Eyes are windows into the soul, the old saying goes. They are also pathways into the mind, as much of our brain is involved in processing visual input. This dedication to vision is partly why much of AI research is likewise focused on machine vision. But do artificial neural networks (ANN) actually work like the gray matter that inspired them? A recently published research paper (DOI: 10.1126/science.aav9436) builds a convincing argument for “yes”.
Neural nets were named because their organization was inspired by biological neurons in the brain. But as we learned more and more about how biological neurons worked, we also discovered artificial neurons aren’t very faithful digital copies of the original. This cast doubt whether machine vision neural nets actually function like their natural inspiration, or if they worked in an entirely different way.
This experiment took a trained machine vision network and analyzed its internals. Armed with this knowledge, images were created and tailored for the purpose of triggering high activity in specific neurons. These responses were far stronger than what occurs when processing normal visual input. These tailored images were then shown to three macaque monkeys fitted with electrodes monitoring their neuron activity, which picked up similarly strong neural responses atypical of normal vision.
Manipulating neural activity beyond their normal operating range via tailored imagery is the Hollywood portrayal of mind control, but we’re not at risk of input injection attacks on our brains. This data point gives machine learning researchers confidence their work still has relevance to biological source material, and neuroscientists are excited about the possibility of exploring brain functions without invasive surgical implants. Artificial neural networks could end up help us better understand what happens inside our brain, bringing the process full circle.
[via Science News]
Insulin pumps are a medical device used by people with diabetes to automatically deliver a measured dose of insulin into their bloodstream. Traditionally they have involved a canula and separate connected pump, but more recent models have taken the form of a patch with a pump mounted directly upon it. When [Pete Schwamb]’s daughter received one of these pumps, an Omnipod, he responded to a bounty offer for reverse engineering its RF protocol. As one of the people who helped create Loop, an app framework for controlling insulin delivery systems, he was in a particularly good position to do the work.
The reverse engineering itself started with the familiar tale of using an SDR to eavesdrop on the device’s 433MHz communication between pump and control device. Interrogating the raw data was straightforward enough, but making sense of it was not. There was a problem with the CRC algorithm used by the device which had a bug involving a bitwise shift in the wrong direction, then they hit a brick wall in the encryption of the data. Hardware investigation revealed a custom chip in the device, and there they might have stalled.
But the international reverse engineering community is not without resources and expertise, and through the incredible work of a university researcher in the UK (whose paper incidentally includes a pump teardown) they were able with an arduous process supported by many people to have the firmware recovered through decapping the chip. Even once they had thus extracted the encryption code and produced their own software their problems were not over, because communication issues necessitated a much better antenna on the RileyLink Bluetooth bridge boards that translated Bluetooth from a mobile phone to 433 MHz for the device.
This precis doesn’t fully encapsulate the immense amount of work over several years by a large group of people with some very specialist skills that reverse engineering the Omnipod represents. To succeed in this task is an incredible feat, and makes for a fascinating write-up.
Thanks [Alex] for the tip.
Every few years, or so we’re told, [Scott] revisits the idea of building an electrocardiogram machine. This is just a small box with three electrodes. Attach them to your chest, and you get a neat readout of your heartbeat. This is a project that has been done to death, but [Scott]’s most recent implementation is fantastic. It’s cheap, relying on the almost absurd capability for analog to digital conversion found in every sound card, and the software is great. It’s the fit and finish that makes this project shine.
The hardware for this build is simply an AD8232, a chip designed to be the front end of any electrocardiogram. This is then simply connected to the microphone port of a sound card through a 1/8″ cable. For the exceptionally clever, there’s a version based on an op-amp. It’s an extraordinarily simple build, but as with all simple builds the real trick is in the software. That’s where this project really shines, with custom software with graphics, and enough information being displayed to actually tell you something.
We’ve seen a number of sound card ADCs being used for electrocardiograms in the past, including some from the Before Times; it makes sense, sound cards are the cheapest way to get a lot of analog data very quickly. You can check out [Scott]’s demo video out below.
Continue reading “Sound Card ADCs For Electrocardiograms”
As 3D printing becomes more and more used in a wide range of fields, medical science is not left behind. From the more standard uses such as printing medical equipment and prosthetics to more advanced uses like printing cartilages and bones, the success of 3D printing technologies in the medical field is rapidly growing.
One of the last breakthrough is the world’s first 3D vascularised engineered heart using the patient’s own cells and biological materials. Until now, scientists have only been successful in printing only simple tissues without blood vessels. Researchers from Tel Aviv University used the fatty tissue from patients to separate the cellular and acellular materials and reprogrammed the cells become pluripotent stem cells. The extracellular matrix (ECM) was processed into a personalized hydrogel that served as the basis from the print.
This heart is made from human cells and patient-specific biological materials. In our process these materials serve as the bioinks, substances made of sugars and proteins that can be used for 3D printing of complex tissue models… At this stage, our 3D heart is small, the size of a rabbit’s heart, but larger human hearts require the same technology.
After being mixed with the hydrogel, the cells were efficiently differentiated to cardiac or endothelial cells to create patient-specific, immune-compatible cardiac patches with blood vessels and, subsequently, an entire heart that completely matches the immunological, cellular, biochemical and anatomical properties of the patient. The difficulty of printing full-blown organs were being tackled for a long time and we already talked about it in the past.
The development of this technology may completely solve both the problem of organ compatibility and organ rejection.
That first glimpse of a child in the womb as a black and white image on a screen is a thrilling moment for any parent-to-be, made possible by several hundred thousand dollars worth of precision medical instrumentation. This ultrasound machine cobbled together from eBay parts and modules is not that machine by a long shot, but it’s still a very cool project that actually gives a peek inside the skin.
The ultrasound transducer used by [stoppi71] in this build has an unusual source: a commercial paint-thickness meter. Cue the jokes about watching paint dry, but coatings measurement is serious stuff. Even so, the meter in question only ran about $40 on eBay, and provided the perfect transducer for the build. The sender needs a 100V pulse at about 5 MHz, so [stoppi71] had some fun with a boost converter and a 74121 Schmitt-trigger one-shot driving a MOSFET to switch the high voltage. On the receive side, the faint echo is sent through a three-stage amp using AD811 op amps before going through an LM7171 op amp acting as a rectifier and peak detector. Echos are sent to an Arduino Due for display on a 320×480 LCD. The resolution isn’t great, but the video below shows that it’s enough to see reflections from the skin of [stoppi71]’s forearm and from the bones within.
[stoppi71] says that he was inspired to tackle this build by Murgen, an open-source ultrasound project. That project got further refined and entered into the “Best Product” category in the 2018 Hackaday Prize. We like that because focusing on turning projects into products is what this year’s Hackaday Prize is all about.
Continue reading “Simple Ultrasound Machine Shows The Skeleton Lurking Inside Us All”
Can you electronically enhance your brain? I’m not talking about surgically turning into a Borg. But are there electronic methods that can improve various functions of your brain? Fans of brainwave entrainment say yes.
There was an old recruiting ad for electrical engineers that started with the headline: The best electronic brains are still human. While it is true that even a toddler can do things our best computers struggle with, it is easy to feel a little inadequate compared to some of our modern electronic brains. Then again, your brain is an electronic device of sorts. While we don’t understand everything about how it works, there are definitely electric signals going between neurons. And where there are electric signals there are ways to measure them.
The tool for measuring electric signals in the brain is an EEG (electroencephalograph). While you can’t use an EEG to read your mind, exactly, it can tell you some pretty interesting information, such as when you are relaxed or concentrating. At its most basic we’ve seen toys and simple hobby projects that purport to be “mind controlled” but only at an incredibly rudimentary level.
Brainwave entrainment is a hypothesis that sending low frequency waves to your brain can give your mind a nudge and sync up brain activity with the equipment measuring it. The ability to synchronize with the brain could yield much better measurements for a meaningful interface between modern electronics and electric storm of thought happening in your head.
Continue reading “Brain Hacking with Entrainment”
One of the vast untapped potentials of medicine is the access to imaging equipment. A billion people have difficulty getting access to an x-ray, and that says nothing about access to MRIs or CAT scans. Over the past few years, [Jean Rintoul] has been working on a low-cost way to image the inside of a human body using nothing more than a few electrodes. It can be done cheaply and easily, and it’s one of the most innovative ways of bringing medical imaging to the masses. Now, this is a crowdfunding project, aiming to provide safe, accessible medical imaging to everyone.
It’s called Spectra, and uses electrical impedance tomography to image the inside of a chest cavity, the dielectric spectrum of a bone, or the interior of a strawberry. Spectra does this by wrapping an electrode around a part of the body and sending out small AC currents. These small currents are reconstructed using tomographic techniques, imaging a cross-section of a body.
[Jean] gave a talk about Spectra at last year’s Hackaday Superconference, and if you want to look at the forefront of affordable medical technology, you needn’t look any further. Simply by sending an AC wave of around 10kHz through a body, software can reconstruct the internals. Everything from lung volume to muscle and fat mass to cancers can be detected with this equipment. You still need a tech or MD to interpret the data, but this is a great way to bring medical imaging technology to the people who need it.
Right now, the Spectra is up on Crowd Supply, with a board that can be configured to use 32 electrodes. Measurements are taken at 160,000 samples/sec, and these samples have 16-bit resolution. This is just the acquisition hardware, though, but the software to do tomographic reconstruction is open source and also readily available.
In terms of bringing medical imaging to the masses, this is a very impressive piece of work, and is probably the project from last year’s Hackaday Prize that has the best chance of changing the world.