Greetings Everyone,
I would like to share an interesting paper I stumbled upon.
An implantable network of EMG threads powdered and managed by the method from the paper is what I would consider a fantastic way to gain an additional prosthetic limb, a haptic VR controller based on your hand EMG readings, or something similar.
I’ve seen the embedded temp sensor but this seems far far more practical.
Cheers,
I only gave it a cursory read but it immediately struck me as possibly being written by AI… there is a lot of word salad in there.
Also I think this paper sort of looks a little bit like patent filings which have never actually been constructed. They are just a series of ideas that skip over absolutely huge technical hurdles. Case in point, this idea of a semi-implantable has a single massive flaw which is the transdermal junction.
Notice how there is nothing special shown at the transition from external to internal through the skin? That is a major biological challenge they just skip over by just drawing a line through the skin.
If I get some time later I will give it a more intense read but papers like this take a lot of mental effort to read because you’re trying to decipher the word salad and at the same time identify and enumerate the things they completely skip over.
Yea, I’m almost out of my Master’s program and the fluf is just white noise at this point. I agree that the external aspects of that particular test unit are not ideal, but it’s what they had available for the rabbit based tests and they were decent with laying out the equipment they used. Stimulation would be impractical but EMG sensing with a fully internal implant shouldn’t be undoable.
Well the idea of implantable EMG sensors is nothing new. I’ve even seen some in real life, they just use inductive power and communications. When you start getting into emissive concepts like wireless power transmission, things get a little weird.
I’ve read a couple of the references from the first paper and there are some interesting coupling methods between sensor and receiver which bypass the complications of wireless in addition to power, like acoustic coupling, which is really interesting.
I’m always game to explore alternatives but so far inductive seems to be the rock solid way to go about it. Acoustic (ultrasonic) is interesting due to ultrasound machines being an almost ubiquitous diagnostic tool in medical offices big and small, but it’s finicky as heck and prolonged use invites micromechanical failure. Most everything I’ve seen as an alterative is fine for spot checking a sensor from time to time but for continuous operation, inductive is hard to beat.
What alternatives have you seen that could provide a serious challenge to inductive power transfer? Data can be RF broadcast if not modulated via inductive field… but getting practical power transfer through means other than inductive seems to be a particularly difficult.
In the paper I linked, they are powering the implant using surface electrodes similar to tens machine(spelling?)
Here:
From the paper
“The technique consists in the delivery of bursts of high-frequency currents through pairs of external textile electrodes and the development of thread-like devices, which can be deeply implanted by minimally invasive procedures. The electric currents, due to the conductive properties of the tissues, produce a voltage gradient that can be picked up by thin (<1 mm) and elongated (~3 cm) implants.”
I mean… to me, this sounds like they are just going to shock the person with unshielded electrodes and hope the implant picks up some of the current. They aren’t wrong… the human body is like a bag of salty water… plenty of conductivity in tissue… but running current through tissue is just shocking people. This would be a much more effective way to artificially animate a person like a digital puppet than power an implant… which is also cool and I want to see happen 
It also does not address how the “threads” that run from external devices to “deeply implanted” within tissue will deal with the constant risk of infiltration and infection through the transdermal incision / puncture. Prosthetics users who have bone mounts that protrude through the skin are constantly battling inflammation and infection in these areas. It is the primary reason things like dermal anchors always fail eventually…
But… this concern is really only related to Fig 2 in the original paper linked;
That seems… impractical at best… at least for long term implantation.
However… in reading more in-depth, I think I get the idea… Figure 1 outlines the concept best.
In this figure, the external bands are producing a rapidly alternating RF electric field and act like a kind of split di-pole with the implants aligned along the field, embedded within the muscle. A gradient will indeed be present and you should have enough energy to do real work with it… so this is indeed a new and different way of powering an implant.
Thanks for sharing!
Always a pleasure to brainstorm with interesting people.
While I eagerly await my subdermal IEMG sensor net, I welcome any novel ideas for achieving similar methods of generating actionable control signals from implants.
Even something like hal effect array wrist-cones detecting tiny magnet implants is worth consideration. Something free of environmental interference from light for traditional light reliant tracking technologies, IR, or even Lidar.
