We report a novel vibration control technique of an artificial auditory cochlear epithelium that mimics the function of outer hair cells in the organ of Corti. The proposed piezoelectric and trapezoidal membrane not only has the acoustic/electric conversion and frequency selectivity of the previous device developed mainly by one of the authors and colleagues, but also has a function to control local vibration according to sound stimuli. Vibration control is achieved by applying local electrical stimuli to patterned electrodes on an epithelium made using micro-electro-mechanical system technology. By choosing appropriate phase differences between sound and electrical stimuli, it is shown that it is possible to both amplify and dampen membrane vibration, realizing better control of the response of the artificial cochlea. To be more specific, amplification and damping are achieved when the phase difference between the membrane vibration by sound stimuli and electrical stimuli is zero and π, respectively. We also demonstrate that the developed control system responds automatically to a change in sound frequency. The proposed technique can be applied to mimic the nonlinear response of the outer hair cells in a cochlea, and to realize a high-quality human auditory system.

No abstract

A novel feedback control technique for the local oscillation amplitude in an artificial cochlear sensory epithelium that mimics the functions of the outer hair cells in the cochlea is successfully developed and can be implemented with a control time on the order of hundreds of milliseconds. The prototype artificial cochlear sensory epithelium was improved from that developed in our previous study to enable the instantaneous determination of the local resonance position based on the electrical output from a bimorph piezoelectric membrane. The device contains local patterned electrodes deposited with micro electro mechanical system (MEMS) technology that is used to detect the electrical output and oscillate the device by applying local electrical stimuli. The main feature of the present feedback control system is the principle that the resonance position is recognized by simultaneously measuring the local electrical outputs of all of the electrodes and comparing their magnitudes, which drastically reduces the feedback control time. In this way, it takes 0.8 s to control the local oscillation of the device, representing the speed of control with the order of one hundred times relative to that in the previous study using the mechanical automatic stage to scan the oscillation amplitude at each electrode. Furthermore, the intrinsic difficulties in the experiment such as the electrical measurement against the electromagnetic noise, adhesion of materials, and fatigue failure mechanism of the oscillation system are also shown and discussed in detail based on the many scientific aspects. The basic knowledge of the MEMS fabrication and the experimental measurement would provide useful suggestions for future research. The proposed preliminary prototype high-speed feedback control can aid in the future development of fully implantable cochlear implants with a wider dynamic range.

Recently, electrical sensing techniques for single objects, such as nanoparticles, biomolecules, and viruses, have attracted a great deal of attention. To achieve both high throughput and high measurement accuracy, target objects need to be quickly transported to a small sensing section embedded in a fluidic channel. In the present study, we propose a novel method to improve the signal-to-noise (S/N) ratio of electrical signals of single particles, using optical tweezers and a microchannel. Optically trapping a 2 μm microparticle in a micro-orifice that has a comparable dimension of 3.0 μm (W), 2.5 μm (H), and 3.0 μm (L), the electrical signal from a small target particle that passes by the microparticle is sharpened and separated from the background noise. By irradiation with nearinfrared light, the micro-orifice can be switched between opening and closing by optical tweezers, which works effectively to bring target particles to the sensing section using liquid flows and electrophoretic transport. As a result, the S/N ratio of electrical sensing of the smaller particle is improved by a factor of 5. The present microfluidic chip enables us to electrically detect particles of several hundreds of nanometers. Based on the present method, identification of single nanoparticles will also be feasible by using machine learning in the near future.

We report a novel mathematical model of an artificial auditory system consisting of a micro‐machined cochlea and the auditory nerve response it evokes. The modeled micro‐machined cochlea is one previously realized experimentally by mimicking functions of the cochlea [Shintaku et al, Sens. Actuat. 158 (2010) 183–192; Inaoka et al, Proc. Natl. Acad. Sci. USA 108 (2011) 18390–18395]. First, from the viewpoint of mechanical engineering, the frequency characteristics of a model device were experimentally investigated to develop an artificial basilar membrane based on a spring–mass–damper system. In addition, a nonlinear feedback controller mimicking the function of the outer hair cells was incorporated in this experimental system. That is, the developed device reproduces the proportional relationship between the oscillation amplitude of the basilar membrane and the cube root of the sound pressure observed in the mammalian auditory system, which is what enables it to have a wide dynamic range, and the characteristics of the control performance were evaluated numerically and experimentally. Furthermore, the stimulation of the auditory nerve by the micro‐machined cochlea was investigated using the present mathematical model, and the simulation results were compared with our previous experimental results from animal testing [Shintaku et al, J. Biomech. Sci. Eng. 8 (2013) 198–208]. The simulation results were found to be in reasonably good agreement with those from the previous animal test; namely, there exists a threshold at which the excitation of the nerve starts and a saturation value for the firing rate under a large input. The proposed numerical model was able to qualitatively reproduce the results of the animal test with the micro‐machined cochlea and is thus expected to guide the evaluation of micro‐machined cochleae for future animal experiments.