Computer-Aided Diagnosis and Feature-Guided Data Reduction Systems in Mammography
Abstract
We have performed extensive evaluation of the computer detection programs and the GUI this year. The mass detection program has been evaluated with over 300 mammograms at the University of Michigan and the Georgetown University. The microcalcification detection program was evaluated with 260 mammograms. The small-scale reading experiment simulating the pilot CAD reading of screening mammograms by four experienced mammography radiologists, we found that the CAD could improve the detection of cancer cases, but there might be a very small increase in the call-back rate. We expect that the pilot clinical study will provide information if the increase is statistically significant. Two CADView workstations have been implemented at the University of Michigan and the Georgetown University. The pilot clinical study in our off-line screening mammography clinics has begun and will collect data for the analysis of the effects of CAD on radiologists' reading.
Document Details
- Document Type
- Technical Report
- Publication Date
- Oct 01, 1999
- Accession Number
- ADA381681
Entities
People
- Heang-pint Chan
Organizations
- University of Michigan