Steffi Kreuzfeld

Fuzzy Evaluation of Heart Rate Signals for Mental Stress Assessment

Mental stress is accompanied by dynamic changes in autonomic nervous system (ANS) activity. Heart rate variability (HRV) analysis is a popular tool for assessing the activities of autonomic nervous system. This paper presents a novel method of HRV analysis for mental stress assessment using fuzzy clustering and robust identification techniques. The approach consists of 1) online monitoring of heart rate signals, 2) signal processing (e.g., using the continuous wavelet transform to extract the local features of HRV in time-frequency domain), 3) exploiting fuzzy clustering and fuzzy identification techniques to render robustness in HRV analysis against uncertainties due to individual variations, and 4) monitoring the functioning of autonomic nervous system under different stress conditions. Our experiments involved 38 physically fit subjects (26 male, 12 female, aged 18-29 years) in air traffic control task simulations. The subjective rating scores of mental workload were assessed using NASA task load index. Fuzzy clustering methods have been used to model the experimental data. Further, a robust fuzzy identification technique has been used to handle the uncertainties due to individual variations for the assessment of mental stress.

Heart Rate Variability and Blood Pressure during Dynamic and Static Exercise at Similar Heart Rate Levels

Aim was to elucidate autonomic responses to dynamic and static (isometric) exercise of the lower limbs eliciting the same moderate heart rate (HR) response. Method: 23 males performed two kinds of voluntary exercise in a supine position at similar heart rates: static exercise (SE) of the lower limbs (static leg press) and dynamic exercise (DE) of the lower limbs (cycling). Subjective effort, systolic (SBP) and diastolic blood pressure (DBP), mean arterial pressure (MAP), rate pressure product (RPP) and the time between consecutive heart beats (RR-intervals) were measured. Time-domain (SDNN, RMSSD), frequency-domain (power in the low and high frequency band (LFP, HFP)) and geometric measures (SD1, SD2) as well as non-linear measures of regularity (approximate entropy (ApEn), sample entropy (SampEn) and correlation dimension D2) were calculated. Results: Although HR was similar during both exercise conditions (88±10 bpm), subjective effort, SBP, DBP, MAP and RPP were significantly enhanced during SE. HRV indicators representing overall variability (SDNN, SD 2) and vagal modulated variability (RMSSD, HFP, SD 1) were increased. LFP, thought to be modulated by both autonomic branches, tended to be higher during SE. ApEn and SampEn were decreased whereas D2 was enhanced during SE. It can be concluded that autonomic control processes during SE and DE were qualitatively different despite similar heart rate levels. The differences were reflected by blood pressure and HRV indices. HRV-measures indicated a stronger vagal cardiac activity during SE, while blood pressure response indicated a stronger sympathetic efferent activity to the vessels. The elevated vagal cardiac activity during SE might be a response mechanism, compensating a possible co-activation of sympathetic cardiac efferents, as HR and LF/HF was similar and LFP tended to be higher. However, this conclusion must be drawn cautiously as there is no HRV-marker reflecting “pure” sympathetic cardiac activity.

Fuzzy Techniques for Subjective Workload-Score Modeling Under Uncertainties

This paper deals with the development of a computer model to estimate the subjective workload score of individuals by evaluating their heart-rate (HR) signals. The identification of a model to estimate the subjective workload score of individuals under different workload situations is too ambitious a task because different individuals (due to different body conditions, emotional states, age, gender, etc.) show different physiological responses (assessed by evaluating the HR signal) under different workload situations. This is equivalent to saying that the mathematical mappings between physiological parameters and the workload score are uncertain. Our approach to deal with the uncertainties in a workload-modeling problem consists of the following steps: 1) The uncertainties arising due the individual variations in identifying a common model valid for all the individuals are filtered out using a fuzzy filter; 2) stochastic modeling of the uncertainties (provided by the fuzzy filter) use finite-mixture models and utilize this information regarding uncertainties for identifying the structure and initial parameters of a workload model; and 3) finally, the workload model parameters for an individual are identified in an online scenario using machine learning algorithms. The contribution of this paper is to propose, with a mathematical analysis, a fuzzy-based modeling technique that first filters out the uncertainties from the modeling problem, analyzes the uncertainties statistically using finite-mixture modeling, and, finally, utilizes the information about uncertainties for adapting the workload model to an individuals physiological conditions. The approach of this paper, demonstrated with the real-world medical data of 11 subjects, provides a fuzzy-based tool useful for modeling in the presence of uncertainties.

Fuzzy Filtering for Physiological Signal Analysis

This study suggests the use of fuzzy-filtering algorithms to deal with the uncertainties associated to the analysis of physiological signals. The signal characteristics, for a given situation or physiological state, vary for an individual over time and also vary among the individuals with the same state. These random variations are due to the several factors related to the physiological behavior of individuals, which cannot be taken into account in the interpretation of signal characteristics. Our approach is to reduce the effect of random variations on the analysis of signal characteristics via filtering out randomness or uncertainty from the signal using a nonlinear fuzzy filter. A fuzzy-filtering algorithm, which is based on a modification of filtering algorithm of Kumar et al. [M. Kumar, N. Stoll, and R. Stoll, IEEE Trans. Fuzzy Syst., vol. 17, no. 1, pp. 150-166, Feb. 2009], is proposed for an improved performance. The method is illustrated by studying the effect of head-up tilting on the heart-rate signal of 40 healthy subjects.

Tri-axial high-resolution acceleration for oxygen uptake estimation: Validation of a multi-sensor device and a novel analysis method.

We validated a multi-sensor chest-strap against indirect calorimetry and further introduced the total-acceleration-variability (TAV) method for analyzing high-resolution accelerometer data. Linear regression models were developed to predict oxygen uptake from the TAV-processed multi-sensor data. Individual correlations between observed and TAV-predicted oxygen uptake (V̇O2) were strong (mean r = 0.94) and bias low (1.5 mL·min(-1)·kg(-1), p < 0.01; 95% confidence interval: 8.7 mL·min(-1)·kg(-1); -5.8 mL·min(-1)·kg(-1)); however, caution should be taken when a single-model value is used as a surrogate for V̇O2.

Flexible R&D integration platform of process informatics for automated medical applications and mobile data acquisition

In medical studies, an increasingly decentralized automated information acquisition occurs. Innovations assume for example a multi-parameter mobile sensor system combined with compact mobile computers (such as PDAs) communicating via Bluetooth or other wireless interfaces. Data sets from manual investigations and automated examination procedures arise in distributed systems. This results in above-average requirements of configuration for medical R&D applications. On the other hand in medical research projects involving test persons, there is a need for accumulating data from various examinations, for selecting data sets according to any criteria and for using the data on different targets. Fulfilling these demands, an open Web-based integration platform was developed. It is based on a workflow-oriented information management for laboratory and examination characteristic processes, picking up experiences of flexible, hierarchical laboratory automation. A process mapping and communication framework is proposed for workflow documentation as well as systems coupling with mobile medical data acquisition. It can be freely configured for application in medical examination by including process automation. Challenges of a universal, decentralized acquisition of arbitrary even time-based process parameters are archived by using service-oriented communication protocols. An exemplary project presentation illustrates potential and advantages of the developed integration platform.

System for Flexible Field Measurement of Physiological Data of Operators Working in Automated Labs

The strain of manual, automated, and semi-automated tasks in high-throughput chemical screening procedures offer new challenges for occupational and preventive medicine. Therefore, field inquiries need to measure several physiological parameters to determine amounts of strain. These include blood pressure, heart rate, and breathing parameters. These parameters can be registered by specialized mobile devices and sensor systems, but before they can be evaluated, it is essential to understand the activity patterns that represent the workloads. Activity status is gathered by those being tested with the help of a Nokia 3660 cell phone. A self-developed software module runs on the mobile phone and allows users to enter an activity setting and any subjective conditions via a short questionnaire. All data are stored together with a time stamp. After finishing an experiment (i.e., completing a working day), the collected data are transmitted to a central data server. Activity data and physiological data are stored in a central database in a merged data set. Access is provided via an Intranet or Internet connection. A graphical visualization tool, which can be shared from any location via Internet, helps to analyze the experimental data. Occupational medicine field studies in the working environment of chemical laboratories benefit from the data acquisition system and especially from the semiautomated self-monitoring.

A fuzzy filtering based system for maximal oxygen uptake prediction using heart rate variability analysis

An attempt has been made to predict the maximum oxygen uptake (VO2max) of an individual from a submaximum load test. The aim is to predict the VO2max from an analysis of the heart rate signal collected during the first 3 minutes of an incremental bicycle ergometer exercise test. The features of the heart rate signal are extracted in frequency domain using the continuous wavelet transform. The relationships between the signal features and corresponding VO2max are complicated by the uncertainties arising due to several factors (e.g. age, gender, etc.) related to individual behavior. A fuzzy filter is used for separating uncertainties from the modeling problem and then assessing the worst effect of uncertainties on the predicted VO2max. The method is studied with the experimental data of 49 subjects (24 females, 25 males, ag ed 28–56 years).

24-Hour ambulatory monitoring of complex physiological parameters with a wireless health system: Feasibility, user compliance and application

In this paper, the feasibility of a wireless health system will be presented. The developed system is based on a sensor electronic module, a smartphone and a process management system that handles expert models for individual health estimation and delivers results from the primary data. The system was evaluated with experiments on N = 128 volunteers who were monitored in real-world settings. Ecological momentary assessment was triggered automatically by the smartphone several times per day. All participants confirmed the feasibility and ease of use of the developed system.

Comment on "Cytokines and oxidative stress status following a handball game in elite male players".

In a well-conducted study, Marin et al. [1] reported significant alterations of oxidative stress biomarkers, antioxidant capacity, and indices of muscular damage in elite handball players after a friendly match. The authors were surprised by the marked increase of muscular damage indirectly assessed by serum creatine kinase (CK) in experienced players, which increased from about 80 U/L at baseline to 150 U/L 24 hours after game. However, average CK values reported here are quite low in comparison to reference values of athletic populations [2] and do not exceed reference values used in clinical practice [3, 4]. Although containing lots of eccentric exercises like abrupt stopping or landing after jumping and the risk of muscle injury due to direct contact with other players, the average 24-hour postmatch value reported by Marin and colleagues reached only 20% of the upper reference level of swimmers. Swimmers generally exhibit low CK levels because of the non-weight-bearing, noncontact, and concentric nature of their sport [2, 5, 6]. We investigated twenty-one elite handball players to obtain representative values during a regular play-off season and found 12-hour postmatch CK values of 347 U/L (SEM: 43 U/L) and values of 255 U/L (SEM: 38 U/L) during a normal training week (60 hours after match). The relatively low values reported by Marin et al. are due to the study design, which included the abstinence from handball training and games before a friendly match for 2 and 4 days, respectively. Furthermore, serum CK is not always an (indirect) marker of muscular damage, but rather reflects increased rates of muscle turnover, stimulated by muscle use [6–8]. Thus, a 2-fold and even higher increase in response to exercise is not surprising [6, 9, 10].