作者: R. C. Warren , C. J. Darwin
DOI: 10.1259/0007-1285-62-741-817
关键词: Artificial intelligence 、 Prior probability 、 Observer (special relativity) 、 Abnormality 、 Receiver operating characteristic 、 Detection theory 、 Iterative reconstruction 、 Computer graphics 、 Radiography 、 Mathematics 、 Pattern recognition
摘要: Abstract The ability of an observer to detect variations in size a geometrical image feature have been investigated using receiver operating characteristic (ROC) analysis. Three types were constructed computer graphics: disc-shaped targets variable radius, model chest radiographs showing heart diameter and arterial angiograms with vessel width. Five factors investigated: experience, variation detectability theoretical signal-to-noise ratio, the prior probability presence abnormality, viewing distance, uncertainty location abnormality. In all but one experiment, excellent agreement was found between measured detectabilities predictions signal detection theory, providing initial practice session included for each observer. No significant six different probabilities two distances, reduction fo...