Dear AuntMinnie Member,
A new study is fueling debate over the accuracy of computer-aided detection (CAD) in screening mammography. A group from a Utah healthcare system found that installing CAD enabled them to find more breast cancers without significantly boosting patient recall rates.
The researchers reported on their experiences after implementing mammography CAD software, studying its impact on breast screening in a population of more than 43,000 patients, according to an article we're featuring in our Advanced Visualization Digital Community. They wanted to assess CAD's impact on the recall rate, which some studies have found tends to rise after CAD implementation.
The researchers found that CAD did not have a statistically significant impact on the recall rate at their facilities. CAD also found some cancers that otherwise would have gone undetected, particularly in younger women with dense breast tissue. Get the details on the study by clicking here.
In another article in the community, radiologists from Stanford University in California put their homegrown CAD application to the test in a dataset of multidetector CT chest images. They wanted to see if there were any patterns in cases when radiologists don't "believe" CAD markings.
The group found that certain types of lesions did indeed create discrepancies between what the CAD application found suspicious and what radiologists believed was worth working up. Learn more by clicking here, or visit our Advanced Visualization Digital Community at av.auntminnie.com.
