This is a frustration, but burnout among doctors (senior and junior) is largely due to either corporatisation of private care or defunding of public care, leading to unmanageable workloads, scapegoating after patient harm events, and loss of autonomy.
It's a compiler for the APL language, that generates GPU code. What's interesting about it is that it's also hosted on the GPU, and it also has a very original architecture representing trees as arrays of pointers.
This is a very valuable contribution to computer science that has been overlooked because of how obscure the implementation language is.
The idea of that flattening a tree into an array with relative indexing to represent the structure opens up a whole bag of performance tools.
Parallel processing is just one (major!) aspect. It also enables GPU processing and unlocks SIMD instructions. Not to mention that instead of millions of tiny inefficient mallocs, this can use large contiguous arrays.
An itch I’ve been meaning to scratch is implementing something like Equality Saturation on top of this data structure in the style of EGG (e-graphs good).
Combine that with the Category Theory work done recently in the Haskell world for arbitrary structure to structure functional maps and you could make a compiler that takes a bunch of term rewrite rules and spits out parallel GPU code that can apply optimisation rewrites at enormous speed.
“The best thing about the computer revolution is that it hasn’t happened yet.”
They’ve forgotten the classic error of engaging only with client at the management and IT and ignoring clinical requirements until UAT, and a clinician revolt ends in project failure, adverse media coverage, millions of wasted dollars, and ongoing patient harm from unsatisfied requirements and persistent legacy systems.
Also needs an asterisk about not relying on your customer’s BAU resource to deliver projects, particularly those with large integration components, and one about not buying a business for its product as a replacement for yours which is EoL, then immediately trying to push it on your customers as an “upgrade” when you’ve lost half the acquisition’s engineering talent and you don’t understand the product.
Not OP, but I am a radiologist, and that 3T image is not representative of a state-of-the-art clinical scanner. The 11.7T does have extremely high spatial resolution but contrast resolution is compromised.
As others have said the clinical utility of >3T is very dubious, and most real world MR advances in the last 5 years have been AI (DL/ML not generative) reconstruction for lower field strengths.
That high a field strength has significant safety concerns, particularly for implants and pacemakers etc.
The safety aspect you point to is just so painful.
I was looking at a 5T scanner from United Imaging (interesting stories there…). Is there a single implant that’s been tested for these novel magnets? Checking out implants is painful enough at 1.5T and 3T.
reply