Ml. Magnusson et al., Range of motion and motion patterns in patients with low back pain before and after rehabilitation, SPINE, 23(23), 1998, pp. 2631-2639
Study Design. Data were collected form 27 patients who were participating i
n a rehabilitation program for chronic low back pain. The patients were tes
ted on day 2 and day 11 of a 2-week rehabilitation program.
Objectives. To determine specific characteristics of trunk motion associate
d with long-term dysfunction caused by low back pain of various origin, to
determine if a neural network analysis system can be effective in distingui
shing between patterns, and to determine if the rehabilitation has an effec
t on range and pattern of motion.
Summary of Background Data. There is a lack of objective measures for evalu
ating the efficacy of rehabilitation programs. Numerous studies have establ
ished the difficulty of evaluating low back pain. Existing techniques, such
as imaging methods, are in many cases either very rough and inaccurate or
expensive and ineffective. A technique for evaluation of motion patterns in
low back pain was developed based on analysis of dynamic motion features s
uch as shape, velocity, and symmetry of movements.
Methods. Dynamic motion data were collected before and after rehabilitation
from 27 patients with low back pain by using a triaxial goniometer. Range
of motion and features of the movement, such as shape, velocity, and repeti
tiveness, were extracted for analysis.
Results. Motion features showed significant improvement after the rehabilit
ation program.
Conclusions. A neural network based on kinematic data is an excellent model
for classification of low back pain dysfunction. Such a system could marke
dly improve the management of low back pain for an individual patient.