Effects of interval treadmill training on spatiotemporal parameters in children with cerebral palsy: a machine learning approach

Background Children with cerebral palsy (CP) have reduced step length, reduced symmetry, and greater step width compared to their peers. Short-burst interval locomotor treadmill training (SBLTT) is a novel rehabilitation paradigm for children with CP that may improve spatiotemporal outcomes. However, for interventions like SBLTT, quantifying rehabilitation responses and optimizing therapy parameters remains challenging. Machine learning and causal modeling provide a platform to quantify step-by-step changes during gait training to understand mechanisms driving individual responses. Research question What is the direct effect of SBLTT on step length, asymmetry, and step width in children with CP? Methods We recruited four children with spastic CP, ages 4-13. Each participant received 24 sessions of SBLTT over 8-12 weeks, with spatiotemporal outcomes monitored with an instrumented treadmill. We used Bayesian Additive Regression Trees (BART) to model the direct effect of therapy parameters on step length, step length asymmetry, and step width. Additionally, we generated in silico data for 150 virtual participants to quantify the quality of BART models to capture rehabilitation progression. Results After SBLTT, participants’ step lengths increased by 26 ± 13% (pre-post effect). Controlling for treadmill speed, time in session, limb, and treadmill incline with BART demonstrated that SBLTT directly increased step length for three participants (direct effect: 13.5 ± 4.5%), while one participant decreased step length (−11.6%). SBLTT had minimal effects on step length asymmetry and step width. Virtual datasets demonstrated that BART could accurately predict step length progression (R2 = 0.73) and plateaus in progression (R2 = 0.87), with better model fit for participants with less step-to-step variability. Significance Tools such as BART can leverage step-by-step data collected during gait training to monitor progression and optimize rehabilitation protocols. This work can help personalize rehabilitation and understand the causal mechanisms driving individual responses. ### Competing Interest Statement The authors have declared no competing interest. ### Clinical Trial NCT04467437 ### Funding Statement This work was supported by Seattle Children′s Hospital CP Research Pilot Study Fund 2020 Award, UW Rehabilitation Medicine Walter C, and Anita C. Stolov 2021 Research Fund, and NSF Graduate Research Fellowship Program Award [DGE-1762114]. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: The IRB Committee of the University of Washington gave ethical approval for this work. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes All data produced in the present work are contained in the manuscript.