DEXA is very useful at determining bone density in adults, Z-scores, T-scores etc. This is done by imaging the lumbar spine and hip in country. Sometimes this is supplemented by imaging the shoulder too. However, by scanning the non-dominant hand ONLY, would it be possible to establish scores which helps us determine bone age?

The standard in planar imaging but it can be rather inaccurate. Can we be more accurate with DEXA while achieving a lower dose?

Also, any thoughts about how one would go about conducting an experiment without hitting an ethical landmine?

In children it seems to be possible: doi: 10.1259/bjr/23858213