TY - JOUR
T1 - Computed Tomography Pulmonary Angiography in Acute Pulmonary Embolism: The Effect of a Computer-assisted Detection Prototype Used as a Concurrent Reader
AU - Wittenberg, Rianne
AU - Peters, Joost F.
AU - van den Berk, Inge A. H.
AU - Freling, Nicole J. M.
AU - Lely, Rutger
AU - de Hoop, Bartjan
AU - Horsthuis, Karin
AU - Ravesloot, Cécile J.
AU - Weber, Michael
AU - Prokop, W. Mathias
AU - Schaefer-Prokop, Cornelia M.
AU - Lelij, R.J.
PY - 2013
Y1 - 2013
N2 - Purpose: To assess the effect of computer-assisted detection (CAD) on diagnostic accuracy, reader confidence, and reading time when used as a concurrent reader for the detection of acute pulmonary embolism in computed tomography pulmonary angiography.Materials and Methods: In this institutional review board-approved retrospective study, 6 observers with varying experience evaluated 158 negative and 38 positive consecutive computed tomography pulmonary angiographies (mean patient age 60 y; 115 women) without and with CAD as a concurrent reader. Readers were asked to determine the presence of pulmonary embolism, assess their diagnostic confidence using a 5-point scale, and document their reading time. Results were compared with an independent standard established by 2 readers, and a third chest radiologist was consulted in case of discordant findings.Results: Using logistic regression for repeated measurements, we found a significant increase in readers' sensitivity (P <0.001) without loss of specificity (P=0.855) with the effects being reader dependent (P <0.001). Sensitivities varied from 68% to 100% without CAD and from 76% to 100% with CAD. A 2-way analysis of variance showed a small but significant decrease in reading time (P <0.001), with the duration varying between 24 and 208 seconds without CAD and between 17 and 196 seconds with CAD, and a significant increase in readers' confidence scores using CAD as a concurrent reader (P <0.001).Conclusions: CAD as a concurrent reader has the potential to increase readers' sensitivity and confidence with a decrease in reading time without loss of specificity. The differences between readers, however, require further evaluation of CAD as a concurrent reader in a larger trial before stronger conclusions can be drawn
AB - Purpose: To assess the effect of computer-assisted detection (CAD) on diagnostic accuracy, reader confidence, and reading time when used as a concurrent reader for the detection of acute pulmonary embolism in computed tomography pulmonary angiography.Materials and Methods: In this institutional review board-approved retrospective study, 6 observers with varying experience evaluated 158 negative and 38 positive consecutive computed tomography pulmonary angiographies (mean patient age 60 y; 115 women) without and with CAD as a concurrent reader. Readers were asked to determine the presence of pulmonary embolism, assess their diagnostic confidence using a 5-point scale, and document their reading time. Results were compared with an independent standard established by 2 readers, and a third chest radiologist was consulted in case of discordant findings.Results: Using logistic regression for repeated measurements, we found a significant increase in readers' sensitivity (P <0.001) without loss of specificity (P=0.855) with the effects being reader dependent (P <0.001). Sensitivities varied from 68% to 100% without CAD and from 76% to 100% with CAD. A 2-way analysis of variance showed a small but significant decrease in reading time (P <0.001), with the duration varying between 24 and 208 seconds without CAD and between 17 and 196 seconds with CAD, and a significant increase in readers' confidence scores using CAD as a concurrent reader (P <0.001).Conclusions: CAD as a concurrent reader has the potential to increase readers' sensitivity and confidence with a decrease in reading time without loss of specificity. The differences between readers, however, require further evaluation of CAD as a concurrent reader in a larger trial before stronger conclusions can be drawn
U2 - https://doi.org/10.1097/RTI.0b013e3182870b97
DO - https://doi.org/10.1097/RTI.0b013e3182870b97
M3 - Article
C2 - 23486230
SN - 0883-5993
VL - 28
SP - 315
EP - 321
JO - Journal of thoracic imaging
JF - Journal of thoracic imaging
IS - 5
ER -