Blood oxygen sensors seem relatively cheap and low power.

I wonder if they could use that as the feedback mechanism.

Ideally if the sensors are small, low power, and cheap enough CO2 and lactic acid levels would also be good to check on to increase bloodflow.