Advances in Science and Technology Vol. 85

Abstract: The interconnection of electronics and textile circuits is still a main challenge for the fabrication of reliable smart textiles. This paper investigates the thermoplastic adhesive bonding technology. Electronic modules are bonded to textile substrates with a thermoplastic non-conductive adhesive (NCA) film. The modules are placed onto textile circuits with an NCA-film in-between. By applying pressure and heat, the adhesive melts and contact partners touch. Subsequently cooling solidifies the NCA resulting in an electrical and mechanical contact of the electronic module and the textile circuit. This paper shows the suitability of this technology for knitted, woven, non-woven and embroidered fabrics with metal coated yarns as well as with litz-wires as conductors. Besides, it shows that the interconnection process works well with thermoplastically insulated conductors. In addition, the design of interposers has been improved in respect to contact formation and miniaturization compared to previous publications. The pitch of the contact pads is set to 1.27 mm. A textile display was realized with smart RGB-Pixels, which are controlled by an I²C-bus on a quadrupolic woven substrate. It demonstrates the applicability and the potential of this technology.

Abstract: A Remote Brain Machine Interface (RBMI) can be defined as a means to control a machine that is in a different geographical location than the user. Thus far, simulations for such interfaces using multiple channels of non-invasive EEG signals acquired through tethered systems have been used for control of vehicles in military and exploratory applications, and for ongoing research on RBMI controlled robotic surgery. However, simple applications of RBMI in home automation for the elderly, low cost assistive devices for the disabled, home security etc can be built using fewer and more portable sensor systems. As a case study, we have implemented such an interface using a smartphone for the RBMI. The system consists of a wearable Bluetooth-enabled head band with dry electrodes for EEG and EOG signals, a smartphone to collect and relay the data, a laptop with internet connectivity at a remote location to retrieve the data and generate control commands. In this paper, we describe the information architecture, the design of the wearable nanosensors and algorithms for control command generation based on EEG and EOG. A selected demonstration will be shown.

Abstract: The average age of European population is growing year by year and the social impact of this trend will affect the future economy of developed countries. Since elderly people suffer from many chronic diseases that limit their capabilities; home assistance is one of the main emerging needs. One common aspect of advanced age is susceptibility to hypothermia due to several age-related physiological factors such as lower metabolism, thinner skin, malnutrition, and delayed perception of hot/cold. Hypothermia can lead to mortality during the night in elderly affected by diseases or disabilities. In this paper we present a pajama that aims to monitor elderly people while sleeping. Such a system, developed in the scope of EU funded project MOBISERV, is able to monitor a 1 lead ECG signal, upper body movements and skin temperature from 2 different sites through a distributed Body Sensor network, and communicate to a main data logger through ANT wireless protocol. The ultimate goal of the system is to prevent the onset of hypothermia by unobtrusively monitoring physiological parameters of a subject and raising alarms if wrong patterns are detected.

Abstract: As part of the goal of developing wearable sensor technologies, we have designed and built a hybrid sensor headset for monitoring brain activity. Through the use of electroencephalography (EEG) and near-infrared spectroscopy (NIRS), the sensor array is capable of monitoring neural activity across the primary motor cortex and wirelessly transmitting data to a computer for real-time processing to generate control signals, which are transmitted to wireless devices for various applications. This paper focuses on current results using this technology for artificial limb control and discusses the development of the headset as well as the neural networks employed for processing motor cortex activity and determining the user’s intentions. Initial results relevant to artificial limb control are presented and discussed, including the performance of the system when actuating an artificial limb with four degrees of freedom. Our headset provides a more natural control mechanism than traditional solutions, through the use of direct brain control. The technology resulting from this research is currently also being investigated for application in areas including phantom limb pain treatment, robotic arm control, general brain-computer interfaces, lie detection, and even a video game interface.

Abstract: For the aged society, the physical activity of daily living is important to improve the quality of life. The simple quantitative evaluation of physical activity as well as rehabilitation is required We have developed wearable inertia sensors as well as evaluating system. We evaluated parameters such as RMS and autocorrelation function of stride and step for common neuro-physiological test in rehabilitation. The normal and fall-risk subjects were performed the 10 m trial. Furthermore obtained data from the wearable motion sensor were compared to the general estimation parameters such as performing time and Activities of Daily Living (ADL) score. In 10 meter trial, the result indicated that the ADL score is weakly correlated to the RMS of acceleration signal. However, the relationship between walking speed and RMS was highly correlated. The ADL score is general daily living activities and the walking is one of daily activities. The walking speed ,RMS and autocorrelation function of step and stride were significantly different between normal and fall-risk subjects. In conclusion, the obtained acceleration and angular velocity signals may help us the evaluation of daily activities and rehabilitation training quantitatively.

Abstract: This paper describes a new implantable measurement system, further developed from an implantable measurement device implemented earlier at Tampere University of Technology (TUT), to assess the psychophysiological state of well-being of dairy-cattle. By measuring single-channel Electrocardiogram (ECG), body temperature and cattle activity we provide veterinarians and animal scientists with a tool to assess cattle stress levels and well-being. This information can, for example, be linked to cattle milk production. The new device processes the ECG signal in real-time to derive the heart rate along with a wireless radio frequency transmission of the data from the implant to a receiver device attached near or on the cattle. By collecting the data through a wireless link we are able to extend the recording period from three weeks of the earlier version of the device up to 3 months. The algorithm for the ECG peak detection is a modified version of the Pan-Tompkins algorithm optimized for cattle application with ECG recordings based on several hundred hours of data collected from previous experiments and recordings. The implantable system was tested with in vivo trials in April 2012.

Abstract: In this paper, we propose a new biological signal-responsive implantable device that triggers direct an anticonvulsive drug into the epileptogenic zone at electrographic seizure onset. We describe the high-performance seizure-onset detection algorithm, low-power circuit technique and focal drug delivery system. The implantable device is composed of a preamplifier, a signal processor, a seizure detector and a micropump. The device records high quality intracerebral electroencephalographic (icEEG) signals using high conductive electrodes and a low noise preamplifier. The recorded signal is processed continuously using low-power technique to detect onset of seizures accurately. The low-power miniaturized micropump is able to deliver sufficient amount of anticonvulsive drug in a short duration (50µL/sec) to epileptogenic zone. The detection algorithm was validated with Matlab tools and a prototype device was assembled with discrete components in a circular (Ø 40 mm) printed circuit board. The device was validated offline using the icEEG recordings obtained from 3 drug-resistant epilepsy patients. The average seizure detection delay was 10 sec from electrographic seizure onset, well before seizure progression to adjacent functional cortex.

Abstract: With the rapid development of micro systems technology and microelectronics, smart implantable wireless electronic systems are emerging for the continuous surveillance of relevant parameters in the body and even for closed-loop systems with a sensor feed-back to drug release systems. With respect to diabetes management, there is a critical societal need for a fully integrated sensor array that can be used to continuously measure a patient’s blood glucose concentration, pH, pCO2 and colloid oncotic pressure twenty four hours a day on a long-term basis. In this work, thin films of metabolite-specific or “smart” hydrogels were combined with microfabricated piezoresistive pressure transducers to obtain “chemomechanical sensors” that can serve as selective and versatile wireless biomedical sensors and sensor arrays for a continuous monitoring of several metabolites. Sensor response time and accuracy with which sensors can track gradual changes in glucose, pH, CO2 and ionic strength, respectively, was estimated in vitro using simulated physiological solutions. The biocompatibility and hermeticity of the developed multilayer encapsulation for the microsensor array has been investigated concerning the long-term stability and enduring functionality that is desired for permanent implants.

Abstract: Wireless sensor networks (WSNs) are becoming more and more attractive because of their flexibility. In particular, WSNs are being applied to a user body in order to monitor and detect some activities of daily living (ADL) performed by the user (e.g., for medical purposes). This class of WSNs are typically denoted as body sensor networks (BSNs). In this paper, we discuss BSN-based human activity classification. In particular, the goal of our approach is to detect a sequence of activities, chosen from a limited set of fixed known activities, by observing the outputs generated by accelerometers and gyroscopes at the sensors placed over the body. In general, our framework is based on low-complexity windowing-&-classification. First, we consider the case of disjoint (in the time domain) activities; then, we extend our approach to encompass a scenario with consecutive non-disjoint activities. While in the first case windowing is separate from classification, in the second case windowing and classification need to be carried out jointly. The obtained results show a significant detection accuracy of the proposed method, making it suitable for healthcare monitoring applications.