Detailed Use of Funds

Detailed use of the funds generated with this campaign

The AIM of the Foundation is to implant people who are locked-in, that is, paralyzed and mute, so as to restore speech.

Summary:

The funds will be used as follows: [1] provide an increase in the number of neural signals from implanted electrodes, [2] modify the electronics needed to read and amplify the signals, [3] modify the existing speech decoding system and [4] implant the system in locked-in persons in 2022.  These improvements are outlined in the following paragraphs:

Step 1:  The neurotrophic electrode consists of a small hollow glass cone that contains a set of electrical contacts and is coated with compounds that attract brain tissue to extend into the cone (and thus come into a close and stable position near the electrode contacts).  The current electrode, consisting of Teflon insulated gold wires inside the glass cone, has proven long-term efficacy in animal and human studies, including one study in a patient who survived for 13 years. This subject produced signals from identifiable specific neurons that could be identified for a decade until he became to ill to record1,2. A postmortem analysis after 13 years indicated no scar formation and abundant myelinated axons that allowed recording3. His recordings and those from Dr. Kennedy’s brain allowed decoding of the speech signals.

An improved version of the electrode was developed with the Neuronexus company (the NeuroneXus Neurotrophic Electrode (NXN), figure 1). In this electrode, 16 channels are available in a configuration shown in figure 2. This electrode is currently undergoing testing in animals, to determine how many signals can be recorded. We expect that 100 individual nerve cells ought to be recorded, compared to about 15 to 20 with the previous electrode design. This development step is already underway, utilizing funding from Dr. Kennedy, and is expected to be complete in about 4-6 months.

Step 2.  After completion of step 1, the study will be expanded with donated funds to involve additional animal studies, to provide a definitive basis for on-going development of the electrode. These extended studies will be to FDA standards. This component of the work is expected to take about one year and will cost approximately $200,000.

Step 3.  This step serves to further develop the electronic components needed to record the signals and to transmit them to external devices for further processing.  Ideally, four NXN Electrodes need to be implanted to cover the full extent of the human speech cortex.  To capture the signals from four electrodes, we will need a 4 x 16 (= 64) channel electronics package. The amplifiers are very small (1×2 cm) and will be buried under the scalp on the skull.  The amplified signals will be transported, using insulated wires under the skin, to a second unit that will be located under the skin of the chest wall. This unit will contain a 25 year battery and a transmitter to send the data to an external computer for production of speech (see figure 3 below).

Step 3 will take five years and will cost approximately $5,000,000. 

Step 4.  After completion of the first three steps, the system should be ready for assessment by the FDA regarding implantation in a human.  The Silent Speech Fund will fund the FDA application process, as well as the costs for each patient. Implantation does not need to wait for FDA permission because implantation can be done in other countries, such as Belize, after local permission has been obtained. Implants may be performed without the electronics, which is a common practice in present day brain machine implants. Thus patient implants may be done in the next year or two.

Summary of financial needs:

2022: $200,000

2023: $350,000

2024: $5,000,000

References:

  1. Kennedy P.R., 1 Dinal S. Andreasen,2 Jess Bartels,1 Princewill Ehirim,3 E. Joe Wright1, Steven Seibert1, Andre Joel Cervantes4  Validation of Neurotrophic Electrode long-term recordings in human cortex. Handbook of Biomedical Engineering. 2017.
  2. Kennedy P.R., Gambrell C, Ehirim P, and Cervantes A. Advances in the development of a speech prosthesis. Book chapter in Brain-Machine Interfaces: Uses and Developments accepted 2017.
  3. Gearin M and Kennedy PR. Histological confirmation of myelinated neural filaments within the tip of the Neurotrophic Electrode after a decade of neural recordings. Front. Hum. Neurosci. 21 April 2020. 
  4. Kennedy PR. The cone electrode: A long-term electrode that records from neurites grown onto to recording surface. J. Neurosci. Methods, 29:181-193, 1989
  5. Ganesh, A, Cervantes, A J and Kennedy PR, Slow firing single units are essential for optimal decoding of silent speech. Submitted to Computer Speech and Language 2020.