Answers are grounded in this paper and cite their sources.
Screening: Interpreting screening mammograms is difficult; radiologists must maintain specificity while reading films with a very low incidence of cancer. Estimates suggest sensitivity is no higher than 75%. CAD: Commercial systems are now available that prompt radiologists to look at suspicious regions. Such systems achieve high sensitivity but at low specificity. Evaluations: Evaluations of these CAD systems reveal that they do well at detecting microcalcifications but less well at detecting masses. Impact on decision-making: The low specificity of the computer prompts seems to hinder their capacity to influence radiologists' decision-making. Prospective trials: Only one prospective trial has been carried out of a CAD system for mammography. A replication of this trial in a UK setting is awaited.
No takes yet. Share an insight, caveat, or question.
Paul Taylor (2002) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: