Observer Performance with Varying Radiation Dose and Reconstruction Methods for Detection of Hepatic Metastases

Abstract

Purpose To estimate the ability of lower dose levels and iterative reconstruction (IR) to display hepatic metastases that can be detected by radiologists. Materials and Methods Projection data from 83 contrast agent-enhanced CT examinations were collected. Metastases were defined by histopathologic analysis or progression and regression. Lower radiation dose configurations were reconstructed at five dose levels with filtered back projection (FBP) and IR (automatic exposure control settings: 80, 100, 120, 160, and 200 quality reference mAs [QRM]). Three abdominal radiologists circumscribed metastases, indicating confidence (confidence range, 0-100) and image quality. Noninferiority was assessed by using jackknife alternative free-response receiver operating characteristic (JAFROC) analysis (noninferiority limit, -0.10) and reader agreement rules, which required identification of metastases identified at routine dose, and no nonlesion localizations in patients negative for metastases, in 71 or more patient CT examinations (of 83), for each configuration. Results There were 123 hepatic metastases (mean size, 1.4 cm; median volume CT dose index and size-specific dose estimate, 11.0 and 13.4 mGy, respectively). By using JAFROC figure of merit, 100 QRM FBP did not meet noninferiority criteria and had estimated performance difference from routine dose of -0.08 (95% confidence interval: -0.11, -0.04). Preset reader agreement rules were not met for 100 QRM IR or 80 QRM IR, but were met for doses 120 QRM or higher (ie, size-specific dose estimate ≥ 8.0 mGy). IR improved image quality (P < .05) but not reader performance. Other than 160 QRM IR, lower dose levels were associated with reduced confidence in metastasis detection (P < .001). Conclusion For detection of hepatic metastases by using contrast-enhanced CT, dose levels that corresponded to 120 quality reference mAs (size-specific dose estimate, 8.0 mGy) and higher performed similarly to 200 quality reference mAs with filtered back projection. © RSNA, 2018 Online supplemental material is available for this article.

Figure 1:

Specialized computer workstation that shows the reader axial (middle) and coronal (right) images in a patient with a small neuroendocrine metastasis (circle on middle image).

Figure 2:

Forest plot shows estimated differences in jackknife alternative free-response receiver operating characteristic figure of merit between observer performance at routine dose (200 quality reference mAs [QRM]) and lower radiation doses. Points represent the estimated difference with bars representing the estimated 95% confidence intervals of the differences; data in parentheses are 95% confidence intervals. FBP = filtered back projection; IR = iterative reconstruction.

Figure 3:

Images in a 69-year-old man with metastatic colon cancer who underwent contrast-enhanced abdominopelvic CT. Green circles indicate reference lesions marked by the reference reader and red circles indicate liver lesions interpreted as metastases by reader 3. Images corresponding to, A, 200 quality reference mAs (QRM) filtered back projection (FBP), B, 160 QRM iterative reconstruction (IR), C, 160 QRM FBP, and, D, 120 QRM IR, with three of three abdominal radiologists identifying the larger segment VII (black arrows in E and F), and three, two, none, and two abdominal radiologist readers detecting the smaller segment VIII metastasis, respectively (white arrows in E and F). E, Image without reader or reference markings at routine dose (200 QRM FBP); and, F, the image without reader markings corresponding to 120 QRM IR.

Figure 4:

Contrast-enhanced CT images in a 61-year-old woman with a small colorectal cancer metastasis (arrows) and focal fat adjacent to the falciform ligament (arrowheads). Routine-dose images, A, without and, B, with reference standard and reader markings (B, green is reference standard and red is reader). At the routine radiation dose, the three subspecialized readers identified the small metastasis (arrow in A). C and D, Half-dose (ie, by using an automatic exposure control setting of 100 quality reference mAs), C, without and, D, with reference standard and reader markings (D, green is reference and red is reader). At this dose level, only one reader was able to identify this lesion as a metastasis.

Figure 5:

Diagnostic image quality rating at each dose level and reconstruction combination. Numbers along the horizontal axis are automatic exposure control settings (ie, quality reference milliampere-seconds [QRM]) for each dose-reconstruction configuration. Image ratings are significantly different from the reference dose (P < .05), and 160 QRM with iterative reconstruction (IR, red) has a higher image quality compared with routine-dose 200 QRM with filtered back projection (FBP, blue), but all other dose-reconstruction configurations have worse image quality (P = .01). Each black dot represents the mean diagnostic image quality rating for each dose level and reconstruction combination.

Figure 6a:

(a) Pooled generalized estimating equation (GEE) estimates of lesion-level sensitivity versus reader confidence for hepatic metastases at different dose-reconstruction configurations. The denominator for sensitivity was the 123 hepatic metastases in the study. Sensitivity fell as reader confidence increased. At confidence of 0, the sensitivity represented pure detection (spatial colocalization). At confidence of 100, the lesion was detected and assigned a confidence score of 100. At any reader confidence threshold, sensitivity was lower for lower dose configurations. At any sensitivity, primary task confidence was lower at lower dose levels. (b) Individual performance of each reader (gray lines) is shown along with the pooled estimates (black line) for four example dose levels. FBP = filtered back projection, IR = iterative reconstruction, QRM = quality reference milliampere-second.

Figure 6b:

(a) Pooled generalized estimating equation (GEE) estimates of lesion-level sensitivity versus reader confidence for hepatic metastases at different dose-reconstruction configurations. The denominator for sensitivity was the 123 hepatic metastases in the study. Sensitivity fell as reader confidence increased. At confidence of 0, the sensitivity represented pure detection (spatial colocalization). At confidence of 100, the lesion was detected and assigned a confidence score of 100. At any reader confidence threshold, sensitivity was lower for lower dose configurations. At any sensitivity, primary task confidence was lower at lower dose levels. (b) Individual performance of each reader (gray lines) is shown along with the pooled estimates (black line) for four example dose levels. FBP = filtered back projection, IR = iterative reconstruction, QRM = quality reference milliampere-second.