Walking function disorders are typical for patients after cerebral stroke. A cerebral stroke
normally affects one hemisphere and causes a hemiplegic syndrome. The gait of hemiplegic
patients has very specific features: reduced walking speed, increased double stance phase,
and reduced amplitude of movement in the leg joints. Biofeedback technology (BFB) is
currently considered effective and promising for training walking function, including in
patients after cerebral stroke. The technology is based on capturing a physiological
parameter and presenting it to the patient in a perceivable form, so that the subject can
understand its changes and respond appropriately. BFB can be used independently or as part of
rehabilitation therapy. Nevertheless, efficiency, as noted by most authors, remains the
subject of discussion. This is due to the fact, that at the previous stage of development of
these systems, there was no technical capability to use the specific biomechanical gait
parameters as targets for training. Therefore, more general parameters-such as walking speed,
step frequency, etc.-were and are still used. This circumstance is attributable to the very
nature of the main biomechanical gait parameters, which require special means of recording.
One of the significant technical difficulties in BFB implementation is the need for accurate
and fast registration of the gait parameters in real-time to use them for biofeedback. At the
same time, the use of portable sensors for BFB training goals can represent a certain
solution to technical problems. In recent years, owing to its important advantages, wearable
IMU technology (systems using inertial measurement units) has been widely applied for
capturing biomechanical gait parameters. Investigators used a system that was originally
developed with our participation for targeted training based on biofeedback according to the
biomechanical parameters of gait. The use of inertial technology and artificial intelligence
technology has made it possible to use biomechanical parameters of gait (time and general
gait parameters, EMG and kinematics of leg's joints) for biofeedback in a very low-cost and
practically convenient way. Biofeedback training courses based on target biomechanical gait
parameters are being studied. For targeted biofeedback training, various biomechanical
parameters are used: parameters of the gait cycle, EMG or kinematics of joint movements. The
number of sessions is 8-11 for each patient. Clinical gait analysis is carried out before and
after a course of training. Changes in biomechanical parameters that occurred at the end of
the training course are assessed in comparison with those before training, and both statuses
(before and after training) are compared with similar gait parameters in a group of healthy
adults.
Stroke patients participated in the study in Federal Center of Cerebrovascular Pathology and
Stroke FMBA in Moscow, Russia. The study was approved by a local ethic committee and followed
principles of the Declaration of Helsinki.
Single-blinded controlled clinical trial. The study involved stroke patients with hemiparesis
(no more than 3 points on a scale Rankin). Experimental groups are determined by the target
training parameter (time and general gait parameters, EMG and kinematics of leg's joints).
One target parameter is using for one group. The number of sessions is 8-11 for each patient
during three weeks of hospital stay. The duration of each session for each patient on each
training day varies according to his well-being and current exercise tolerance, but does not
exceed 30 minutes of training in one session. . Changes in biomechanical parameters that
occurred at the end of the training course are assessed in comparison with those before
training, and both statuses (before and after training) are compared with similar gait
parameters in a group of healthy adults.