This is precisely why you do it in water - the water-skin contact is effectively perfect, as is the water-transducer interface, and the body of water is easily characterizable; in effect you are scanning one large object that consists of a body of water that just happens to have a human body in it, and then extracting the body from that scan.

The water is a clever impedance matching trick. The contrast in density between air and human flesh is high, so the waves all reflect off the surface rather than penetrating and reflecting off the internal structures we care about.

That's why normally you're concerned with really good transducer contact (squeezing out any air) or use a gel to match impedance.

I'm a bit rusty on CT, but I'd guess the resolution is proportional to the total number of transducers in the array (e.g. larger sensing surface equals tighter resolution) since you're basically taking a Fourier transform of the incident wave.

This paper, “Whole Cross-Sectional Human Ultrasound Tomography”, goes into more detail.

https://arxiv.org/pdf/2307.00110