Researchers attempt to get a quicker mammography second opinion
Duke University medical physicists are taking a page out of Google’s playbook by developing an application that is able to retrieve mammography studies and other information from a database in a matter of three seconds. The information can be used as a sort of second opinion that compares the mammography image currently being reviewed with each breast image available in the database. This approach is Google-like in that it orders results dynamically by relevance to certain pieces of key information, much like the popular search engine does with keywords.
In a pilot study, the approach allowed computers to maintain their high level of accuracy while performing faster analysis, which will be important as databases amass more information that has to be sorted speedily, according to a release of the study’s findings.
Knowledge-based computer-assisted detection (CAD) systems compare mammogram images to those of known cases of breast cancer in order to aid radiologists in their diagnosis. To avoid the systems getting bogged down with information, the researchers are presenting a Knowledge-Based Computer Assisted Detection (KB-CAD) system that analyzes breast masses using the principles of information theory.
The system uses an approach that compares new unknown cases with a restrictive search for that information that seems most useful, instead of every similar case in the database. Thus, the speed of results is much faster.
In the recent pilot study, the Duke researchers applied their technique to a database of 2,300 mammography images. With this type of indexing, the researchers compared a sample image to the top 600 most informative, cutting down their CAD system's processing time by one-fourth, to less than 3 seconds per query. The researchers expect to launch a larger study in a year to evaluate the clinical impact of this new approach.
This study was presented on August 1 at the 48th Annual Meeting of the American Association of Physicists in Medicine in Orlando.
In a pilot study, the approach allowed computers to maintain their high level of accuracy while performing faster analysis, which will be important as databases amass more information that has to be sorted speedily, according to a release of the study’s findings.
Knowledge-based computer-assisted detection (CAD) systems compare mammogram images to those of known cases of breast cancer in order to aid radiologists in their diagnosis. To avoid the systems getting bogged down with information, the researchers are presenting a Knowledge-Based Computer Assisted Detection (KB-CAD) system that analyzes breast masses using the principles of information theory.
The system uses an approach that compares new unknown cases with a restrictive search for that information that seems most useful, instead of every similar case in the database. Thus, the speed of results is much faster.
In the recent pilot study, the Duke researchers applied their technique to a database of 2,300 mammography images. With this type of indexing, the researchers compared a sample image to the top 600 most informative, cutting down their CAD system's processing time by one-fourth, to less than 3 seconds per query. The researchers expect to launch a larger study in a year to evaluate the clinical impact of this new approach.
This study was presented on August 1 at the 48th Annual Meeting of the American Association of Physicists in Medicine in Orlando.