Objectives: This study examines recognition performance to depend on image context and time order error. Recognition of standard images is a basic process in medical image analysis. Methods: After the presentation of a standard square, 20 subjects identified the standard within a variety of 7 squares. The choice was between the standard and either 3 smaller and 3 larger squares, 5 smaller and 1 larger square, or 5 larger and 1 smaller square (context conditions). Results: Multilevel regression analysis showed large individual differences in judgments (P < .001). Context induced assimilation of judgments to the medium-sized square within response options (P < .001). Negative time order error in rapid judgments caused an underestimation of the standard (P < .001). Conclusions: Assimilation of judgments and time order error might be a threat to the reliability of medical image analysis. Some procedural recommendations are derived to reduce bias and increase patient safety in radiology.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.