Machine Learning and Statistical Analyses of Sensor Data Reveal Variability Between Repeated Trials in Parkinson’s Disease Mobility Assessments
2024

Variability in Parkinson's Disease Mobility Assessments

Sample size: 312 publication 10 minutes Evidence: moderate

Author Information

Author(s): Khalil Rana M., Shulman Lisa M., Gruber-Baldini Ann L., Shakya Sunita, Hausdorff Jeffrey M., von Coelln Rainer, Cummings Michael P.

Primary Institution: University of Maryland

Hypothesis

How does test-retest reliability vary in mobility tasks for individuals with Parkinson's disease compared to controls?

Conclusion

The study found that repeating mobility tasks may not improve diagnostic accuracy, suggesting that a single trial could suffice for assessments.

Supporting Evidence

  • Total duration of mobility tasks showed good to excellent reliability (ICC = 0.75 to 0.95).
  • Variability in subtask duration was greater than total task duration.
  • Machine learning models indicated that repeating trials generally provided little additional benefit for diagnostic accuracy.

Takeaway

This study looked at how well people with Parkinson's disease can repeat mobility tests. It found that doing the tests more than once might not help doctors make better decisions.

Methodology

The study assessed test-retest reliability of mobility tasks using machine learning models and statistical metrics on data from 262 PD participants and 50 controls.

Limitations

The study's findings may not generalize to other motor or cognitive assessments beyond the specific mobility tasks evaluated.

Participant Demographics

{"controls":50,"PD":262,"age":{"controls":"64.1 ± 9.8","PD":"66.9 ± 9.3"},"gender":{"controls":"38% male","PD":"62% male"}}

Digital Object Identifier (DOI)

10.3390/s24248096

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication