© 1999 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Nanomedicine, Volume I: Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999

7.4.6.3 Auditory Outmessaging

The human ear consists of an air-filled external auditory canal terminating on the ~1 cm² tympanic membrane (~100 microns thick), whose vibrations are transmitted through the bony ossicles of the ~2 cm³ middle ear to the membranous oval window of the inner ear (Fig. 7.3). From there, sound energy enters the cochlea as compressional waves in the perilymph, an incompressible watery fluid that fills most of the ~100 mm³ spiral cochlear tube. This ~35-mm long tube resembles a snail shell, coiled two and a half times around the modiolus, itself the center axis along which pass the cochlear nerves and blood vessels. About 24,000 hair cells (multifiber stereocilia) on the basilar membrane under the tectorial membrane (Fig. 7.4) inside the cochlea resonate in response to different acoustic frequencies. These cells make complex synaptic contacts with neurons of the spiral ganglion, which passes this aural information into the auditory (8th cranial) nerve and thence to the brain for decoding.

A large nanodevice (~20-50 microns) stationed in the perilymph of the cochlea can "speak" directly into the human ear, loud enough to be heard. Total acoustic power delivered to the tympanic membrane during normal audition is ~10^-4 pW (0 dB) at the threshold of hearing, ~0.1 pW for whispering (30 dB), and ~100 pW for normal conversation (60 dB). At the low frequencies (30-10,000 Hz) common to human speech, micron-scale acoustic radiators are very inefficient (Section 7.2.2.1) although there are virtually no attenuation losses (Eqn. 4.52). Even so, from Eqn. 7.7 a cochlear nanodevice able to apply an input power of 2000 pW to an acoustic omnidirectional piston radiator 20 microns in diameter can generate 10 KHz sound waves at an output power ~ 0.1 pW, approximately the intensity level of whispering. At a more comfortable 3 KHz (the peak sensitivity of human hearing), P_in must rise to ~22,000 pW or piston diameter must increase to 45 microns to maintain P_out ~ 0.1 pW, or a duty cycle substantially less than 100% must be employed. If linked to a communication network, the cochlear nanodevice can receive information to be audibilized from the network. This device can also be used for acoustic inmessaging by allowing it to intercept sound waves before they reach the basilar membrane, then passing this information immediately to the network (which thus receives the signal sooner than the brain). Perilymph contains glucose for power at roughly serum concentrations; alternatively, the cochlear device can absorb and store the natural ambient acoustic energy present in the auditory canal (Section 7.4.8).

Similarly, a nanorobot located in the endolymph-filled scala media, the innermost of the three parallel chambers of the cochlea, can attach itself to the reticular lamina of the organ of Corti and physically manipulate the stereocilia to produce the desired audible stimulus. Stereocilia are spaced over the organ of Corti at a linear density of ~1 cell/micron along the 30 mm length of the cochlear spiral tube.⁵⁸⁵ Continuously stimulating all ~100 cells within reach of a 20-micron diameter nanorobot with bidirectional 40-micron long extensible manipulators should produce a narrow-frequency tone up to ~0.4% of maximum pressure amplitude, close to normal conversational levels. Hair cells have a membrane time constant of ~0.5 millisec ,⁷⁷² in theory permitting a transfer rate of up to ~2000 bits/sec if all stereocilia are manipulated coherently as a single channel. Careful design should prevent unwanted stimulation of tinnitus.

Direct neural stimulation of some of the ~31,000 spiral ganglion cells allows transduction of information into the human auditory system. Each ganglial cell carries temporal and amplitude information on a narrow band of acoustic frequencies received by the ear, essentially a crude Fourier transform of the original audio signal. Properly positioned nanorobots can superimpose artificial signals on this natural traffic which may be perceived by the patient as a frequency-reconstructed voice of limited tonal range speaking words or numbers. Nanorobots stationed at the saccule, an otolithic organ in the vestibule of the inner ear, can stimulate direct ultrasonic hearing in humans up to ~108 KHz.¹³⁷² In 1998, cochlear implant devices were in wide use, with audible sounds transmitted by FM radio waves to an electrode array in the inner ear, providing direct electrical stimulation of the auditory nerves.^1891-1893

A review of current technology in artificial speech synthesis^1653,1654 is beyond the scope of this text.

Last updated on 19 February 2003