Computer-aided mammography: a case study of coping with fallibility in a skilled decision-making task.

Abstract

Breast screening requires radiologists to exercise keen perceptual skills to find what may be faint, small features, and sophisticated interpretative skills to classify them correctly. Understandably, radiologists sometimes make errors, and evidence suggests that employing a computer prompting aid can reduce them. To investigate prompting aid requirements, we have studied both current reading practices and radiologists reading with prompts. These studies have enabled us to understand better how radiologists manage errors in current practice, and how they deal with prompting aid errors. They also show that such aids may get used in ways quite different from those originally envisaged.