deFOG--A real time system for detection and unfreezing of gait of Parkinson's patients.

Freezing of gait (FOG) is a common complication in movement disorders, typically associated with the advanced stages of Parkinson's disease. Auditory cues might be used to facilitate unfreezing of gait and prevent fall related injuries. We present a wearable, unobtrusive system for real-time gait monitoring, which consists of an inertial wearable sensor and wireless headset for the delivery of acoustic cues. The system recognizes FOG episodes with minimum latency and delivers acoustic cues to unfreeze the gait. We present design of a system for the detection and unfreezing of gait (deFOG), and preliminary results of the feasibility study. In a limited test run of 4 test cases the system was able to detect freezing of gait with average latency of 332 ms, and maximum latency of 580 ms.

Avatar - a multi-sensory system for real time body position monitoring.

Virtual reality and computer assisted physical rehabilitation applications require an unobtrusive and inexpensive real time monitoring systems. Existing systems are usually complex and expensive and based on infrared monitoring. In this paper we propose Avatar, a hybrid system consisting of off-the-shelf components and sensors. Absolute positioning of a few reference points is determined using infrared diode on subject's body and a set of Wii Remotes as optical sensors. Individual body segments are monitored by intelligent inertial sensor nodes iSense. A network of inertial nodes is controlled by a master node that serves as a gateway for communication with a capture device. Each sensor features a 3D accelerometer and a 2 axis gyroscope. Avatar system is used for control of avatars in Virtual Reality applications, but could be used in a variety of augmented reality, gaming, and computer assisted physical rehabilitation applications.

Alterations in human EEG activity caused by extremely low frequency electromagnetic fields.

This study has investigated whether extremely low frequency (ELF) electromagnetic fields (EMFs) can alter human brain activity. Linearly polarised magnetic flux density of 20 muT (rms) was generated using a standard double Helmholtz coils and applied to the human head over a sequence of 1 minute stimulations followed by one minute without stimulation in the following order of frequencies 50, 16.66, 13, 10, 8.33 and 4 Hz. We collected recordings on 33 human volunteers under double-blind counter-balanced conditions. Each stimulation lasted for two minutes followed by one minute post-stimulation EEG recording. The same procedure was repeated for the EMF control sessions, where the order of control and exposure sessions was determined randomly according to the subject's ID number. The rest period between two conditions (exposure and control) was 30 minutes. The results indicate that there was a significant increase in Alpha1, Alpha2, and Beta1 at the frontal brain region, and a significant decrease in Alpha2 band in parietal and occipital region due to EMF exposure.

A WBAN System for Ambulatory Monitoring of Physical Activity and Health Status: Applications and Challenges.

Recent technological advances in sensors, low-power integrated circuits, and wireless communications have enabled the design of low-cost, miniature, lightweight, intelligent physiological sensor platforms that can be seamlessly integrated into a body area network for health monitoring. Wireless body area networks (WBANs) promise unobtrusive ambulatory health monitoring for extended periods of time and near real-time updates of patients' medical records through the Internet. A number of innovative systems for health monitoring have recently been proposed. However, they typically rely on custom communication protocols and hardware designs, lacking generality and flexibility. The lack of standard platforms, system software support, and standards makes these systems expensive. Bulky sensors, high price, and frequent battery changes are all likely to limit user compliance. To address some of these challenges, we prototyped a WBAN utilizing a common off-the-shelf wireless sensor platform with a ZigBee-compliant radio interface and an ultra low-power microcontroller. The standard platform interfaces to custom sensor boards that are equipped with accelerometers for motion monitoring and a bioamplifier for electrocardiogram or electromyogram monitoring. Software modules for on-board processing, communication, and network synchronization have been developed using the TinyOS operating system. Although the initial WBAN prototype targets ambulatory monitoring of user activity, the developed sensors can easily be adapted to monitor other physiological parameters. In this paper, we discuss initial results, implementation challenges, and the need for standardization in this dynamic and promising research field.

On Spectral Analysis of Heart Rate Variability during Very Slow Yogic Breathing.

Very slow yogic breathing techniques provide valuable insights into mechanisms of autonomous nervous system regulation that are usually not available for human subjects. This paper presents results of eight sessions of Nadi Shodhana Pranayama practiced at rate of one breath per minute. We characterized statistic and spectral measures of heart rate variability before, during, and after exercises. Significant changes include increase of VLF frequencies caused by slow breathing and decrease in average interbeat interval from 959.3 to 904.1 ms (t(7) = -7.5, p<0.001). We present the results of HRV analysis and analyze origins of characteristic frequency components. The most prominent changes of the exercise include significant increase of respiratory sinus arrhythmia (RSA) and LF/HF ratio, and decrease of breathing frequency after the exercise against the state before the exercise. The maximum LF frequency decreased from 0.0919 Hz to 0.07125 Hz (t(7) = -3.255, p < 0.01), indicating the decrease of average breathing rhythm from 5.5 breaths/min to 4.3 breaths/min. In addition, the state after the exercise is characterized by disappearance of VLF frequencies from the spectrum, and a significant increase of LF/HF from 14.33 to 50.93 (t(7) = 2.461, p <.05).

Reconfigurable intelligent sensors for health monitoring: a case study of pulse oximeter sensor.

Design of low-cost, miniature, lightweight, ultra low-power, intelligent sensors capable of customization and seamless integration into a body area network for health monitoring applications presents one of the most challenging tasks for system designers. To answer this challenge we propose a reconfigurable intelligent sensor platform featuring a low-power microcontroller, a low-power programmable logic device, a communication interface, and a signal conditioning circuit. The proposed solution promises a cost-effective, flexible platform that allows easy customization, run-time reconfiguration, and energy-efficient computation and communication. The development of a common platform for multiple physical sensors and a repository of both software procedures and soft intellectual property cores for hardware acceleration will increase reuse and alleviate costs of transition to a new generation of sensors. As a case study, we present an implementation of a reconfigurable pulse oximeter sensor.

Patient monitoring using personal area networks of wireless intelligent sensors.

A wearable device for monitoring multiple physiological signals (polysomnograph) usually includes multiple wires connecting sensors and the monitoring device. In order to integrate information from intelligent sensors, all devices must be connected to a Personal Area Network (PAN). This system organization is unsuitable for longer and continuous monitoring, particularly during the normal activity. For instance, monitoring of athletes and computer assisted rehabilitation commonly involve unwieldy wires to arms and legs that restrain normal activity. We propose a wireless PAN of intelligent sensors as a system architecture of choice, and present a new design of wireless personal area network with physiological sensors for medical applications. Intelligent wireless sensors perform data acquisition and limited processing. Individual sensors monitor specific physiological signals (such as EEG, ECG, GSR, etc.) and communicate with each other and the personal server. Personal server integrates information from different sensors and communicates with the rest of telemedical system as a standard mobile unit. We present our prototype implementation of Wireless Intelligent SEnsor (WISE) based on a very low power consumption microcontroller and a DSP-based personal server. In future we expect all components of WISE integrated in a single chip for use in a variety of new medical applications and sophisticated human computer interfaces. Existing growth of wireless infrastructure will allow a range of new telemedical applications that will significantly improve the quality of health care.

Thermistor-based breathing sensor for circadian rhythm evaluation.

One of the most frequently used methods to sense breathing pattern is to detect airflow using a nasal thermistor or a thermocouple sensor. Prolonged, minimally intrusive measurement of the breathing pattern is particularly important for polysomnography, sleeping disorders, stress monitoring, biofeedback techniques, and circadian rhythm analysis. Although most applications require only breathing pattern, some applications and diagnostic procedures require monitoring of the rhythm of change of dominant nostril. In this paper we present our design of a differential thermistor-based breathing sensor for prolonged monitoring during the normal activity. The system is designed using a low power microcontroller Texas Instruments MSP430F149 with an on-chip A/D converter for data acquisition and signal processing. We use wireless RF link to a PC for long-term data acquisition and storage. Precise measurement requires decreasing zero and sensitivity errors of the measurement. We discuss signal processing methods, calibration and parameters used to characterize breathing patterns necessary for circadian breathing rhythm evaluation.

Tactical audio and acoustic rendering in biomedical applications.

Complexity of biomedical data requires novel sophisticated analysis and presentation methods. Sonification is used as a new information display in augmented reality systems to overcome problems of existing human-computer interface (e.g., opaque or heavy head-mounted displays, slow computer graphics, etc.). A novel taxonomy of sonification methods and techniques is introduced. We present our experience with tactical audio and acoustic rendering in biomedical applications. Tactical audio as an audio feedback is used as support for precise manual positioning of a surgical instrument in the operating room. Acoustic rendering is applied as an additional information channel and/or warning signal in biomedical signal analysis and data presentation.