Really interesting idea. But breathomics seems tricky — breath signatures can shift with diet, time of day, stress, exercise, even the room you’re in. How do you think a stable baseline could be built across people (or even within one person) without very tight controls? Could this avoid the pitfalls that stalled other non-invasive diagnostics like urine or saliva tests?
Pretty wild to see how the DOE labs certify the stockpile without live testing anymore, just layers of modeling, diagnostics, and some very calculated inference.
LANL’s particle accelerator shuts down about every 28 days because a tiny tungsten filament burns out under heat, cesium, and ion bombardment. Researchers are testing graphene coatings—atomic-scale “armor”—to keep it running longer.
I would love to hear if you think this kind of nano-protection could be generalized to other extreme environments (fusion, space, etc).
You're going to cringe when you hear this, but a week ago I uploaded recent lab results to ChatGPT, and received a thorough reviews of the results. Then I asked if there was a connection between my labs and a medical issue a sibling of mine is having. It clearly outlined how they weren't related and provided prompts for me to ask my doctor as a follow up.
Of course, I should verify everything from CGPT (and I shouldn't share my labs with it), but medicinal AI seems like it could soon be a game changer for patients.
