On the Unification of Common Actigraphic Data Scoring Algorithms

Area of Science:

• Biomedical Engineering
• Sleep Science
• Signal Processing

Background:

• Actigraphy is a common, cost-effective method for analyzing human movement patterns.
• Sleep and circadian rhythm research frequently utilizes actigraphy data.
• Existing sleep-wake scoring algorithms for actigraphy lack a unified mathematical framework.

Purpose of the Study:

• To investigate and mathematically describe seven common sleep-wake scoring algorithms for actigraphy.
• To propose a unified mathematical framework for five of these algorithms.
• To elucidate the underlying principles and empirical factors influencing algorithm design.

Main Methods:

• Mathematical framework development.
• Analysis of Cole-Kripke, Sadeh (two versions), Sazonov, Webster, UCSD, and Scripps Clinic algorithms.
• Identification of algorithmic equivalencies through mathematical modeling.

Main Results:

• Five common sleep-wake scoring algorithms are mathematically unified.
• These five algorithms are demonstrated to be equivalent to low-pass FIR filters with similar characteristics.
• Explanations are provided for empirical factors influencing algorithm design.

Conclusions:

• A robust mathematical framework enhances the understanding of actigraphy sleep-wake scoring algorithms.
• The equivalence to low-pass FIR filters offers novel insights into algorithm operation.
• This work provides a foundational understanding for future development and application of actigraphy algorithms.