Cancer Missed on Mammography
Abstract
Observer error in reading screening mammograms has been identified as a significant factor in delayed diagnosis of breast cancer, with up to 30% of potentially detectable cancers being overlooked. Computer-aided detection (CADe) has been developed to aid radiologists in the detection task, and pre-clinical studies have shown that CADe applied to digitized mammography films can flag about 50% of radiologists' observational oversights. The purpose of this investigation is to test how many additional cancers are detected by radiologists using CADe, in an observer study using an enriched mixture of cancers. Based on an initial pilot study, we estimate that we need 360 cases containing 60 cancers with 10 radiologists. The CADe system used in this study has a sensitivity of 53% with 0.48 false positives per image for the cases used in this study. To date five radiologists have completed the study. All were experienced radiologists spending 100% of their clinical time on breast imaging. For these readers, we did not measure a statistically significant increase in performance when using CADe. Several less experienced radiologists are now enrolled in the study. We will be able to examine the effect of CADe as a function of reader experience.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 2004
- Accession Number
- ADA433868
Entities
People
- Robert Nishikawa
Organizations
- University of Chicago