Optimal time lags from causal prediction model help stratify and forecast nervous system pathology

Area of Science:

• Neuroscience
• Biomedical Engineering
• Computational Biology

Background:

• Current neurological disorder diagnosis relies on observational criteria, often leading to symptom heterogeneity within diagnostic groups.
• A need exists for automated methods to stratify patient populations for personalized treatment and prognosis.
• Existing approaches lack precision in identifying distinct subtypes of gait pathologies.

Purpose of the Study:

• To develop and validate new methods for automatically stratifying a diverse population cohort, including healthy individuals and those with neurological disorders.
• To establish a framework for tailoring clinical interventions and forecasting disease severity based on identified patient clusters.
• To define and analyze novel internal motor timings derived from biorhythmic motion analysis.

Main Methods:

• Utilized a simple walking task to capture micro-fluctuations in biorhythmic motions.
• Applied non-linear causal network connectivity analyses in both temporal and frequency domains.
• Integrated stochastic mapping with causal prediction principles and the theory of reafference to operationalize results.

Main Results:

• Successfully stratified a mixed cohort of healthy controls and patients with neurological disorders.
• Identified distinct clusters of gait pathologies based on internal motor timings derived from motion analysis.
• Demonstrated the potential for personalized clinical interventions based on these identified clusters.

Conclusions:

• The developed methods offer a novel approach to objective neurological disorder stratification through gait analysis.
• Internal motor timings provide a new biomarker for understanding and classifying neuromotor control differences.
• This work renovates the theory of internal models in neuromotor control, paving the way for precision medicine in neurology.