🧵 View Thread
🧵 Thread (4 tweets)
I'm still waiting for someone to figure out how to use a swarm of bees to parallelize breast cancer detection. https://t.co/LPOqhYvodX
There's been some interesting advancements in deep learning https://t.co/AfEgnXunB8, but there are limitations on what kinds of cancers that works well on https://t.co/t5hgcdE51o
We can seemingly train bees to detect cancer entirely by scent, and we don't quite yet have good enough digital noses to replace bees https://t.co/elosfZgFgsThe hard part for scaling this out would be scaling out / automating the act of maintaining a trained colony IMHO.
Not sure if anyone tried cuttlefish yet, but their eyes are fascinating. Despite only having cells to detect dark/light, many seem to use shape of eye & (what I would describe as) a frequency saccade to differentiate colors.https://t.co/j3HtmIVHR0 + are good at imitating shapes.