A testbed for optimizing electrodes embedded in the skull or in artificial skull replacement pieces used after injury

Abstract

Background: Custom-fitted skull replacement pieces are often used after a head injury or surgery to replace damaged bone. Chronic brain recordings are beneficial after injury/surgery for monitoring brain health and seizure development. Embedding electrodes directly in these artificial skull replacement pieces would be a novel, low-risk way to perform chronic brain monitoring in these patients. Similarly, embedding electrodes directly in healthy skull would be a viable minimally-invasive option for many other neuroscience and neurotechnology applications requiring chronic brain recordings.

New method: We demonstrate a preclinical testbed that can be used for refining electrode designs embedded in artificial skull replacement pieces or for embedding directly into the skull itself. Options are explored to increase the surface area of the contacts without increasing recording contact diameter to maximize recording resolution.

Results: Embedding electrodes in real or artificial skull allows one to lower electrode impedance without increasing the recording contact diameter by making use of conductive channels that extend into the skull. The higher density of small contacts embedded in the artificial skull in this testbed enables one to optimize electrode spacing for use in real bone.

Comparison with existing methods: For brain monitoring applications, skull-embedded electrodes fill a gap between electroencephalograms recorded on the scalp surface and the more invasive epidural or subdural electrode sheets.

Conclusions: Embedding electrodes into the skull or in skull replacement pieces may provide a safe, convenient, minimally-invasive alternative for chronic brain monitoring. The manufacturing methods described here will facilitate further testing of skull-embedded electrodes in animal models.

Keywords: Brain monitoring; Brain-machine interface (BMI); Cranioplasty; Electrocorticogram (ECoG); Impedance; Skull-embedded electrodes; Traumatic brain injury (TBI).

Figure 1

A) The 8 × 8 grid of dots indicate the approximate placement of the 64 recording channels of all SREAs used in this study. The implant location was chosen to target upper-limb areas of the motor and premotor cortices for other ongoing studies (ArS = arcuate sulcus; CS = central sulcus). B) Example 3D macaque skull computer model generated from a CT scan. C) Example 3D macaque brain model generated from its co-registered MRI scan, D) Example computer model of a larger SREA with pre-formed electrode holes for direct manufacturing out of biocompatible material (not part of this study).

Figure 2

A) 3D-printed plastic skull. B) Silicon rubber is poured into the plastic skull covering the target area for the SREA. C) In design I, a sewing needle was used to thread electrode wire through the rubber in a grid pattern (figure shows insertion in progress with a grid overlay). Note the shadows on the rubber mold itself show there are ridges on the inner lumen of the skull that reflect the sulci/gyri of the underlying cortex. These ridges can be used to target electrodes over the desired anatomical landmarks. D) In design II, small ~2mm loops of deinsulated wire were inserted into slits in the rubber mold. Pouring acrylic over this configuration results in an SREA with small loops of deinsulated wires sticking into the cranial cavity (inset). These wires were then pushed flat against the acrylic before final implantation. E) Diagrams of the steel tubes with multistranded deinsulated wire inside used as electrodes in designs III (left) and IV (right). F) In designs IV and V, ‘place holder’ tubes made from hypodermic needles were inserted in a grid pattern into the rubber mold. Once completed, the full grid of tubes resulted in a skull-shaped piece of acrylic with an 8x8 grid of holes in which electrodes were later inserted. G) The left panel shows a close-up of the cured acrylic with holes made using the hypodermic needles in design four. Four of the holes have already been filled with the tube electrodes each containing a loop of deinsluated multistraded wire. The right panel shows the untrimmed molded acrylic after all the tubes were inserted. H) The final design V SREA with two expanded views of the laser-etched contact of one example electrode.

Figure 3

Cut-away illustration showing how the SREAs were implanted and secured. Once an SREA was placed within the craniotomy, titanium bone connector strips comprised of a string of connected metal rings were screwed to the skull to make ‘legs’ that extended under the scalp. The remainder of each bone strip was bent up and incorporated into the acrylic headcap that sealed the craniotomy as well as secured the connectors in place within a protective metal chamber.

Figure 4

Array densities compared by reanalyzing the same data using different subsets of the original 64 recording electrodes. Black dots visually indicate the three different uniform electrode spacing options compared. Note, in the half-density option, the assessment was repeated with both possible half-density subsets (i.e. black dots and grey dots). For the 1/8^th density option, all possible subsets were assessed that have the same relative spatial density pattern shown by the eight black dots in the plot on the right.

Figure 5

Box and whisker plots showing the median (black center line) and upper/lower quartiles (grey box top/bottom) of the 1.3KHz impedance values collected from all intact measured contacts combined over all data collection sessions for each SREA design in (A) the early (<3 mos.) and (B) late (>8 mos.) post-implantation epochs. Lower column labels indicates the SREA design number; columns ‘Va’ and ‘Vb’ indicate the flat and laser-etched electrodes in design V respectively. Whiskers indicate 1.5 times the interquartile range. Individual tick marks denote any data that fell outside 1.5 times the interquartile range. The numbers above each column indicate the number of testing days included in that data. Note, to estimate the number of channels outside the 1.5 interquartile range on a given day, one would divide the number of tick marks by the number of days above each column.

Figure 6

Percentage of tested channels that were assumed to have a bad connection due to 1.3KHz impedance values being > 10 times the median value for each design in the first recorded epoch. Dots are for individual testing sessions. Note, many values were the same between days resulting in some dots plotted on top of each other. Bars indicate the means of these percentage values across all recorded sessions. For design V, Va=flat and Vb=laser etched.