Article Text

Download PDFPDF

OP0238 Measurement of spinal mobility in axial spondyloarthritis using inertial sensors: reliability and validation preliminary results
  1. FJ Mata-Perdigon1,
  2. IR Martínez Sanchez2,
  3. JL Garrido-Castro1,
  4. C González-Navas1,
  5. LM Fernandez-Ahumada2,
  6. DA Gomez1,
  7. VC Perez-Guijo1,
  8. P Font-Ugalde1,
  9. E Collantes-Estevez1
  1. 1Maimonides Institute for Biomedical Research of Cordoba
  2. 2University of Cordoba, Cordoba, Spain


Background Axial spondyloarthritis (axSpA) is a chronic rheumatic disease that causes reduction of mobility in the patients' spine. There are several indices to analyze this mobility: BASMI, which lacks precision and sensitivity to change according to different authors, and UCOASMI (1) based on motion capture, which needs extensive resources that limit its practical applicability. Inertial measurement unit sensors (IMU) give, in real time, the 3D orientation of any anatomical place of the patient. It could be a sensitive, flexible and cheap technology, useful for assessing mobility in AxSpa, but validation studies are needed.

Objectives To assess reliability and validity of inertial sensors for measuring spinal mobility in patients with axSpA.

Methods 14 subjects: 7 patients with axSpA (5 male and 2 female, age 51.4±6.7 years, evolution time 25.4±11.3 years, 85.7% B27 positive) and a control group of 7 healthy individuals matched in gender and age were recruited. Cervical and lumbar movements were evaluated using 3 IMU sensors (located at forehead, D3 and L4) and a 3D motion capture system synchronously. A test/retest was performed at 5 minutes in the same day with the IMUs and in two days with both systems. Measurements of metrology, BASMI and UCOASMI indices were obtained. An index, iUCOASMI, was calculated using the same measurements used for UCOASMI, but obtained by inertial sensors.

Results Table shows mean values (SD) for each range of movement expressed in degrees. BASMI, UCOASMI and iUCOASMI indexes are also included. Intraclass correlation coefficient (ICC) is indicated as α: >0.98 – Excellent, β: 0.95–0.98 – Very good and γ: 0.7–0.95 – Good, δ: <0.7 – Bad. RMSE error was less than 10° for all measures. There was good correlation (p<0.01) between iUCOASMI with BASFI, BASG, UCOASMI and BASMI. Graph shows results of linear regression between measures obtained with both system (for example: cervical frontal flexion obtained by motion capture and IMUs have a R2 of 0.97) and between iUCOASMI with UCOASMI and BASMI.

Table 1

Conclusions The IMU system measured range of movement, showing good ICC both in the same day and in the two days test/retest. The iUCOASMI, has also shown an excellent correlation with UCOASMI, and with BASMI. Therefore, these kind of systems, based on IMU, may be useful for analyzing spinal mobility in patients with axSpA in a more accurate and reliable way compared with conventional metrology, and more flexible and cheap than other advanced systems, improving their practical applicability.


  1. Validation of a new objective index to measure spinal mobility: the University of Cordoba Ankylosing Spondylitis Metrology Index (UCOASMI). Garrido-Castro JL, et al. Rheumatol Int. 2014 Mar;34(3):401–6.


Acknowledgements This study was supported by PIN-0079–2016 research project (Consejería de Salud de la Junta de Andalucía, SPAIN).

Disclosure of Interest None declared

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.