Search

Abstract Introduction Conflicting results persist regarding the effectiveness of robotic-assisted gait training (RAGT) for functional gait recovery in post-stroke survivors. We used several machine learning algorithms to construct prediction models for the functional outcomes of robotic neurorehabilitation in adult patients. Methods and materials Data of 139 patients who underwent Lokomat training at Taipei Medical University Hospital were retrospectively collected. After screening for data completeness, records of 91 adult patients with acute or chronic neurological disorders were included in this study. Patient characteristics and quantitative data from Lokomat were incorporated as features to construct prediction models to explore early responses and factors associated with patient recovery. Results Experimental results using the random forest algorithm achieved the best area under the receiver operating characteristic curve of 0.9813 with data extracted from all sessions. Body weight (BW) support played a key role in influencing the progress of functional ambulation categories. The analysis identified negative correlations of BW support, guidance force, and days required to complete 12 Lokomat sessions with the occurrence of progress, while a positive correlation was observed with regard to speed. Conclusions We developed a predictive model for ambulatory outcomes based on patient characteristics and quantitative data on impairment reduction with early-stage robotic neurorehabilitation. RAGT is a customized approach for patients with different conditions to regain walking ability. To obtain a more-precise and clearer predictive model, collecting more RAGT training parameters and analyzing them for each individual disorder is a possible approach to help clinicians achieve a better understanding of the most efficient RAGT parameters for different patients. Trial registration: Retrospectively registered.

Abstract Background We designed a novel ankle foot orthosis (AFO), namely, ideal training AFO (IT-AFO), with motion feedback on the hemiparetic lower limb to improve ambulation in individuals with stroke-related hemiplegia. We, therefore sought to compare the kinematic parameters of gait between IT-AFO with and without dynamic control and conventional anterior-type AFO or no AFO. Methods Gait parameters were measured using the RehaWatch® system in seven individuals with hemiplegia (mean 51.14 years). The parameters were compared across four conditions: no AFO, conventional anterior AFO, IT-AFO without dynamic control, and IT-AFO with dynamic control, with three trials of a 10-m walk test for each. Results The dorsiflexion angle increased during the swing phase when the IT-AFO was worn, and it was larger with dynamic control. These data can confirm drop foot improvement; however, the difference between the parameters with- and without-AFO control conditions was not significant in the swing phase. The IT-AFO with or without dynamic control enhanced the loading response to a greater extent between the hemiparetic and unaffected lower limbs than conventional AFO or no AFO. The duration of the stance phase on the hemiparetic lower limb was also longer when using IT-AFO with and without dynamic control than that when using conventional AFO, which improved asymmetry. User comfort and satisfaction was greater with IT-AFO than with the other conditions. Conclusions The IT-AFO with dynamic control improved gait pattern and weight shifting to the hemiparetic lower limb, reducing gait asymmetry. The difference with and without dynamic control of IT-AFO is not statistically significant, and it is limited by sample size. However, this study shows the potential of IT-AFO in applying positive motion feedback with gait training. Trial registration Taipei Medical University-Joint Institutional Review Board. N201510010 . Registered 12 February 2015. http://ohr.tmu.edu.tw/main.php .