We have shown previously, in the context of computer‐aided diagnosis (CAD), that information derived from multiple images of the same patient can be used to improve diagnostic performance. In that work, we ignored the correlation among multiple images of the same patient. In the present study, we investigate theoretically, within the framework of receiver operating characteristic (ROC) analysis, the effect of correlation on three methods for combining quantitative diagnostic information from two images: taking the average, the maximum, and the minimum of a pair of normally distributed decision variables. We assume, as in our previous work, that the quantitative diagnostic information obtained from the two images of a given patient can be transformed monotonically to two latent decision variables that are normally distributed. Similar to the situation of uncorrelated images, we found that (1) the average always improves the area under the ROC curve (AUC) compared to the single‐view image; (2) the maximum and the minimum can also, but not always, improve the AUC; and (3) each method can be the best method in certain situations. In addition, as the correlation strength increases, the average performs the best less often, whereas the maximum and the minimum perform the best more often. These theoretical results are illustrated with analysis of a mammography study.
This paper describes the application of the area and contrast of mammographic microcalcifications to computer-aided diagnostic schemes. Image contrast (measured in differences in optical density on the film) is converted to radiation contrast (in terms of log x- ray exposure) by correcting for the characteristic curve of the screen-film system and by correcting for the loss in contrast caused by the blurring by the screen and the film digitizer. From the radiation contrast, we estimate an effective thickness of a microcalcification that would have produced the corresponding radiation contrast. By examining the relationship between effective thickness and size of computer-detected signals (potential microcalcifications), the false-positive rate of our automated detection scheme can be reduced from 2.5 to 1.5 false clusters per image, while maintaining a sensitivity of 85%. We have also conducted two preliminary studies for which the extraction technique may be beneficial. The first was for classifying clusters as either benign or malignant. Four features were identified: the standard deviation in area, thickness, and effective volume of microcalcifications within a given cluster, and the mean effective volume of microcalcifications within the cluster. The second study was for developing a quantitative measure of the subtlety of appearance of microcalcifications in mammograms. We have found that the product of the area and image contrast summed over all microcalcifications within a cluster correlates well with human subjective impression of subtlety.
Objectives The objective of this study was to test the hypothesis that the metabolic tumor volume (MTV) of primary non-small-cell lung cancer is not sensitive to differences in 18F-fluorodeoxyglucose (18F-FDG) uptake time, and to compare this consistency of MTV measurements with that of standardized uptake value (SUV) and total lesion glycolysis (TLG). Methods Under Institutional Review Board approval, 134 consecutive patients with histologically proven non-small-cell lung cancer underwent 18F-FDG PET/computed tomography scanning at about 1 h (early) and 2 h (delayed) after intravenous injection of 18F-FDG. MTV, SUV, and TLG of the primary tumor were all measured. Student’s t-test and Wilcoxon’s signed-rank test for paired data were used to compare MTV, SUV, and TLG between the two scans. The intraclass correlation coefficient (ICC) was used to assess agreement in PET parameters between the two scans and between the measurements made by two observers. Results MTV was not significantly different (P=0.17) between the two scans. However, SUVmax, SUVmean, SUVpeak, and TLG increased significantly from the early to the delayed scans (P<0.0001 for all). The median percentage change between the two scans in MTV (1.65%) was smaller than in SUVmax (11.76%), SUVmean(10.57%), SUVpeak(13.51%), and TLG (14.34%); the ICC of MTV (0.996) was greater than that of SUVmax (0.933), SUVmean (0.952), SUVpeak (0.928), and TLG (0.982). Interobserver agreement between the two radiologists was excellent for MTV, SUV, and TLG on both scans (ICC: 0.934–0.999). Conclusion MTV is not sensitive to common clinical variations in 18F-FDG uptake time, its consistency is greater than that of SUVmax, SUVmean, SUVpeak, and TLG, and it has excellent interobserver agreement.
Purpose To determine whether prostate‐specific antigen (PSA) levels adjusted by prostate and zonal volumes estimated from magnetic resonance imaging (MRI) improve the diagnosis of prostate cancer (PCa) and differentiation between patients who harbor high‐Gleason‐sum PCa and those without PCa. Materials and Methods This retrospective study was Health Insurance Portability and Accountability Act (HIPAA)‐compliant and approved by the Institutional Review Board of participating medical institutions. T 2 ‐weighted MR images were acquired for 61 PCa patients and 100 patients with elevated PSA but without PCa. Computer methods were used to segment prostate and zonal structures and to estimate the total prostate and central‐gland (CG) volumes, which were then used to calculate CG volume fraction, PSA density, and PSA density adjusted by CG volume. These quantities were used to differentiate patients with and without PCa. Area under the receiver operating characteristic curve (AUC) was used as the figure of merit. Results The total prostate and CG volumes, CG volume fraction, and PSA density adjusted by the total prostate and CG volumes were statistically significantly different between patients with PCa and patients without PCa ( P ≤ 0.007). AUC values for the total prostate and CG volumes, and PSA density adjusted by CG volume, were 0.68 ± 0.04, 0.68 ± 0.04, and 0.66 ± 0.04, respectively, and were significantly better than that of PSA ( P < 0.02), for differentiation of PCa patients from patients without PCa. Conclusion The total prostate and CG volumes estimated from T 2 ‐weighted MR images and PSA density adjusted by these volumes can improve the effectiveness of PSA for the diagnosis of PCa and differentiation of high‐Gleason‐sum PCa patients from patients without PCa. J. MAGN. RESON. IMAGING 2015;42:1733–1739.
A summary is not available for this content so a preview has been provided. Please use the Get access link above for information on how to access this content.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)