< Previous | Home | Up | Table of Contents | Next >

Soft Uploading:

Design #1: Neuron-Silicon Hybrid

From my analysis this design is highly unlikely to be sufficient for soft uploading (ie memory tranfer).

However, this design is currently being explored by various research centers, for BCI purposes. Also, development of techniques of neuron reprograming would be beneficial to neurotransplantation.

The thinnest unmyelinated axons in humans are ~0.2µm in diameter; If myelinated ~1µm (inner diameter).


Spatial / Temporal Resolution

Both spatial and temporal resolutions may be limited by intrinsic characteristics of the axon, such as the refractory period and conduction speed. Can these parameters be modified?

Turn axon into ionic electrode? If we eliminate (eg by gene knock-out) voltage-gated ion channels along the axon, action potentials will be abolished. Further, if we eliminate all ion channels, the axon will become electrically isolated from the outside. Then it can be used as an ionic conductor with resistance in the GOhm range (if axon diameter ~ .3µm and axon length ~2mm). Signal transmission speed will be in the 100m/s range (calculated using transmission line theory with various assumptions), similar to that of normal axons. The plot of transmission speed vs frequency vs resistance is as shown. From the graph we can see that axoplasmic resistance should be reduced in order to increase the speed.

[Voltage attenuation may be a problem; what's its value?]

Junction

Need to create a junction that can initiate action potentials on the target cell, as well as report the target cell's activation (target's action potential) in the antidromic direction. Also this junction needs to report cell-type information of the target cell. This suggests that gap junctions may be a better candidate than chemical synapses.

Neuron Reprograming

The path-finding apparatus must be reprogramed. Also the axon should be made non-branching.


Comparison

A=solved, B=optimistic, C=maybe, X=difficult, ?=no clue

Requirements a. Neuron Hybrid b. Micromachine
R1.1. 2-way transmission C A
R1.2. Read Vm C B
R1.3. Write AP C B
R1.4. Temporal & digital resolution X A
R1.5. Electrical insulation A A
R1.6. Chemical insulation A C
R1.7. Impedance matching C A
R2.1. Cell-type recognition B C
R2.2. Cell-type reporting C C
R3.1. Non-damage A C
R3.2. Small volume displacement C B
R3.3. Penetration A C
R3.4. Even distribution ? C
R3.5. Path-finding B C
R3.6. Block-detour B C
R3.7. Cell-type targeting B C
R3.8. Probing density C B
R3.9. Site targeting C C
R3.10. Junction formation C ?
R4.1. Mechanical stability A A
R4.2. Biochemical stability C ?
R4.3. Electrical stability A A

< Previous | Home | Up | Table of Contents | Next >

Dec/2003 Yan King Yin