Application of invariant integrals to elastostatic inverse problems