The Use of Motion Detection Systems in Training of Patients with Neurological Disorders: A Systematic Review

Background: This review aims to examine which motion detection systems are used in training in neurological rehabilitation, what their content and outcome of training is. Methods: A computerised systematic review was conducted... [ view full abstract ]

Background: This review aims to examine which motion detection systems are used in training in neurological rehabilitation, what their content and outcome of training is.
Methods: A computerised systematic review was conducted in PubMed, Cinahl, Cochrane Database and IEEE. The following key words were used: Motion, Movement, Detection, Capture, Kinect, Rehabilitation, Nervous System Diseases, Multiple Sclerosis, Stroke, Spinal Cord. The methodological quality of the studies was scored by the Van Tulder’s Quality assessment. The descriptive assessment is reported by training systems, patient group and ICF level of training and outcome.
Results: A total of nineteen studies were selected (mean Van Tulder score = 7.84 ± 4.00). Microsoft Kinect was used in thirteen studies. Fourteen studies included persons with stroke. Upper limb training was performed in twelve studies, while four studies focused on lower limb and three the whole body. ICF activity level was the main focus in intervention and outcome measures. An individualised training was not reported. Most commonly used feedback was visual-auditory feedback. Most prevailing combination of training parameters is four weeks of training, three training sessions per week, training for one hour per day. All but one study reported improvement on one or more ICF levels.
Conclusion: In neurological rehabilitation, motion detection systems are promising tools for increasing treatment intensity and support improvement on one of more ICF levels. However, to explore the use of motion detection systems in clinical practice more research is needed by performing larger multicentre RCTs with long-term follow-up using an individualised task-oriented approach.