Development, accuracy and reliability of a three dimension postural analysis system

• Grant number: 10/12500-0
• Support Opportunities: Regular Research Grants
• Start date: December 01, 2010
• End date: November 30, 2012
• Field of knowledge: Health Sciences - Physiotherapy and Occupational Therapy
• Principal Investigator: Amélia Pasqual Marques
• Grantee: Amélia Pasqual Marques
• Host Institution: Faculdade de Medicina (FM). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated researchers: Arnaldo Diniz Gonçalves Neto

Abstract

Postural assessment is regarded as the initial physical examination basic procedure to be conducted on patients. However, most existing systems for postural assessment analyze the posture only in two dimensions, not allowing the quantification of spins on the Y axis. These systems do not use the methodology of real-time scanning of anatomical markers, which could increase the error during the process. Other systems such as computed tomography or the use of ultrasound do assess posture in three dimensions but are not easily accessible. Hypothesis: We believe a system for postural assessment in three dimensions based on infrared sensors have good reliability, accuracy and is useful for clinical use and scientific research as it allows the evaluation of rotation on the Y axis and evaluation of vertebral curvatures in three dimensions. Objectives: This study aims to develop a postural analysis system in three dimensions based on the automatic scanning of anatomical points for body sensors and infrared emitters at low cost and verify its accuracy and reliability. Methods: A system will be developed, composed of a set of infrared sensors (controls of the Nintendo Wii videogame), a pen-emitting infrared light, a software capable of receiving the coordinates X, Y and Z of infrared sensors and interpret them in five grades freedom: rotation X-axis (flexion and extension), the y-axis rotations (rotation to the right and left), Z axis rotation (lateral tilt to the right and left) translations in the X axis (axial rotation to the right and left ) and Z axis translations (anterior and posterior). No translations will be measured on the Y axis, since those would be possible with the use of X-rays. To determine accuracy, three blind raters previously trained in the system will evaluate a mannequin of head, trunk and pelvis positioned at different known angles. The differences between the known values and those found will be considered as system error. The evaluations will be repeated after two weeks and the values will be used to calculate the intra-rater and inter-rater reliability. Statistical analysis: For inter-rater and intra-rater reliability calculus it will be used the ICC (2,1) and ICC (3,1) respectively. It will also adopted an ± = 0.05. (AU)