Xeromammographic false negatives were analyzed to ascertain the cause of the errors and determine what corrective measures could be taken. Of 52 cancers miscalled, 52% were not visualized and 48% were categorized as misinterpretations. The causes of these errors are discussed. The error rate in all but 8% of the former group appears to be irreducible, yet errors discussed in the latter group are thought to be correctable in a high percentage of cases. Recommendations to reduce the number of false negatives in this group are presented.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1148/133.2.297 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!