Abstract
Background: Performing mammary ultrasonography with prior knowledge of the regions to which special attention should be paid makes it possible to obtain results with higher sensitivity.
Objectives: To test ultrasonographers for their mammogram-reading skill and to determine the test sensitivity.
Methods: We tested 15 ultrasonographers for their mammogram-reading skills on a total of 50 cases comprising 19 involving cancerous lesions, 4 involving benign lesions, and 26 normal cases (direction, MLO 1). For the question regarding diagnostic detection, the examinees were asked to show by illustration up to 2 sites per breast where attention should be paid when performing ultrasonography, and in the qualitative question, they were asked to select the right answer from among 4 options: "no abnormalities," "possible lesions," "benign lesions," or "possible malignant lesions."
Results: The mean sensitivity of the 15 ultrasonographers was 89% (74-100%). The sensitivities of 7 of them were 90% or higher, and 13 of the 15 attained sensitivities of 80% or higher. As for the qualitative diagnosis, there was some variability of the results among the readers.
Discussion: Although no special training is now given, it is beneficial to employ ultrasonographers with high mammogram-reading skills. Because reading by ultrasonographers increases the sensitivity of ultrasonography, a different concept is required from that of physicians, who place emphasis on specificity when reading.
Conclusion: It will be possible to obtain more accurate results if ultrasonographers read mammograms before performing ultrasonography.
