The proposed experiments will investigate the perceptual processes that observers (both non-radiologists and radiologists) use to detect and to discriminate between features on CT images. Features that simulate hepatic lesion and blood vessels will be superimposed on single and multiple images obtained from actual CT scans of water phantoms and clinical patients with normal livers. Experiments with single CT images will study how the physical """"""""signal-to-noise ratio"""""""" affects the observer's ability to: (1) detect features on images with distracting or confusing features, (2) detect pairs of features that differ in size and spatial separation, and (3) discriminate between pairs of features that differ in relative size, contrast or edge sharpness. Experiments with multiple CT images will study the observer's ability to combine information that is distributed across several contiguous CT images. In addition to investigating the standard methods of displaying an entire series of CT images simultaneously, these experiments will also develop and compare temporal display method that provide a rapid sequential presentation of the CT images. The analysis of observer performance will generate ROC curves from confidence ratings about the presence of a feature, or about the difference between two features on the CT image. The fitted ROC curves provide indices of the obervers' detection and discrimination abilities, independent of any changes in their decision criteria, which can be compared across different conditions of the experiment. The observer's ability to detect and to discriminate will be compared with a model based on the matched-filter detector, which combines the information from the physical image with an estimate of the likelihood that a feature is present at a specified location. The results will improve our understanding of the perceptual processes involved in the detection and discrimination of features on CT images and will contribute to the development of theoretical models that can be used to predict observer performance.
| Swensson, R G; Judy, P F (1996) Measuring performance efficiency and consistency in visual discriminations with noisy images. J Exp Psychol Hum Percept Perform 22:1393-415 |
| Seltzer, S E; Judy, P F; Swensson, R G et al. (1994) Flattening of the contrast-detail curve for large lesions on liver CT images. Med Phys 21:1547-55 |
| Judy, P F; Swensson, R G; Nawfel, R D et al. (1992) Contrast-detail curves for liver CT. Med Phys 19:1167-74 |
| Coughlin, B F; Seltzer, S E; Swensson, R G et al. (1992) Practices and attitudes about cathode-ray tube-based and film-based image interpretation. J Digit Imaging 5:50-3 |
| Seltzer, S E; Swensson, R G; Nawfel, R D et al. (1991) Visualization and detection-localization on computed tomographic images. Invest Radiol 26:285-94 |
| Seltzer, S E; Swensson, R G; Judy, P F et al. (1988) Size discrimination in computed tomographic images. Effects of feature contrast and display window. Invest Radiol 23:455-62 |
| Judy, P F; Swensson, R G (1987) Display thresholding of images and observer detection performance. J Opt Soc Am A 4:954-65 |
| Kijewski, M F; Judy, P F (1987) The noise power spectrum of CT images. Phys Med Biol 32:565-75 |
| Seltzer, S E; Judy, P F (1987) Gaining insight: improved images enhance understanding. Radiographics 7:1221-40 |
