PURPOSE: To test the accuracy of a combined radiologist-computer syste
m in the diagnosis with magnetic resonance (MR) imaging of cancer of t
he prostate gland. MATERIALS AND METHODS: The combined system was deve
loped and tested by four specialists in prostate MR imaging and five r
adiologists expert in body MR imaging. Each group read MR images obtai
ned in 100 proved cases of prostate cancer. The images were obtained f
rom two sources, and all were obtained with an endorectal surface coil
. Prostate MR specialists ranked imaging features of cases to develop
a checklist for image interpretation. Features with greatest diagnosti
c value were incorporated in the combined system. Accuracy measures we
re derived from the area index of the receiver operating characteristi
c curve for the combined system and compared with those of radiologist
s working alone. RESULTS: Body MR radiologists had a mean baseline acc
uracy of 0.67; mean accuracy of their combined system was 0.80. The pr
ostate MR specialists, when they rated the features in each case, had
a mean accuracy of 0.81; the accuracy of their combined system was 0.8
7. CONCLUSION: A combined radiologist-computer system substantially im
proved accuracy of body MR radiologists in the diagnosis of prostate c
ancer. High levels of accuracy were also achieved by the system with p
rostate MR specialists.