Nonlinear Analysis of Motor Activity Shows Differences between Schizophrenia and Depression: A Study Using Fourier Analysis and Sample Entropy

Ole Bernt Fasmer,Erik R. Hauge,Jan Øystein Berle,Fred Holsten,Ketil J. Oedegaard

doi:10.1371/journal.pone.0016291

Ole Bernt Fasmer, Erik R. Hauge + Show 3 more

Open Access

https://doi.org/10.1371/journal.pone.0016291

Copy DOI

Journal: PLoS ONE	Publication Date: Jan 28, 2011
Citations: 73	License type: CC BY 4.0

Affiliation: Haukeland University Hospital, University of Bergen

Abstract

The purpose of this study has been to describe motor activity data obtained by using wrist-worn actigraphs in patients with schizophrenia and major depression by the use of linear and non-linear methods of analysis. Different time frames were investigated, i.e., activity counts measured every minute for up to five hours and activity counts made hourly for up to two weeks. The results show that motor activity was lower in the schizophrenic patients and in patients with major depression, compared to controls. Using one minute intervals the depressed patients had a higher standard deviation (SD) compared to both the schizophrenic patients and the controls. The ratio between the root mean square successive differences (RMSSD) and SD was higher in the schizophrenic patients compared to controls. The Fourier analysis of the activity counts measured every minute showed that the relation between variance in the low and the high frequency range was lower in the schizophrenic patients compared to the controls. The sample entropy was higher in the schizophrenic patients compared to controls in the time series from the activity counts made every minute. The main conclusions of the study are that schizophrenic and depressive patients have distinctly different profiles of motor activity and that the results differ according to period length analysed.

Highlights

Assessing the motor activity of the patients has always been an integral part of a psychiatric evaluation
The motor activity was significantly lower in the schizophrenic patients and in patients with major depression, compared to controls (Table 4, Figures 1, 2, 3, 4, 5 for 300 min periods, and Table 5, Figures 6, 7, 8, 9, 10, 11 for 2 weeks periods)
As described previously [22] there was a significantly reduced motor activity in both patient groups compared to the controls

Summary

Introduction

Assessing the motor activity of the patients has always been an integral part of a psychiatric evaluation. The possibility of an objective registration of the motor activity as it varies over hours and days seems not to have lead to its widespread use in clinical practice, neither in diagnosis nor for monitoring symptom severity or treatment effect. Methods from nonlinear dynamics have been applied to a number of subjects encompassing numerous areas in physiology and clinical medicine [4] [5] [6] [7] [8] [9] [10] These methods range from studying the randomness versus chaos by measuring the nonlinearity in a signal to estimating the fractal dimension of its corresponding attractor. Basic information of nonlinear analysis as well as various purposes of applications can be found in the references mentioned above

Methods

Results

Conclusion