PURPOSE: Area under a receiver operating characteristic (ROC) curve (A
(z)) is widely used as an index of diagnostic performance. However, A(
z) is not a meaningful summary of clinical diagnostic performance when
high sensitivity must be maintained clinically. The authors developed
a new ROC partial area index, which measures clinical diagnostic perf
ormance more meaningfully in such situations, to summarize an ROC curv
e in only a high-sensitivity region. MATERIALS AND METHODS: The mathem
atical formulation of the partial area index was derived from the conv
entional binormal model. Statistical tests of apparent differences in
this index were formulated analogous to that of A(z). One common stati
stical test involving the partial area index was validated by computer
simulations under realistic conditions. RESULTS: An example in mammog
raphy illustrates a situation in which the partial area index is more
meaningful than A(z) in measuring clinical diagnostic performance. CON
CLUSION: The partial area index can be used as a more meaningful alter
native to the conventional A(z) index for highly sensitive diagnostic
tests.