial platforms including an accelerometer, magnetometer and gyroscope (G-sensor, BTS, Italy), and with an optoelectronic system (SmartDX, BTS, Italy).
Methods: Thirty young healthy volunteers (22±2.5 years old, 15 male, 15 female) were enrolled. They did a single session acquisition in the Movement Analysis and Rehabilitation Laboratory of Humanitas Research Hospital. Subjects were in a standardized standing position with a fixed 10 cm intramelleolar distance. In order to collect different conditions of CoM displacement, subjects performed four tasks, one minute each, in bipodalic stance, opened and closed eyes, and in monopodalic stance, on the right and left leg. The inertial sensor, with a retroreflective marker over it, was positioned to the sacrum at S2 level and fixed with medical tape. The two system were synchronized by a trigger box and the signals were sampled at 100 Hz. Total, medio-lateral (ML) and anterior-posterior (AP) length of trace, ML and AP speed, the confidence ellipse of CoM , the root mean squared (RMS) and range in ML and AP direction were computed. For the statistical analysis, the four tasks were overall considered. The Intraclass Correlation Coefficients (ICCs) and Bland-Altman plot were used to assess systems' agreement.
Results: ICCs revealed an excellent correlation for ML speed (r=0.93), AP speed (r=0.95), total length (r=0.91), good correlation for ML length (r=0.88), AP length (r=0.89), ML RMS (r=0.77), moderate correlation for AP RMS (r=0.65) , confidence ellipse (r=0.748) and ML range (r=0.699), poor correlation for AP range (r=0.47) (always p 0.001). The Bland-Altman plot revealed a good accuracy between the two systems. The Bland-Altman plot showed the smallest mean difference for AP speed (-0.0003 m, 95% limit of agreement: -0.0032 to 0.0025, 6 outliers) and the greatest mean difference for confidence Ellipse (0.37 m2, 95% limit of agreement: -0.1463 to 0.2204, 5 outliers).
Conclusion(s): The inertial sensor can offer an accurate method to study CoM displacement in the variables of AP-ML speeds, AP-ML lengths, total length, AP-ML RMS and confidence ellipse. It can be applied in several clinical conditions in which is important to assess these variables of CoM displacement.
Implications: The inertial sensor provides a wearable, low-cost, no time-consuming CoM's evaluation. The inertial sensors can be used in different clinical environments and it can support functional evaluation of balance.
Keywords: center of mass, inertial sensor, balance
Funding acknowledgements: No souce of funding or support was received.
Topic: Robotics & technology; Robotics & technology
Ethics approval required: Yes
Institution: Humanitas Research Hospital
Ethics committee: Ethics Commitee for Human Investigation of Humanitas Research Hospital
Ethics number: 181/18
All authors, affiliations and abstracts have been published as submitted.
