Improved CEEMDAN Based Speech Signal Analysis Algorithm for Mental Disorders Diagnostic System: Pitch Frequency Detection and Measurement

Improved CEEMDAN Based Speech Signal Analysis Algorithm for Mental Disorders Diagnostic System: Pitch Frequency Detection and Measurement

Alan K. Alimuradov (Penza State University, Penza, Russia), Alexander Yu. Tychkov (Penza State University, Penza, Russia), Andrey V. Kuzmin (Penza State University, Penza, Russia), Pyotr P. Churakov (Penza State University, Penza, Russia), Alexey V. Ageykin (Penza State University, Penza, Russia) and Galina V. Vishnevskaya (Penza State University, Penza, Russia)
DOI: 10.4018/IJERTCS.2019010102

Abstract

An automated algorithm for pitch frequency measurement for diagnostic systems of borderline mental disorders is developed. It is based on decomposition of a speech signal into frequency components using an adaptive method for analyzing of non-stationary signals, improved complete ensemble empirical mode decomposition with adaptive noise (improved CEEMDAN), and isolating the component containing pitch. A block diagram for the developed algorithm and a detailed mathematical description are presented. A research of the algorithm using the formed verified signal base of healthy patients, and male and female patients with psychogenic disorders, aged from 18 to 60, is conducted. The research results are evaluated in comparison with the known algorithms for pitch frequency measurement. In accordance with the results of the study, the developed algorithm for pitch frequency measurement provides an accuracy increase in determination of borderline mental disorders: for the error of the first kind, on the average, it is more accurate by 10.7%, and for the second type error by 4.7%.
Article Preview

Introduction

Currently, an assessment of human mental health is a socially significant problem for every state, since it is directly related with the formation of a healthy lifestyle of the population. According to the World Health Organization, current socially significant diseases, which are the main cause of temporary disability, invalidity and mortality, negatively affecting the socioeconomic factors of the state development, are directly related to the mental health of the population (Alimuradov, 2015).

An assessment of borderline mental disorders is of particular importance in those branches of human activity that involve an increased risk to human life and the risk of economic consequences (operators of control systems with a high degree of responsibility: pilots, astronauts, servicemen, airport dispatchers, dispatchers of hazardous production facilities, e.g., nuclear power plants, thermal power plants, chemical industry facilities, etc.).

In the last decades, the research in the field of mental (psycho-emotional) state assessment has been actively supported by international funds and grants of organizations: Remote Assessment of Disease and Relaps in Central Nervous System Disorders, RADAR CNS (#115902), Foundation/Grant Organization: EU H2020 / EFPIA Innovative Medicines Initiative (IMI) (Shuller, 2017); Emotion Sensitive Assistance System, EmotAsS (#16SV7213), Foundation/Grant Organization: BMBF IKT2020-Grant (Sozial- und emotionssensitive Systeme für eine optimierte Mensch-Technik-Interaktion) (Fraunhofer, 2015); Promoting Early Diagnosis of Rett Syndrome through Speech-Language Pathology, (#16430), Foundation/Grant Organization: Österreichische Nationalbank (OeNB) Jubiläumsfonds (OeNB Project P16430, 2015).

Currently, various experimental and statistical techniques and differentiation of signal processing methods on accessible recording channels of the human body reactions are used for the detection of borderline mental disorders. Methods for evaluation, implemented on the basis of video data reflecting mimic and gestural changes (Alimuradov, Tychkov, Ageykin, Churakov, Kvitka, & Zaretskiy, 2017; Alimuradov, 2017); signals reflecting parameters of physiological activity of a human body (electroencephalography, electrocardiography, electromyography, etc.) (Alimuradov, Tychkov, Frantsuzov, & Churakov, 2015; Agrafioti, 2011; Barabanschikov & Zhegallo, 2014); biochemical blood parameters (Bobkov, 2013; Camacho, & Harris, 2008); parameters of handwriting and keyboard writing of texts (Cheveigne, & Kawahara, 2002; Colominasa, Schlotthauera, & Torres, 2014); parameters of oculography (eye tracking) (Darley, Aronson, & Brown, 1969; Davydov, Kiselev, Kochetkov, & Tkachenya, 2011) are of particular interest.

An essential shortcoming limiting a wide practical application of these methods is the obligatory condition of contact recording/sampling/writing, which certainly affects the mental state, which it is no longer possible to effectively evaluate. The most promising and adaptive (automated in real time and free activity) is the method based on the analysis of speech signals (SS) (Dorry, 2016; Eun-Joo, & Kwang-Seok, 2015; Fant, 1964).

Nowadays, the market of speech technologies is represented by commercial voice analysis systems for assessing human psychoemotional state (mental disorders). The highest practical popularity has been achieved by lingWAVES software and hardware modular complex (WEVOSYS, Germany) (lingWaves, 2014), and Sense software and hardware complex (NEMESYSCO, Israel) (Nemesysco, 2016). The greatest interest in these systems is represented by automated algorithms and the used methods for speech signal processing. However, due to trade secrets, manufacturers do not provide such information. Hence, the development of new automated algorithms for speech signal processing in real time, which raise system effectiveness for diagnosing borderline mental disorders, is urgent. The fundamentals of proposed solutions are given in (Alimuradov, A., Tychkov, A., Kuzmin, A., Churakov, P., Ageykin, A., & Vishnevskaya, G. 2017).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 11: 4 Issues (2020): Forthcoming, Available for Pre-Order
Volume 10: 4 Issues (2019): 2 Released, 2 Forthcoming
Volume 9: 2 Issues (2018)
Volume 8: 2 Issues (2017)
Volume 7: 2 Issues (2016)
Volume 6: 2 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing