A noncontact input interface using image processing for people with disabilities of the fingers, such as rheumatism, collagen disease, and osteoarthritis, has been developed. The input interface is simply composed of a CCD camera and PC. Taken with the CCD camera, the fingers were appropriately detected using image processing in this paper. An on-screen keyboard and control box were also made. In order to evaluate a performance of the input interface, three healthy subjects input some example sentences using it. From the experimental results, it was clear that the beginners could directly input them in Microsoft Word without any difficulty, and the input characteristics with the least error were experimentally shown in this paper.

View full abstract

We investigated NIR tomographic imaging based on the pulsed-ultrasound modulation of diffuse light field in turbid media theoretically and experimentally. The difficulty of the method is the very low S/N which cannot be overcome by the conventional signal-averaging because of the slowly-varying light phase on the detector. We proposed a simple model for the detected signal and validated it experimentally. Furthermore, we proposed the frequency-doubling-and-averaging method to recover the signal, and demonstrated its effectiveness through the phantom experiment. We present a 2-D tomographic image of the high-scattering inclusion in the scattering medium.

View full abstract

Daily health monitoring is very important to detect disease in early stage and to keep health. However, there are a lot of difficulties in it. The purpose of this study is to develop an easy and simple system that collects long-term vital sign (ECG in daily life) at home. In order to realize this, we aimed at downsizing and power saving by using reflective modulation for the ECG transmission system. In general, RFID (Radio Frequency IDentification) system uses ASK modulation, however, we employed analog modulation, PFM-AM to solve the problem that occurred in analog signal transmission. The variable impedance antenna, which we developed, allows analog signal transmission without transmitter like conventional RFID system. In order to show the availability of the proposed system, we performed the evaluation experiments with prototype system. It was shown from the experiments that the system could transmit ECG signal in real time, and that the power consumption for the tag type sensor was sufficiently low.

View full abstract

An optical method recently proposed for non-invasive in vivo blood glucose concentration (BGL) measurement, named “Pulse Glucometry”, was combined and compared with four multivariate analyses for constructing calibration models: Principal Component Regression (PCR), Partial Least Squares Regression (PLS), Artificial Neural Network (ANN), Support Vector Machines Regression (SVMsR). A very fast spectrophotometer for “Pulse Glucometry” provides the total transmitted radiation spectrum (I_λ) and the cardiac-related pulsatile component (ΔI_λ) superimposed on I_λ in human fingertips over a wavelength range from 900 to 1700 nm with resolution of 8 nm in 100 Hz sampling. From a family of I_λs measured, which include information relating to blood constituent such as BGL values, differential optical densities (ΔOD_λs, where ΔOD_λ=Log (1+ΔI_λ/I_λ)) were obtained and normalized by the ΔOD_λ values at 1100 nm. Finally, the 2nd derivatives of the normalized ΔOD_λs (Δ²OD_λs) along wavelengths were calculated as regressors. Subsequently, calibration models from paired data sets of regressors (the values of Δ²OD_λs) and regressand (the corresponding known BGL values) were constructed with PCR, PLS, ANN and SVMsR. The results show that each calibration model provides a relatively good regression with a modified 5-fold cross validation for total 95 paired data, in which the BGLs ranged from 100.7-246.3 mg/dl. The results were evaluated by the Clarke error grid analysis and all data points obtained from all calibration models fell within the clinically acceptable regions (region A or B). Among them, ANN and SVMsR calibration provided the best plot distributions (in ANN; Region A: 77 plots (81.1%), B: 18 plots (18.9%). in SVMsR; Region A: 78 (82.1%), B: 17 (17.9%)). Total calculation time of SVMsR is about 100 times shorter than ANN. These results suggest that a calibration model using SVMsR is highly promising for “Pulse Glucometry.”

View full abstract

We're developing walking robot with the object of bringing the walking of robot close to that of human utilizing pendular motion of legs. First, human gait analysis has been conducted in the study. In particular, we measured joint angle variation and myoelectric signal at a normal walking pace, focusing on the ankle mobility. From the result of the measurement, we realized that ankle mobility of supporting leg not only provides energy in the direction of travel by kicking back but also leads idling leg to swing forward. Then, we conducted modeling of ankle-driven walking and reproduced the motion using highly simplified model. As the result, walking driven only by ankle was realized in simulation. Therefore, it is expected that the system could not only be used for walking robot but also be applied to walking aids or artificial legs, because it reproduces a walking utilizing pendular motion of legs effectively due to the hip of free joint, and it is possible to reduce the number of actuators.

View full abstract

For application of a cultured muscle to medical engineering, both to control C2C12 myoblast differentiation and to investigate a contraction mechanism of C2C12 myotube are important. The muscle can be derived from a cultured C2C12 myoblast in an incubator. It is important to evaluate a level of C2C12 differentiation quantitatively. In this study, we propose an efficient evaluation method by cellular membrane potential measurement instead of the conventional methods that take time and are higher cost.

View full abstract

A living body is receiving quite a lot of information from the external world through sensitive organs for life-supporting. Recently, various physical sensors which measure physical quantity have been developed rapidly. On the other hand, chemical sensors have not developed enough and they are limited to a pH meter, an ion selective electrode and an enzyme sensor. It is difficult to develop sensors for the sense of smell and taste became they have to sense various characters concurrently from chemical substances. We have been studying a possibility of taste sensing using a single artificial lipid membrane. We found that the conductance and transmittance show different characteristics in taste solutions. And we have suggested a way of the taste evaluation by showing features of electrical and optical characteristics as response patterns on a six axis radar-chart. Furthermore, applying the principal component analysis to the response patterns, we have also shown them on the two dimensional coordinate. In this study, we measured the spectrum characteristics of membrane for the purpose of discrimination in the commercial drinks and the condiments. We defined autocorrelation functions in order to evaluate the spectrum characteristics and we utilized Mahalanobis distance as a way of discriminant analysis. We selected two autocorrelation coefficients (hereafter cited as ACC1 and ACC2). One is ACC1 which was put as the value of x-axis and the other is ACC2 which was put as the value of y-axis. We had plots represented from the two ACCs on the two dimensional coordinate. We had known groups obtained by some sample plots. Mahalanobis distances were measured between those plots and the center of gravities of the known groups. As a result, the commercial drinks and the condiments were classified correctly in each category. Therefore, this method was effective to classify category among the commercial drinks and the condiments.

View full abstract

For computer simulations of physiological experiments by physiological models, a machine-readable description of experimental protocols is useful. Although description formats for cell physiological models are already available, there is no specialized representation format for experimental protocols. Here, we propose an XML-based language, PEPML (Physiological Experimental Protocol Markup Language), and a simulation method of experimental protocols in PEPML. In the PEPML, conditions and procedures of an experimental protocol are procedurally described as a list of events, each of which consists of a condition for execution and an action to be executed. By specifying the references of variables in a protocol using an ontology, the protocol can be applied to various models without editing tasks. The proposed simulation method allows unambiguous numerical calculation of a procedural experimental protocol, which is applied to a declaratively represented model by interpretation of the protocol as a set of boundary conditions for the model variables. PEPML and the proposed method allow applying both multiple protocols to a single model and a single protocol to multiple models. This feature facilitates the efficient simulation for verification, comparison and utilization of physiological models.

View full abstract

The heart is an organ with highly complex structure and its pump function is affected by various factors. Many reports showed the strong relation between the excitation-propagation phenomenon in the heart and its pump function. But the experimental studies often failed to provide the mechanistic insights due to the complex cross-talks built in these processes. In this study, we focus on the relation between the activation time and the contraction of the tissue. The contraction simulation coupled with the excitation propagation simulation was performed using the finite element model of the ventricular ring. We performed the simulation with 7 types of stimulation with various pacing sites in the tissue, and compared the pump functions by the area ejection fraction. The results showed that when the activation time became long, the area ejection fraction became nonlinearly small. The results of this study showed the excitation propagation is one of the important parameter of the left ventricular wall motion.

View full abstract

In this paper, we proposed a new guidance scheme of the GPS-GIS-based speech guidance system using the adaptive fuzzy inference neural network (AFINN). Geographic Information System (GIS) calculates some attributes of user-to-landmark relations, so-called, landmark celebrity, landmark facade, landmark area, representative point of latitude and longitude (the center of gravity) of the landmark, user's latitude and longitude from GPS, and user's direction. These attributes were transformed to six basic elements, such as the celebrity, the facade relative bearing, the Euclidean distance between the landmark and user, the area size, the direction relative bearing and the user's individual data, respectively. The AFINN, that has the above six elements as the input data, calculates a level of importance (LOI) corresponding to a “cognitive distance”. We considered that acceptable guidance was achieved by using the LOI. We applied the analytic hierarchy process (AHP) to the leaning data of the AFINN. Therefore, the guidance system reflected user's preference, and obtained the guidance suitable for diverse users. Numerical simulations suggested that the AFINN should be available for determining the guidance satisfying each user. Finally, the visually impaired pedestrians evaluated the guidance system from the user's viewpoints.

View full abstract

We investigated effects of the temporal waveform of complex sounds, which were delivered repeatedly to subjects, on the latency and amplitude of N1m and P2m responses of auditory evoked fields. Two components appeared in early and late latencies in P2m response, which led us to analyze these components separately. It was found that the change in the periodic waveform of sounds, while the frequency spectrum was kept nearly constant, affected the amplitude of N1m but not P2m. The latency and adaptation, i.e., how the response amplitude was attenuated by repetition of stimulus sounds, were affected by the phase and periodicity of the waveform, in different manners for the two components of P2m. The results indicated that the N1m response was sensitive mostly to the periodicity, while the two P2m components were dependent both on the phase and periodicity of complex sounds. We discussed possible relation between the processing of sound features and the neural substrates of N1m and P2m responses.

View full abstract

It is very important for elderly people to maintain good walking ability for good quality of life. This paper focuses on the characteristics of balance function and walking in elderly and extracts the characteristic indices. Seven young to middle-aged subjects and seven elderly subjects were participated in this study. They were all paid volunteers with informed consent. The plantar pressure was measured by a pressure sensor plate at quiet standing posture with open eyes and closed eyes, functional reach test (FRT), forward lean and backward lean (FB), and walking. Maximum lengths between back and forth of center of plantar pressure at FB and at FRT in the elderly were shorter than those in the young to middle-aged. Those results suggested that it is difficult to maintain balance forward and backward in the elderly. The plantar pressure of the elderly while walking did not have two peaks. It suggested that the elderly walking form became a shuffle. The gait pace and the gait speed did not have significant differences between the young to middle-aged and elderly, but the plantar pressure parameters did. As a result, this study showed that the plantar pressure parameters can be used as characteristic indexes of elderly balance function and walking.

View full abstract

In an aging society, an increase is seen in the number of elderly person living alone. As the incidence of disease and injury increases with aging, a family who lives apart from an elderly parent may be concerned about that person's safety. Conversely, an aging parent may be interested in the lives of their children who live apart from them. To improve the relationship between family members who share such a situation, we have developed an inexpensive telemonitoring system using the television's operating state (TVOS). The system is comprised of sensor and communication sections. The sensor section consists of a current sensor for a television (TV) and an infrared sensor for a remote controller (RC), and outputs the TVOS, i.e., outputs indicating whether the TV is on or off and the RC is used. The communication section consists of a set of power-line communications modems and a computer with an Internet connection. We installed such a system in both an elderly person's home and that of his family living at a distant location. The computer at the family's end requires the TVOS data from the computer at the elderly person's end to be transmitted via the Internet once per minute. A feasibility test, conducted over one year, was performed in two pairs of parent-child families, four families, living apart. The results showed that (1) differences in the TVOS pattern was obtained by each family, and (2) the TVOS pattern differed between business days and non-business days in the same family. The present system does not specifically describe the health condition of an elderly parent living alone, but if that individual's family is familiar with the parent's viewing habits, they would be able to identify any change that might indicate a need for assistance.

View full abstract