Architectural distortion, the non-mass but potentially ominous clinical feature observed in many breast imaging procedures, is less likely to signal malignancy when it’s detected on screening mammography rather than diagnostic mammography or when it does not correlate with a subsequent targeted ultrasound exam.