I spent the weekend dissassembling a hematology analyzer capable of measuring 17 parameters of the immune system from a large drop of blood. Nabbed it on ebay for cheap.
Was able to unclog a few of the lines and get it running, learning how it works by shooting cells past a laserbeam and looking at scatter. Overall, the complexity of the machine is far simpler than that of a simple laser printer, but its retail value is $10k as a veterinary instrument. Machines for human patients are way more expensive.
My dissappointment however was discovering that it refuses to run unless its barcode scanner can see manufacturer-validated reagent bottles. The chemicals needed to make it run are simple salt-waters that can be easily made for $10 of various powders, but the manufacturer sells reagent sets for $1000k each. Now its clear that getting any useful data out of it will require hacking the ARM embedded system onboard.
Why am I doing this? Certainly not for any kind of regulated medical treatment, but I personally want to know how my immune system changes on a day to day basis. Datasets like this dont exist on such a high frequency sampling rate. What I find however, is that the decades-old tech that powers this relatively simple machine is being held back by an overly restrictive regulatory environment and price gouging of chemical reagents.
Why the artificial construction of a $100 lab test, when the actual value to perform such a test is much much lower? Why can't we live in a world where such self-experimentation is allowed outside of the strict guidelines of diagnostics?
I happen to do some IT field support for my vet - she has one of these machines too, with the same problem of expensive reagents.
The problem space here is: of course anyone can mix the fluids together from powder... but under what conditions? Tested for impurities, contaminants or unevenly mixed/clumped powder? For a clean water source? Regular batch checks to ensure the stuff is up to spec? Mess up one of these things, the system doesn't work/produces false results, a pet or a human dies and the payout sums can be enormous.
The latter risk and the insurance premiums as well as mitigation risks are why medical stuff is so insanely expensive.
And another thing: at least when someone dies, it's over. Bit of a lump sum to relatives of the deceased, funeral cost, that's it. Now, when someone does not die, but gets stuck with brain damage, in an awake coma or bedridden for life, the costs of keeping that person alive can well run into double-digit millions of euros whereas 50 years ago that person would simply die sooner or later. Midwifes in Germany are retiring in droves because insurance premiums have skyrocketed - due to exploding costs for newborns requiring extensive care after e.g. being stuck in the vaginal canal during birth.
That's a pretty awesome project. Instead of hacking the ARM, could you just call up a vet to ask if you can have some of their empty reagent bottles, and fill those with your homemade reagents? (As a bonus, if you can get it to work, the vet would probably be interested in knowing the reagent recipe as well.)
There's a movement towards open-source designs for relatively simple biological instrumentation- perhaps this sort of machine will be in range of those efforts someday soon.
The one neat thing about hacking the ARM (its an at91 system if anyone knows details) is that I can get access to the raw data underneath. IE, the real-time cell counting events. I can envisage some neat deep learning on something like that, if only the dataset existed.
I don't know anything about this particular case, but systems like that based on controlling "consumables" are very common.
Sometimes it's simple price gouging, sure, or at least partly. Often, though, it was in response to capitol equipment budgeting and per-procedure pricing (& insurance). It can be easier for organizations to justify/budget the cost (and pass that on) amortized over N exams than all up front, so device manufacturers figured out ways to do this. These days some devices are entirely funded by contracted use (i.e. free razor, pay for the blades), much like office printers etc.
You are right that the industry as a whole has little interest and no incentive to allow experimentation and re-purposing. Part of this is liability, a lot is lack of any incentive. On the other hand, research lab equipment with similar functionality is often far more accessible (and more expensive)
Maybe you could get your hands on some spent reagent bottles, or even just photos of them. Reset the system's memory in between uses if it requires new bottles each time.
Hacking the controller could be challenging if it's been deliberately hardened.
Heads up: it looks like they have the expiration date embedded in the first barcode and one of the pictured reagents is expired. Have you been able to set the system clock?
I'm aware of the expiration date, but haven't been able to understand the embedding or how the pairs of codes may be "hash validated" in some way, which means that this is a one time use for me unless I can either generate them or access the nvram on the system.
Those reagents are like printer ink and the manufacturer makes plenty of money off of them. Also, the whole thing is supposed to be run with sterilized lines and sterilized water.
That is absolutely awesome, I've always wanted to do some kind of higher frequency analysis of my body's internal processes via blood sampling. It's cool to hear somebody doing that.
Bah, $10k - and yes, that was evidently the retail price for this guy :
Sorry for confusion .... Morning cough syrup hangover. Sometimes I wonder if the drug is worse than the fever! It would be nice if I could monitor my infection with this dang analyzer to see it's rise and fall :p
Was able to unclog a few of the lines and get it running, learning how it works by shooting cells past a laserbeam and looking at scatter. Overall, the complexity of the machine is far simpler than that of a simple laser printer, but its retail value is $10k as a veterinary instrument. Machines for human patients are way more expensive.
My dissappointment however was discovering that it refuses to run unless its barcode scanner can see manufacturer-validated reagent bottles. The chemicals needed to make it run are simple salt-waters that can be easily made for $10 of various powders, but the manufacturer sells reagent sets for $1000k each. Now its clear that getting any useful data out of it will require hacking the ARM embedded system onboard.
Why am I doing this? Certainly not for any kind of regulated medical treatment, but I personally want to know how my immune system changes on a day to day basis. Datasets like this dont exist on such a high frequency sampling rate. What I find however, is that the decades-old tech that powers this relatively simple machine is being held back by an overly restrictive regulatory environment and price gouging of chemical reagents.
Why the artificial construction of a $100 lab test, when the actual value to perform such a test is much much lower? Why can't we live in a world where such self-experimentation is allowed outside of the strict guidelines of diagnostics?