Development of Methods for Computer-Assisted Interpretation of Digital Mammograms for Early Breast Cancer Detection.
Abstract
The goal of tha research was to develop a computer-vision module as an aid to radiologists the specific aims: (1) Further development of advanced computerized schemes for the detection and classification of masses and microcalcifications in digital mammograms, including quantitative analysis of the radiographic characteristics and the decision-making processes used by radiologists in making a decision with respect to the likelihood of malignancy and choosing the appropriate course of action. (2) Development of a dedicated intelligent modular system with man-machine interfaces and fast computation times appropriate for the effective use of the computer-vision schemes. (3) Evaluation of the efficacy and efficiency of the module using a large clinical database. The significance of this research is that if the detectability of cancers can be increased by employing a computer to aid the radiologist's diagnosis, then the treatment of patients with cancer an be initiated earlier and their chance of survival improved. Systematic introduction of computer-vision tools to radiologists that is presented in this proposal requires minimal modification to the current reading habits of radiologists. When digital mammographic imaging units become commonplace, the computer-vision module can be interfaced to electronic, filmless medical imaging reading areas.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 27, 1995
- Accession Number
- ADA295502
Entities
People
- Maryellen Lissak Giger
Organizations
- University of Chicago