← Explore

Towards multifocal ultrasonic neural stimulation II: design considerations for an acoustic retinal prosthesis

Omer Naor, Yoni Hertzberg, Esther Zemel, Eitan Kimmel, Shy Shoham

Journal of Neural Engineering 2012 · 10.1088/1741-2560/9/2/026006

rodenthealthyeeg megotherhistology molecular

Abstract

Ultrasound waves, widely used as a non-invasive diagnostic modality, were recently shown to stimulate neuronal activity. Functionally meaningful stimulation, as is required in order to form a unified percept, requires the dynamic generation of simultaneous stimulation patterns. In this paper, we examine the general feasibility and properties of an acoustic retinal prosthesis, a new vision restoration strategy that will combine ultrasonic neuro-stimulation and ultrasonic field sculpting technology towards non-invasive artificial stimulation of surviving neurons in a degenerating retina. We explain the conceptual framework for such a device, study its feasibility in an in vivo ultrasonic retinal stimulation study and discuss the associated design considerations and tradeoffs. Finally, we simulate and experimentally validate a new holographic method--the angular spectrum-GSW--for efficient generation of uniform and accurate continuous ultrasound patterns. This method provides a powerful, flexible solution to the problem of projecting complex acoustic images onto structures like the retina.

Abstract via europepmc.

Speciesrat (Sprague-Dawley)
Subjects6, 3, 3swept animals
Sessions per subjectnot reported
Randomisednot reported
Blindingnot reported
Sham / controlinactive transducer
Auditory controldeafened subjects
Readout timingboth
Anaesthesiaanaesthetised
Readoutseeg meg, other, histology molecularelectroretinogram (ERG)
Direction of effectexcitatoryFull-field ultrasonic pulses to the eye evoked visual-evoked-potential-like responses at both 0.5 and 1 MHz that were reduced by intravitreal TTX, indicating a neuronal (RGC) origin of the excitation.
Adverse eventsnone observedERG a-wave and b-wave amplitudes and morphology in US-treated eyes were similar to untreated eyes at 3 h and 24 h after stimulation, and histological sections of the treated retina showed no visible damage.

Exposures

Exposure 1: 0.5 MHz full-field ultrasonic stimulation of the eye/retina

Target: retina — “retina
Device: Olympus / Panametrics · Olympus · V301

Pulse timing
Waveformpulsed
Fundamental frequency (kHz)500✓✓
Pulse duration (ms)0.05, 0.1swept✓✓
Pulse repetition frequency (Hz)1,900, 2,000swept✓✓
Duty cycle (%)not reported
Sonication duration (s)0.005, 0.02swept✓✓
Pressure and intensity, by domain
Free-field pressure (kPa)86, 160swept✓✓
Free-field Isppa (W/cm²)0.12, 0.42swept✓✓
Free-field Ispta (W/cm²)0.012, 0.105swept✓✓
In-situ estimatederatingmean or range across subjects
In-situ pressure (kPa)not reported
In-situ Isppa (W/cm²)not reported
In-situ Ispta (W/cm²)not reported
Protocol, in the paper’s words

US transducer excited every 1 or 2 s by burst trains; VEP experiment used burst train durations of 5-20 ms made of single bursts of 50-100 us at a repetition frequency of 1900-2000 Hz (at least 150 repeats averaged). A separate ERG safety experiment applied pulsed US continuously for 90 min, consisting of burst trains spaced 1 s apart with burst train duration 20 ms, single burst duration 100 us and repetition frequency 2000 Hz, selected to represent an upper limit of the dosages used in the 0.5 MHz study.

Exposure 2: 1 MHz full-field ultrasonic stimulation of the eye/retina

Target: retina — “retina
Device: IGT / Imasonic · Imasonic · 3034

Pulse timing
Waveformpulsed
Fundamental frequency (kHz)1,000✓✓
Pulse duration (ms)0.1✓✓
Pulse repetition frequency (Hz)1,667✓✓
Duty cycle (%)not reportedpulse duration × PRF gives 16.67%
Sonication duration (s)0.01, 0.02swept✓✓
Pressure and intensity, by domain
Free-field pressure (kPa)564, 725swept✓✓
Free-field Isppa (W/cm²)5.15, 8.52swept✓✓
Free-field Ispta (W/cm²)0.86, 1.42swept✓✓
In-situ estimatederatingmean or range across subjects
In-situ pressure (kPa)not reported
In-situ Isppa (W/cm²)not reported
In-situ Ispta (W/cm²)not reported
Protocol, in the paper’s words

US transducer excited every 1 or 2 s by burst trains; burst train durations of 10-20 ms made of single bursts of 100 us at a repetition frequency of 1667 Hz (at least 150 repeats averaged).

Flags from extraction

  • n_subjectsPaper reports three separate animal groups (n=6 at 0.5 MHz VEP, n=3 at 1 MHz VEP, and a separate group of three animals for the ERG safety study at 0.5 MHz) but never states a study-wide total; listed as group sizes.
  • exposures[0].in_situ.pressure_kpaValues are described as pressures/intensities 'incident on the cornea', derated from free-field water calibration via the corneal acoustic impedance; unclear whether this should count as the retina target (12% further pressure decrease to the retina is mentioned separately but not applied to these values).
  • exposures[1].in_situ.pressure_kpaSame domain ambiguity as the 0.5 MHz exposure: values are incident-on-cornea estimates, not measured at the retina itself.
  • exposures[0].in_situ.ispta_w_cm2ERG safety experiment states ISPTA of 0.105 W/cm2, slightly above the 0.012-0.083 W/cm2 range given in table 1 for the same 0.5 MHz condition; both values retained as list.
  • sham_typeThe main control demonstrating acoustic specificity ('US in air') moved the coupling cone away from the eye rather than describing a classic inactive-transducer configuration; classified as inactive_transducer (aimed away) but the mechanism is a judgement call.
  • auditory_controlEars were physically sealed with dental elastomer 'to avoid auditory artifacts', classified as deafened_subjects though this is a temporary occlusion rather than deafening.
  • readoutsElectroretinogram (ERG) does not fit cleanly into any vocabulary category and was coded as 'other' with readout_other text.