There was a study that found that, in radiology, human-first assessment resulted in worse outcomes that human-alone. Possibly the human's letting borderline cases through, on the assumption that the machine will catch them.
There's a roundup of such findings here, but they're a mixed bag: https://www.uxtigers.com/post/humans-negative-value I suspect you need careful process design to get better outcomes, and it's not one-size-fits-all.
If it was true, couldn't you get the same effect by taking a biopsy, fragmenting the cells, and injecting them back in? Like a vaccination, in fact. Somebody must have studied that approach already.
First issue is that tumors don't necessarily have to be highly immunogenic, e.g. there're tumors that don't present many neoantigens on the surface. This means immune cells can't easily recognize them. Second issue is that tumor microenvironment evolves to be immunosuppressive. There're many different signals that regulate immune cells activation and simply having antigen-specific cells isn't enough. But as someone said in a sister thread, what you're describing is a basis for multiple clinical trials that combine antigen release with immune activation.
There were reports that if you inject the goo from melting the tumor into another mouse, that mouse became much more resistant to that class of tumor[1], so...
I assume the immune system probably already reacts to this in a specific way. For example, a major bruise has a lot of broken up cells, but doesn't warrant a big immune response.
Major damage tends to cause a much larger immune response than a vaccination. That said, they do have therapeutic cancer vaccines that present proteins from cancer (sometimes patient-specific) with adjuvants to help stimulate the immune response.
Hah. I think of it as a slime mold. There's the main body (bodies?), but it's always shooting out little bits of itself that try weird stuff - founding underwater communes, or climbing mountains in Crocs or something. Most of these offshoots don't have that much of an impact, but occasionally one lucks out and discovers America or peanut butter and the main body saunters off that way.
Yeah we find this type of optimization all over nature. Even radiation is important to create noise. We need it in machine learning. A noisy optimizer is critical for generalized learning. Too much noise and you learn nothing but no noise and you only memorize. So there's a balance
I'm reminded a bit of a robotics professor I had at uni, a very long time ago in the 1990s. When he was at uni in the late 70s he and a couple of friends were working on a robot arm, which didn't work because it kept dropping stuff. They made a more precise linkage for the gripper, and if anything it was worse. They made ever more precise bushings for the pivots, and it just didn't help or made it worse.
Eventually after a conversation in the pub with one of his friends who was studying sports physiotherapy, they ripped out the 47th set of really precise little Teflon bushings and put in new ones made of medical silicone rubber tubing.
Now all the joints were a bit sticky and squashy and wobbly, and it picked everything up perfectly every time.
I get the "outliers are useful" thing you're trying to emphasis. But as someone from a mountainous country, please dont "climb[ing] mountains in Crocs", we regularily get media reports of hopelessly underequipped people having to be rescued with a whole team of people, in the middle of the night, with horrible weather, usually also endangering the people that do the rescue. I guess what I am trying to say is, there is a limit to how silly you can/should be.
reply