Alan - Weeks 8/9

Charlie updated me with some news about the human islets – it looks like the next shipment will be arriving on August 14^th, exactly one day after I leave SF…

But on the bright side, he and Ana will be able to suspend the islets in a hydrogel and load that gel into my hollow fiber cartridges for insulin-glucose response testing. I’m looking forward to seeing some results in the next few weeks, even though I won’t be able to complete the final part of the study.

I’ve also done a lot of deeper CAD work with the backside structures of the iBAP device itself. As a reminder, the device faces some clotting issues because fibrinogen, a clotting protein, is small enough to pass through the silicon membranes. If the flow is stagnant, then this fibrinogen attaches to the internal walls of the device and settles down to begin coagulating, shown in the blue (slower flow velocity) areas of the following image. 

If the flow rate and shear stress along the walls is high enough though, the fibrinogen is ripped off and doesn’t have time to coagulate. So improving the flow path geometry to maintain high-velocity fluid flow is essential to keeping the device alive and un-clotted. 

Since the newest version of the device involves 3D stereolithography (SLA) printing instead of traditional plastic molding and machining, there’s a whole lot more we can do design-wise. SLA printing uses a UV laser to cure plastic resins from the bottom up, instead of ejecting melted plastic in thin layers. Here’s a cool TED talk on the technology, as well as a diagram that explains it better.

The result is that the layer-by-layer resolution is greatly increased. The traditional 3D printers we have at Peddie’s fab lab can’t create smooth curves – they have small steps which make the final product look pixelated. SLA printed parts are far smoother because the laser creates extremely small resolutions – sometimes into several microns per layer. 

This new development in our lab’s manufacturing inspired me to incorporate freeform curved surfaces into the backside structure design, as opposed to the traditional drilled-through holes that are currently used. Here are a few images of what the current design looks like compared to my preliminary concept looks like. It feeds ultrafiltrate from the membranes into a curved central reservoir on each side, which then reaches the islet gel and returns to the vein.

Above: Original design, with strictly defined flow path created from drilled holes. 

Top: Concept backside structure design. Middle: Curved lid creates a reservoir in the device for ultrafiltrate to collect. Bottom: Cross section of device assembly.

Above:  Blood flows between silicon membranes (green), into the ultrafiltrate reservoir, through the islet gel (pink), and into a central cavity for return to the blood vessel. 

Above: Comparison of concept device thickness (~10 mm) to current device thickness ~(13mm). Thinner devices are preferred since the final implantation will likely be in the forearm. 

Clearly, it's still in the very early stages of development as it doesn't have any mounting features yet. But I presented a few concept pictures during lab meeting and Shuvo thought it was a really promising design, so I’ll be spending the next week or so refining it, as well as running CFD to determine if it truly is better than the current design. Last week coming up soon!