Abstract
IntroductionMany physiological signals yield fractal characteristics, i.e., finer details at higher magnifications resemble details of the whole. Evidence has been accumulating that such fractal scaling is basically a consequence of interaction-dominant feedback mechanisms that cooperatively generate those signals. Neurodegenerative diseases provide a natural framework to evaluate this paradigm when this cooperative function declines. However, methodological issues need to be cautiously taken into account in order to be able to provide reliable as well as valid interpretations of such signal analyses.MethodsTwo conceptually different fractal analyses, i.e., detrended fluctuation analysis (DFA) and analysis of cumulative distributions of durations (CDDs), are applied to actigraphy data of 36 geriatric in-patients diagnosed with dementia. The impact of the used time resolution for data acquisition on the assessed fractal outcome parameters is particularly investigated. Moreover, associations between these parameters and scores from the Mini-Mental-State-Examination and circadian activity parameters are explored.ResultsBoth analyses yield significant deviations from (mono-)fractal scaling over the entire considered time range. DFA provides robust measures for the observed break-down of fractal scaling. In contrast, analysis of CDDs results in measures which highly fluctuate with respect to the time resolution of the assessed data which affects also further derived quantities such as scaling exponents or associations with other (clinically relevant) assessed parameters.DiscussionTo scrutinize actigraphic signal characteristics and especially their (deviations from) fractal scaling may be a useful tool for aiding diagnosis, characterization, and monitoring of dementia. However, results may, besides contextual aspects, also substantially depend on specific methodological choices. In order to arrive at both reliable and valid interpretations, these complications need to be carefully elaborated in future research.
Highlights
Many physiological signals yield fractal characteristics, i.e., finer details at higher magnifications resemble details of the whole
The assessed measures for deviations from fractal scaling (i.e., |α12| for fluctuation amplitudes determined via Detrended fluctuation analysis (DFA) and LLR as well as GOF ratios for analysis of cumulative distributions of durations (CDDs)) result in a significant deviation of sample means from the respective values associated with fractal scaling irrespective of the used time resolution of the activity data, see Figure 2
We found evidence that both locomotor activity signal characteristics considered in this work [i.e., activity fluctuations and complementary cumulative distributions of low-activity durations (CDDs)] deviate from fractal scaling over the entire time range from about 1 min to about 10 h for our sample of 36 geriatric in-patients with Alzheimer’s disease (AD)
Summary
Many physiological signals yield fractal characteristics, i.e., finer details at higher magnifications resemble details of the whole. Methodological issues need to be cautiously taken into account in order to be able to provide reliable as well as valid interpretations of such signal analyses It has been recognized in the cognitive sciences that multicomponent systems with a high degree of interaction and feedback give rise to emergent signals which yield fractal scaling indicating self-similarity or rather self-affinity in the case of time series data (Holden et al, 2009; Kello et al, 2010). Human locomotor activity assessed by wrist-actigraphy has repeatedly been found to yield fractal regulation (Hu et al, 2004, 2009, 2013, 2016; Ohashi et al, 2004; Nakamura et al, 2007, 2013a, 2016; Paraschiv-Ionescu et al, 2008; Aybek et al, 2012; Sano et al, 2012)
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
