Ryoko Futami

An approach to a muscle model with a stimulus frequency-force relationship for FES applications

A simplified model of electrically stimulated muscle for use in applications of functional electrical stimulation (FES) is discussed in this paper. The muscle model was required to have both stimulus frequency and stimulus intensity (amplitude/width) inputs. The stimulus frequency versus force relationship of rabbit muscle was modeled first with a small number of model parameters that could be identified by simple experiments in a short time. The model identified was found to be applicable to human muscles. The frequency-force relationships of electrically stimulated fast and slow type muscles were also predicted by the model. The frequency-force model and a simplified model of muscle activation dynamics were used to construct a muscle model that described the summation of muscle contraction. The use of this model decreased the time burden on patients during parameter identification at the clinical site. The clinical applicability of these new model descriptions was suggested through computer simulations.

Sequential grouping of tone sequence as reflected by the mismatch negativity

Abstract.The human sequential grouping that organizes parts of tones into a group was examined by the mismatch negativity (MMN), a component of event-related potentials that reveals the sensory memory process. The sequential grouping is accomplished by the combinations of some factors, e.g., temporal and frequency proximity principles. In this study, auditory oddball stimuli in which each of the stimuli consisted of series of tone bursts, were applied to the subjects, and the MMN elicited by the deviation of the frequency of the last tone in the stimulus was investigated. The relationship between the expected phenomena of sequential grouping of tones and observed magnitudes of MMN was evaluated. It was shown that the magnitudes of MMN changed according to the configuration (number of tones, frequency) of tone sequence to be stored. This result suggested that the sequential grouping of presented tones was achieved on the preattentive auditory sensory memory process. It was also shown that the relative change of MMN magnitudes corresponded to the conditions of sequential grouping, which had been proposed by the auditory psychophysical studies. The investigation of MMN properties could reveal the nature of auditory sequential grouping.

A muscle activation model of variable stimulation frequency response and stimulation history, based on positive feedback in calcium dynamics.

Abstract. Muscle fiber response to a train of variable-frequency pulses includes the potentiation and catch-like effect. For better understanding of these phenomena, we built an activation model with emphasis on the calcium liberation from and re-sequestration into the sarcoplasmic reticulum, including calcium-induced calcium release. The model had two stable equilibrium points in the calcium concentration. Changes from the low to the high equilibrium point could be produced by high-frequency trains of pulses and would account for the potentiation. The model also showed a catch-like effect, as a long-lasting increment of muscle force after the application of a single extra pulse. The increase in force appeared in resting muscle, disappeared when the muscle was potentiated, and reappeared briefly if the stimulation was continued for long periods.

A basic study about multi channel measurement of skin impedance vector loci on the acupuncture points

Acupuncture points have been found to be points of low electrical resistance compared to the surrounding tissue. The significance and detail of the singular electrical properties of acupuncture points are uncertain at present. In this report, we measured the absolute value and the phase of skin impedance on an acupuncture point and non-acupuncture point at many frequencies and current values. Absolute value, phase and vector loci of skin impedance showed dependencies on the amplitude of applied current. So, we examined how the current dependencies changed according to the test subject and measuring points. As a result, by performing reliable measurement of skin impedance vector loci, we found that the suitable frequency range of applied current to the skin is from 10 to 400 Hz and we need to determine the current value for each measurement. Additionally, we applied those results to 4 channel measurements of the skin impedance vector loci around the acupuncture point under the condition that the room temperature and humidity were kept constant.

Basic characteristics of hardware neuron model based on CMOS negative resistance: realization of post-inhibitory rebound firing and its application

We introduced and evaluated the hardware neuron model with CMOS configuration which showed negative resistance characteristics. It was shown that this model had similar properties to those of the biological neuron, and that the post-inhibitory rebound firing (PIR firing) could be reproduced by this model. It was also found that the neural oscillator, consisting of a pair of neurons with inhibitory mutual connections, could be constructed by this hardware neuron model with the help of its PIR firing property. These results show the possibility of designing a new type of neuro-chip inspired by biological neural systems.

Fundamental Study For Rolling-over Motion Of The Body By Functional Electrical Stimulation (FES)

A method to roll-over the paralyzed body by means of Functional Electrical Stimulation (FES) is considered. It is demonstrated that individual joint motions necessary for the rolling-over are realized by electrical stimulation. EMG measurements are also performed to analyze the cooperative activities of the muscles during rolling-over motion in a case where an upper extremity was used. These results of two experiments using normal subjects verifies the fundamental feasibility of body control by FES.

Functional electrical stimulation (FES) for the voluntary rolling-over of the paralyzed body-first proposal and EMG analysis

A method is proposed for achieving voluntary rolling over of the paralyzed body by functional electrical stimulation (FES). Multichannel EMG recordings were made of the rolling-over motion of normal subjects in order to determine which muscles should be the target of FES. It is believed that the proposed method may be more practical and efficient than the use of an expensive patient care robot in order to prevent bedsores (decubitus).<<ETX>>

Temporal Resolution of the Skin Impedance Measurement in Frequency-Domain Method

The temporal-resolution of the frequency-domain method for the identification of the impedance locus depends on the basis frequency used in the current waveform construction, i.e., the higher basis frequency provides the better temporal resolution. The impedance locus can be characterized by the impedance parameters Z0, beta, and taum. The frequency distribution of limited number of data in the impedance locus would affect the accuracy of the estimated impedance parameters. Therefore, we investigated the relationship between the estimation accuracy of the impedance parameters and the frequency coverage of the impedance locus in relatively low to high impedance conditions (dc impedance Z0=51 kOmega-45 MOmega). As the basis frequency, 100 Hz was enough for the usual impedance with Z0 less than 203 kOmega. On the other hand, 10 Hz and 1 Hz were required for the medium-level (Z0=517 kOmega), and high-level (Z0=45 MOmega) impedance, respectively. The required basis frequency, accordingly the temporal resolution, depended much on the central relaxation time taum which affects the frequency distribution on the impedance locus. The results of this study are expected to serve as the reference of the frequency selection in the frequency-domain analysis of the skin impedance

Wearable data collection system for online gait stability analysis.

We had shown in our previous research that the stability assessment and control are essential for generation of faster and more energy efficient functional electrical stimulation (FES) and/or crutch‐assisted gait. The objective of our recent research work has been to design a wearable and portable system for gait stability analysis with online capabilities that is also applicable to crutch‐assisted gait modes. The developed wearable stability assessment system for as yet only biped gait consists of foot switches and goniometers attached to the leg joints. The instantaneous static and dynamic stability is, within the wearable system, assessed from the trajectory of the estimated body center of gravity (COGHAT) and the supporting area shape/size as derived from step length and foot‐floor contact state. We used motion analysis system data as reference for testing the wearable system accuracy. The wearable system was tested on five healthy subjects and one above‐knee amputee. It proved to be reasonably accurate if compared to the classical, motion analysis system based method. However, additional work is required to port the system to the FES assisted and/or crutch assisted gait.

Experimental tests of a musculoskeletal model of the elbow joint for FES applications

Computer model prediction of motions developed by functional electrical stimulation (FES) can be effective for clinical applications. The musculoskeletal model that responds to electrical stimulation is necessary for the model prediction. In this paper, a musculoskeletal model of the elbow joint was tested experimentally on a normal subject with surface electrode stimulation. In the musculoskeletal model, the musculotendon guide that reflects geometrical arrangement of muscles was adopted. Parameters in the muscle model were adapted to the subject based on the force measured under the isometric condition. The output of the tuned model agreed well with experimental results of the single muscle stimulation at high amplitude of stimulation and results of the simultaneous antagonistic stimulation in some stimulus conditions.