Search

Background Coronary artery calcification is a significant contributor to reduced accuracy of coronary computed tomographic angiography (CTA) in the assessment of coronary artery disease severity. The aim of the current study is to assess the impact of a prototype calcium deblooming algorithm on the diagnostic accuracy of CTA. Methods 40 patients referred for invasive catheter angiography underwent CTA and invasive catheter angiography. The CTA were reconstructed using a standard soft tissue kernel (CTASTAND) and a deblooming algorithm (CTADEBLOOM). CTA studies were read with and without the deblooming algorithm blinded to the invasive coronary angiogram findings. Sensitivity, specificity, accuracy, positive predictive value and negative predictive value for the detection of stenosis ≥50% or ≥70% were evaluated using quantitative coronary angiography as the reference standard. Image quality was assessed using a 5-point scale, and the presence of image artifact recorded. Results All studies were diagnostic with 548 segments available for evaluation. Image score was 3.64 ± 0.72 with CTADEBLOOM, versus 3.56 ± 0.72 with CTASTAND (p = 0.38). CTADEBLOOM had significantly less calcium blooming artifact than CTASTAND (12.5% vs. 47.5%, p = 0.001). Based on a 50% stenosis threshold for defining significant disease, the Sensitivity/Specificity/PPV/NPV/Accuracy were 65.9/84.9/27.6/96.6/83.4 for CTADEBLOOM and 75.0/81.9/26.6/97.4/81.4 for CTASTAND using a ≥50% threshold. CTADEBLOOM specificity was significantly higher than CTASTAND (84.9% vs. 81.5%, p = 0.03), with no difference between the algorithms in sensitivity (p = 0.22), or accuracy (p = 0.15). These results remained unchanged when a stenosis threshold of ≥70% was used. Interobserver agreement was fair with both techniques (CTADEBLOOM k = 0.38, CTASTAND k = 0.37). Conclusion In this proof of concept study, coronary calcification deblooming using a prototype post-processing algorithm is feasible and reduces calcium blooming with an improvement of the specificity of the CTA exam.

Purpose After chemotherapy, approximately 50% of patients with metastatic testicular germ cell tumors (GCTs) who undergo retroperitoneal lymph node dissections (RPNLDs) for residual masses have fibrosis. Radiomics uses image processing techniques to extract quantitative textures/features from regions of interest (ROIs) to train a classifier that predicts outcomes. We hypothesized that radiomics would identify patients with a high likelihood of fibrosis who may avoid RPLND. Patients and Methods Patients with GCT who had an RPLND for nodal masses > 1 cm after first-line platinum chemotherapy were included. Preoperative contrast-enhanced axial computed tomography images of retroperitoneal ROIs were manually contoured. Radiomics features (n = 153) were used to train a radial basis function support vector machine classifier to discriminate between viable GCT/mature teratoma versus fibrosis. A nested 10-fold cross-validation protocol was used to determine classifier accuracy. Clinical variables/restricted size criteria were used to optimize the classifier. Results Seventy-seven patients with 102 ROIs were analyzed (GCT, 21; teratoma, 41; fibrosis, 40). The discriminative accuracy of radiomics to identify GCT/teratoma versus fibrosis was 72 ± 2.2% (area under the curve [AUC], 0.74 ± 0.028); sensitivity was 56.2 ± 15.0%, and specificity was 81.9 ± 9.0% ( P = .001). No major predictive differences were identified when data were restricted by varying maximal axial diameters (AUC range, 0.58 ± 0.05 to 0.74 ± 0.03). The prediction algorithm using clinical variables alone identified an AUC of 0.76. When these variables were added to the radiomics signature, the best performing classifier was identified when axial masses were limited to diameter < 2 cm (accuracy, 88.2 ± 4.4; AUC, 0.80 ± 0.05; P = .02). Conclusion A predictive radiomics algorithm had a discriminative accuracy of 72% that improved to 88% when combined with clinical predictors. Additional independent validation is required to assess whether radiomics allows patients with a high predicted likelihood of fibrosis to avoid RPLND.

OBJECTIVE. The purpose of this study was to determine whether integrated 18F-fluorocholine (FCH) PET whole-body MRI (PET/WBMRI) depicts lymph node and distant metastases in patients with high-risk ...

Although a common occurrence, cystic lesions of the pancreatico-biliary tree (PBT) may pose a diagnostic dilemma because they encompass a large number of neoplastic and benign processes with varied clinical symptoms. Knowledge of lesion classification and characterization are essential in making an accurate prospective diagnosis. This is necessary for identifying clinically significant cystic masses, which at times may require invasive intervention from indolent, nonneoplastic lesions, for which surveillance may suffice. Today, there is an arsenal of modalities for assessing the PBT, however, magnetic resonance imaging (MRI) remains at the forefront for characterizing cystic morphology and fluid content, internal septations, solid component, enhancement patterns, as well as assessing the surrounding normal structures. This pictorial review aims to review the spectrum of MRI features, which will aid in the differential diagnoses of cystic lesions of the PBT and mimickers, enabling the radiologist to reach a more confident diagnosis.

PURPOSE: To examine the prognostic implication of fractional flow reserve (FFR) derived from coronary CT (FFR(CT)) in routine clinical practice. MATERIALS AND METHODS: Patients referred for FFR(CT) analysis at a single center between October 2015 and June 2017 were retrospectively included and followed up for rates of invasive angiography and clinical events. Two hundred seven patients underwent successful FFR(CT) analysis with seven lost to follow-up, leaving 200 (mean age ± standard deviation, 62.4 years ± 10.0; 49 [24.5%] women) patients for analysis. At coronary CT angiography, patients were categorized as having significant stenosis (SS) in the presence of a diameter stenosis greater than or equal to 50% (hereafter, SS positive) and flow limitation in the presence of a postlesion (that is, FFR(CT) measured 2 cm to the distal aspect of the lesion) FFR(CT) less than 0.80 (hereafter, FFR(CT) positive). Vessel-oriented clinical events (VOCEs) were defined as vessel-related late revascularization (>90 days), myocardial infarction, and cardiac mortality. RESULTS: At CT angiography, 130 (65%) studies were SS positive and 63 (31.5%) were FFR(CT) positive. At median follow-up of 477 days (range, 252–859 days), there were 26 VOCE end points in 22 patients: 22 revascularizations and four nonfatal myocardial infarctions. VOCE end points occurred in zero of 58 (0%) of SS-negative and FFR(CT) negative patients, in eight of 79 (10.1%) of SS-positive and FFR(CT)-negative patients, in zero of 12 (0%) of SS-negative and FFR(CT)-positive patients, and in 18 of 51 (35.3%) of SS-positive and FFR(CT)-positive patients (log-rank χ(2) = 30.1; P < .001). At multivariable Cox regression, both FFR(CT) (hazard ratio per 0.1 decrease, 1.54 [95% confidence interval: 1.1, 2.2] P = .013) and stenosis (hazard ratio per unit increase, 2.16 [95% confidence interval: 1.25, 3.72] P = .006) were independently associated with VOCE. CONCLUSION: Stenosis and FFR(CT) are independent predictors of intermediate-term outcomes. In the absence of a stenosis greater than 50%, a positive FFR(CT) result is not associated with an increased intermediate risk. © RSNA, 2019 Supplemental material is available for this article. See also commentary by Fairbairn and Bull in this issue.

Introduction Coronary artery calcification is a significant contributor to reduced accuracy of coronary computed tomographic angiography (CTA) in the assessment of coronary artery disease severity. The aim of the current study is to assess the impact of a prototype calcium deblooming algorithm on the diagnostic accuracy of CTA. Methods 40 patients referred for invasive catheter angiography underwent CTA and invasive catheter angiography. CTA studies were read with and without the deblooming algorithm blinded to the invasive coronary angiogram findings. Sensitivity, specificity, accuracy, positive predictive value and negative predictive value for the detection of stenosis ≥50% were evaluated using quantitative coronary angiography as the reference standard. Results All studies were diagnostic with 581 segments available for evaluation. Image score was 3.64±0.72 with CTADEBLOOM, versus 3.56±0.72 with CTASTAND (p=0.38). CTADEBLOOM had significantly less calcium blooming artifact than CTASTAND (12.5% vs. 47.5%, p=0.001). The Sensitivity/Specificity/PPV/NPV/Accuracy were 64.4/85.2/27.6/96.5/83.5 for CTADEBLOOM and 75.0/81.6/25.8/97.5/81.1 for CTASTAND. CTADEBLOOM specificity was significantly higher than CTASTAND (85.2% vs. 81.6%, p=0.017), with no difference between the algorithms in sensitivity (p=0.22), or accuracy (p= 0.09). Interobserver agreement was fair with both techniques (CTADEBLOOMk= 0.38, CTASTAND k= 0.37). Conclusion Coronary calcification deblooming using a prototype post-processing algorithm is feasible and significantly reduces calcium blooming with an improvement of the specificity of the CTA exam.

OBJECTIVE: To describe dynamic (18)F-flumethycholine PET (dPET) and dynamic contrast enhancement MR (DCE MR) parameters in localized high-risk prostate cancer (PCa), and determine whether these differ from normal prostate. Furthermore, to determine whether a correlation exists between dPET and DCE MR parameters. METHODS: 41 consenting patients who underwent prostate DCE MR and dPET were included in this institutionally approved study. Intraprostatic lesions on MR were assigned a PI-RADS v2 score, and focal lesions on PET were documented. All lesions were correlated with pathology. Quantitative and semi-quantitative DCE MR and two-tissue compartmental model dPET parameters were determined and tumor-to-normal gland ratios (T/N) for these parameters were calculated. Finally, dPET and DCE MR correlation was estimated using Spearman correlation coefficients. RESULTS: There were 46 malignant lesions per standard of reference. On dPET, peripheral zone (PZ) tumors had higher K1 (p < 0.001), and a T/N ratio ≥2 was significant (p < 0.001). On DCE MR, the parameters in, kep, Ktrans and quantitative iAUC were higher for PZ and non-PZ tumors than corresponding normal tissue (p < 0.001); for PZ tumors, a T/N ratio ≥ 1.5 for Ktrans and pei was significant (p = 0.0019 and 0.0026, respectively). Moderate Spearman correlation (0.40 < ρ < 0.59) was found between dPET K1 and DCE MR Ktrans and pei. CONCLUSION: In patients with high-risk PCa, quantitative dPET and DCE-MR parameters in primary tumors differ from normal tissue. Only moderate correlation exists between K1 (dPET) and Ktrans and pei (DCE MR). The incremental value of any of these parameters to PI-RADS v2 warrants further investigation. ADVANCES IN KNOWLEDGE: Unique quantitative and semi-quantitative FCH PET/MR parameters in PCa differ from normal gland, and should be further investigated to determine their potential contribution to PI-RADS v2 in the detection of clinically significant PCa.

Purpose: Patients with locally advanced rectal cancer (LARC) achieving pathologic complete response (pCR) to neoadjuvant chemoradiotherapy (CRT) have significantly improved long term survival. Preoperative detection of pCR may enable a conservative therapeutic approach in some patients. The purpose of the current prospective pilot study was to assess multiparametric qualitative and quantitative MR, PET, PET-MR and tumor texture features in predicting pCR to CRT in patients with LARC. Material and Methods: Eighteen LARC patients underwent staging with FDG-PET and MR-rectum and 15 had post-CRT restaging. Response was assessed qualitatively and quantitatively. SUV (tumor/background), SUV/ADC, and tumor texture parameters derived via machine learning algorithms (MLA) from PET and multiple MR sequences and were correlated with histopathology. Results: A third of patients had pCR. Sensitivity, specificity & accuracy of PET, MR and combined PET-MR were 90, 60, & 80; 90, 20 & 66.7; 90, 80 & 86.7, respectively. Differences did not reach statistical significance. Quantitatively, only tumor-muscle (SUV/ADC) ratio improved prediction of pCR. Of all texture features assessed using MLA, only the classifier trained on pre-treatment PET was significant (p = 0.034; accuracy, 92.8%). Combined PET and MR texture features did not improve performance. Conclusion: Combined PET-MR may improve specificity compared with PET or MR alone, although this needs to be validated in a larger cohort. Tumor to muscle SUV/ADC ratios post-therapy and texture features on baseline PET show promise in improving prediction of pCR post-CRT in LARC.

Objective:To describe dynamic 18F-flumethycholine PET (dPET) and dynamic contrast enhancement MR (DCE MR) parameters in localized high-risk prostate cancer (PCa), and determine whether these differ...

The largest randomized controlled trial (RCT) on the effect of FDG-PET on surgical management for metastatic colorectal adenocarcinoma to liver (“PET-CAM”) reported only a modest change in surgical management (8%). To explore the relationship between prior chemotherapy and detection of metastatic disease on PET in patients from PET-CAM. Secondary aim: to determine whether centralized imaging interpretation could have impacted trial results. The study included 120 patients from a single institution. Local PET interpretation (PET-L) was recorded from the original database. Retrospective PET interpretation was performed independently by at least one additional reader (PET-C). The presence of extrahepatic disease (EHD) and significant additional liver metastases (=SALM), defined as metastases not originally planned for resection, was recorded. Patients were stratified to responders to recent chemotherapy (Group R) versus all others (Group O) according to surgical pathology and RECIST criteria. Thirty-seven of 50 patients who received recent chemotherapy (

Aim: To determine the incidence of natural growth or regression of focal nodular hyperplasia (FNH) in the liver. Material and Methods: We retrospectively included 120 consecutive patients who were diagnosed to have FNH on MRI. The mean follow-up duration was 19 months (range: 6-64 months). There were 25 men and 95 women (age range: 18-80 years; mean: 45 years). There were 167 FNH lesions in the 120 patients. MRI images were retrospectively reviewed for interval growth or regression of FNH. The maximum size of the lesions was measured on axial arterial-phase images of the initial and the last MRI examinations. An interval increase or decrease in diameter of over 10% of the initial diameter was considered as positive growth or regression, respectively. The use of Oral contraceptives was also documented. Results: Interval growth was seen in 25/167 nodules (15%) over 7-48 months (mean: 21 months), with increase in size of 0.2-1.7 cm (mean: 0.6 cm) and percentage change of 10.5-340% (mean: 64%). Interval regression was seen in 13/167 (8%) of nodules over 7-63 months (mean: 22 months), with decrease in size of 0.2-0.9 cm (mean: 0.5 cm) and percentage change of 10.4-60% (mean: 24%).Five of 17 (29%) female patients with growing FNH and 25/78 (32%) female patients with non-growing FNH had a history of intake of oral contraceptives (P=0.83). Conclusions: Although FNH is benign and of no clinical significance, a substantial percentage of FNH shows interval growth or regression on long-term follow-up with MRI.

The objective of our study was to assess the association between colorectal liver metastases and intrahepatic bile duct dilatation and compare its frequency with the frequency of intrahepatic bile duct dilatation in other common noncolorectal metastases and hepatocellular carcinoma (HCC).A retrospective review of the radiology database was performed to obtain the records of patients who underwent CT for staging of histologically proven colorectal carcinoma (n = 1,000), noncolorectal carcinomas (n = 1,000), and HCC (n = 226). The CT scans of the 2,226 patients were reviewed for the presence of liver metastases and masses, the presence of intrahepatic bile duct dilatation directly related to the masses, the size of the largest lesion, and the associated caliber of the dilated intrahepatic bile duct.Of the 297 patients with colorectal liver metastases, 49 (16.5%) had intrahepatic bile duct dilatation, whereas of the 263 patients with noncolorectal liver metastases, only eight (3%) had dilated ducts. Thirteen (5.7%) of the 226 patients with HCC had intrahepatic bile duct dilatation. Intrahepatic bile duct dilatation was present in significantly more patients with colorectal liver metastases (p0.05) than those with noncolorectal and HCC metastases. The mean sizes of the lesions and mean caliber of the dilated bile duct from colorectal metastases, noncolorectal metastases, and HCC were 5.3, 3.9, and 5.6 cm, respectively, and 5.1, 4.6, and 4.8 mm, respectively. Overall lesion size irrespective of diagnosis had no significant correlation with the probability of bile duct dilatation (p = 0.16; odds ratio = 1.08; 95% CI, 0.97-1.20). Lesion size also did not have a significantly different effect on the probability of dilatation for the three diagnoses (p = 0.71). The caliber of bile duct dilatation did not significantly differ among the three groups of patients by diagnosis (p = 0.70).Colorectal liver metastases have a significantly higher tendency to cause intrahepatic bile duct dilatation than noncolorectal metastases and HCC due to the characteristic features, such as cholangiocarcinoma, of intrabiliary growth. This association has important diagnostic, surgical, and prognostic implications that make it necessary to highlight this finding while communicating with clinical colleagues.

4559 Background: After chemotherapy, > 50% of patients (pts) with metastatic testicular GCT who undergo retroperitoneal lymph node dissection (RPNLD) for residual masses are found to have fibrosis (F) alone on pathological examination. To minimize overtreatment, better prediction algorithms are needed to identify pts with F who can avoid RPLND. Radiomics uses image processing techniques to extract quantitative textures/features from tumor regions of interest (ROI) to train a classifier that predicts pathological findings. We hypothesized that radiomics may identify pts with a high predicted likelihood of F who may avoid RPLND. Methods: Pts with GCT who had an RPLND for nodal masses > 1cm after first line platinum chemotherapy were included. Preoperative contrast enhanced axial CT images of retroperitoneal ROI were manually contoured. 153 radiomics features trained a radial basis function support vector machine classifier to discriminate between viable GCT /Mature Teratoma (T) vs F. Nested ten-fold cross-validation protocol was employed to determine classifier accuracy. Clinical variables and restricted size criteria were used to optimize the classifier. Results: A total of 82 pts with 102 ROI were analyzed (GCT: 21; T: 41; F: 40). The discriminative accuracy of radiomics to identify GCT/T vs F was 72%(±2.2)(AUC: 0.74 (±0.028); positive predictive value: 67% (48-92%); negative predictive value: 74% (62-84%)(p = 0.001)). No major predictive differences were identified when data was restricted by varying maximal axial diameters (AUC range: 0.58(±0.05) - 0.74(±0.03)). Prediction algorithm using clinical variables alone identified an AUC of 0.71 (±0.15). When these variables were added to the radiomic signature, the best performing classifier was identified when axial tumors were limited to diameter < 2cm (accuracy: 88.2 (±4.4); AUC: 0.80 (±0.05)(p = 0.02)). Conclusions: A predictive radiomics algorithm had an overall discriminative accuracy of 72% that improved to 88% when combined with clinical details. Further independent validation is required to assess whether radiomics, in conjunction with standard clinical predictors, may allow pts with a high predicted likelihood of F to avoid RPLND.