Deep learning in CT colonography: differentiating premalignant from benign colorectal polyps.

OBJECTIVES: To investigate the differentiation of premalignant from benign colorectal polyps detected by CT colonography using deep learning. METHODS: In this retrospective analysis of an average risk colorectal cancer screening sample, polyps of all size categories and morphologies were manually segmented on supine and prone CT colonography images and classified as premalignant (adenoma) or benign (hyperplastic polyp or regular mucosa) according to histopathology. Two deep learning models SEG and noSEG were trained on 3D CT colonography image subvolumes to predict polyp class, and model SEG was additionally trained with polyp segmentation masks. Diagnostic performance was validated in an independent external multicentre test sample. Predictions were analysed with the visualisation technique Grad-CAM++. RESULTS: The training set consisted of 107 colorectal polyps in 63 patients (mean age: 63 ± 8 years, 40 men) comprising 169 polyp segmentations. The external test set included 77 polyps in 59 patients comprising 118 polyp segmentations. Model SEG achieved a ROC-AUC of 0.83 and 80% sensitivity at 69% specificity for differentiating premalignant from benign polyps. Model noSEG yielded a ROC-AUC of 0.75, 80% sensitivity at 44% specificity, and an average Grad-CAM++ heatmap score of ≥ 0.25 in 90% of polyp tissue. CONCLUSIONS: In this proof-of-concept study, deep learning enabled the differentiation of premalignant from benign colorectal polyps detected with CT colonography and the visualisation of image regions important for predictions. The approach did not require polyp segmentation and thus has the potential to facilitate the identification of high-risk polyps as an automated second reader. KEY POINTS: • Non-invasive deep learning image analysis may differentiate premalignant from benign colorectal polyps found in CT colonography scans. • Deep learning autonomously learned to focus on polyp tissue for predictions without the need for prior polyp segmentation by experts. • Deep learning potentially improves the diagnostic accuracy of CT colonography in colorectal cancer screening by allowing for a more precise selection of patients who would benefit from endoscopic polypectomy, especially for patients with polyps of 6-9 mm size.

Machine Learning-based Differentiation of Benign and Premalignant Colorectal Polyps Detected with CT Colonography in an Asymptomatic Screening Population: A Proof-of-Concept Study.

Background CT colonography does not enable definite differentiation between benign and premalignant colorectal polyps. Purpose To perform machine learning-based differentiation of benign and premalignant colorectal polyps detected with CT colonography in an average-risk asymptomatic colorectal cancer screening sample with external validation using radiomics. Materials and Methods In this secondary analysis of a prospective trial, colorectal polyps of all size categories and morphologies were manually segmented on CT colonographic images and were classified as benign (hyperplastic polyp or regular mucosa) or premalignant (adenoma) according to the histopathologic reference standard. Quantitative image features characterizing shape (n = 14), gray level histogram statistics (n = 18), and image texture (n = 68) were extracted from segmentations after applying 22 image filters, resulting in 1906 feature-filter combinations. Based on these features, a random forest classification algorithm was trained to predict the individual polyp character. Diagnostic performance was validated in an external test set. Results The random forest model was fitted using a training set consisting of 107 colorectal polyps in 63 patients (mean age, 63 years ± 8 [standard deviation]; 40 men) comprising 169 segmentations on CT colonographic images. The external test set included 77 polyps in 59 patients comprising 118 segmentations. Random forest analysis yielded an area under the receiver operating characteristic curve of 0.91 (95% CI: 0.85, 0.96), a sensitivity of 82% (65 of 79) (95% CI: 74%, 91%), and a specificity of 85% (33 of 39) (95% CI: 72%, 95%) in the external test set. In two subgroup analyses of the external test set, the area under the receiver operating characteristic curve was 0.87 in the size category of 6-9 mm and 0.90 in the size category of 10 mm or larger. The most important image feature for decision making (relative importance of 3.7%) was quantifying first-order gray level histogram statistics. Conclusion In this proof-of-concept study, machine learning-based image analysis enabled noninvasive differentiation of benign and premalignant colorectal polyps with CT colonography. © RSNA, 2021 Online supplemental material is available for this article.

Papillary vs clear cell renal cell carcinoma. Differentiation and grading by iodine concentration using DECT-correlation with microvascular density.

OBJECTIVES: Various imaging methods have been evaluated regarding non-invasive differentiation of renal cell carcinoma (RCC) subtypes. Dual-energy computed tomography (DECT) allows iodine concentration (IC) analysis as a correlate of tissue perfusion. Microvascular density (MVD) in histopathology specimens is evaluated to determine intratumoral vascularization. The objective of this study was to assess the potential of IC and MVD regarding the differentiation between papillary and clear cell RCC and between well- and dedifferentiated tumors. Further, we aimed to investigate a possible correlation between these parameters. METHODS: DECT imaging series of 53 patients with clear cell RCC (ccRCC) and 15 with papillary RCC (pRCC) were analyzed regarding IC. Histology samples were stained using CD31/CD34 monoclonal antibodies; MVD was evaluated digitally. Statistical analysis included performance of Mann-Whitney U test, ROC analysis, and Spearman rank correlation. RESULTS: Analysis of IC demonstrated significant differences between ccRCC and pRCC (p < 0.001). A cutoff value of ≤ 3.1 mg/ml at IC analysis allowed identification of pRCC with an accuracy of 86.8%. Within the ccRCC subgroup, G1/G2 tumors could significantly be differentiated from G3/G4 carcinomas (p = 0.045). A significant positive correlation between IC and MVD could be determined for the entire RCC cohort and the ccRCC subgroup. Limitations include the small percentage of pRCCs. CONCLUSIONS: IC analysis is a useful method to differentiate pRCC from ccRCC. The significant positive correlation between IC and MVD indicates valid representation of tumor perfusion by DECT. KEY POINTS: • Analysis of iodine concentration using DECT imaging could reliably distinguish papillary from clear cell subtypes of renal cell cancer (RCC). • A cutoff value of 3.1 mg/ml allowed a distinction between papillary and clear cell RCCs with an accuracy of 86.8%. • The positive correlation with microvascular density in tumor specimens indicates correct display of perfusion by iodine concentration analysis.

Optimized management of urolithiasis by coloured stent-stone contrast using dual-energy computed tomography (DECT).

BACKGROUND: We analysed in vitro the appearance of commonly used ureteral stents with dual-energy computed tomography (DECT) and we used these characteristics to optimize the differentiation between stents and adjacent stone. METHODS: We analysed in vitro a selection of 36 different stents from 7 manufacturers. They were placed in a self-build phantom model and measured using the SOMATOM® Force Dual Source CT-Scanner (Siemens, Forchheim, Germany). Each sample was scanned at various tube potentials of 80 and 150 peak kilovoltage (kVp), 90 and 150 kVp and 100 and 150 kVp. The syngo Post-Processing Suite software program (Siemens, Forchheim, Germany) was used for differentiation based on a 3-material decomposition algorithm (UA, calcium, urine) according to our standard stone protocol. RESULTS: Stents composed of polyurethane appeared blue and silicon-based stents were red on the image. The determined appearances were constant for various peak kilovoltage (kVp) values. The coloured stent-stone-contrast displayed on DECT improves monitoring, especially of small calculi adjacent to indwelling ureteral stents. CONCLUSION: Both urinary calculi and ureteral stents can be accurately differentiated by a distinct appearance on DECT. For the management of urolithiasis patients can be monitored more easily and accurately using DECT if the stent shows a different colour than the adjacent stone.

CT colonography: size reduction of submerged colorectal polyps due to electronic cleansing and CT-window settings.

OBJECTIVES: To assess whether electronic cleansing (EC) of tagged residue and different computed tomography (CT) windows influence the size of colorectal polyps in CT colonography (CTC). METHODS: A database of 894 colonoscopy-validated CTC datasets of a low-prevalence cohort was retrospectively reviewed to identify patients with polyps ≥6 mm that were entirely submerged in tagged residue. Ten radiologists independently measured the largest diameter of each polyp, two-dimensionally, before and after EC in colon, bone, and soft-tissue-windows, in randomised order. Differences in size and polyp count before and after EC were calculated for size categories ≥6 mm and ≥10 mm. Statistical testing involved 95% confidence interval, intraclass correlation and mixed-model ANOVA. RESULTS: Thirty-seven patients with 48 polyps were included. Mean polyp size before EC was 9.8 mm in colon, 9.9 mm in bone and 8.2 mm in soft-tissue windows. After EC, the mean polyp size decreased significantly to 9.4 mm in colon, 9.1 mm in bone and 7.1 mm in soft-tissue windows. Compared to unsubtracted colon windows, EC, performed in colon, bone and soft-tissue windows, led to a shift of 6 (12,5%), 10 (20.8%) and 25 (52.1%) polyps ≥6 mm into the next smaller size category, thus affecting patient risk stratification. CONCLUSIONS: EC and narrow CT windows significantly reduce the size of polyps submerged in tagged residue. Polyp measurements should be performed in unsubtracted colon windows. KEY POINTS: • EC significantly reduces the size of polyps submerged in tagged residue. • Abdominal CT-window settings significantly underestimate 2D sizes of submerged polyps. • Size reduction in EC is significantly greater in narrow than wide windows. • Underestimation of polyp size due to EC may lead to inadequate treatment. • Polyp measurements should be performed in unsubtracted images using a colon window.

Improved detection of a tumorous involvement of the mesorectal fascia and locoregional lymph nodes in locally advanced rectal cancer using DCE-MRI.

PURPOSE: The prediction of an infiltration of the mesorectal fascia (MRF) and malignant lymph nodes is essential for treatment planning and prognosis of patients with rectal cancer. The aim of this study was to assess the additional diagnostic value of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for the detection of a malignant involvement of the MRF and of mesorectal lymph nodes in patients with locally advanced rectal cancer. METHODS: In this prospective study, 22 patients with locally advanced rectal cancer were examined with 1.5-T MRI between September 2012 and April 2015. Histopathological assessment of tumor size, tumor infiltration to the MRF, and malignant involvement of locoregional lymph nodes served as standard of reference. Sensitivity and specificity of detecting MRF infiltration and malignant nodes (nodal cut-off size [NCO] ≥ 5 and ≥ 10 mm, respectively) was determined by conventional MRI (cMRI; precontrast and postcontrast T1-weighted, T2-weighted, and diffusion-weighted images) and by additional semi-quantitative DCE-MRI maps (cMRI+DCE-MRI). RESULTS: Compared to cMRI, additional semi-quantitative DCE-MRI maps significantly increased sensitivity (86 vs. 71% [NCO ≥ 5 mm]/29% [NCO ≥ 10 mm]) and specificity (90 vs. 70% [NCO ≥ 5 mm]) of detecting malignant lymph nodes (p < 0.05). Moreover, DCE-MRI significantly augmented specificity (91 vs. 82%) of discovering a MRF infiltration (p < 0.05), while there was no change in sensitivity (83%; p > 0.05). CONCLUSION: DCE-MRI considerably increases both sensitivity and specificity for the detection of small mesorectal lymph node metastases (≥ 5 mm but < 10 mm) and sufficiently improves specificity of a suspected MRF infiltration in patients with locally advanced rectal cancer.

Magnetic resonance colonography for the detection of colorectal neoplasia in asymptomatic adults.

BACKGROUND & AIMS: Colonoscopy is the preferred screening test for colorectal neoplasia; the fecal occult blood test (FOBT) detects neoplasias with low levels of sensitivity. Computed tomographic colonography detects neoplasias with high levels of sensitivity but involves exposure to radiation. We investigated whether magnetic resonance colonography (MRC) can be used to screen for colorectal adenomas and cancers. METHODS: We analyzed data from 286 asymptomatic adults (40-82 years old) who underwent 3 Tesla MRC and colonoscopic examinations on the same day. FOBT was performed before bowel preparation. Colonoscopists were initially blinded to the findings on MRC and unblinded after withdrawal from the respective segments. Sensitivities for adenoma and per-patient sensitivities and specificities were calculated based on the unblinded results of colonoscopy. RESULTS: We detected 133 adenomas and 2 cancers in 86 patients; 37 adenomas were ≥6 mm, and 20 adenomas were advanced. Sensitivities of MRC and colonoscopy for adenomas ≥6 mm were 78.4% (95% confidence interval [CI], 61.8-90.2) and 97.3% (95% CI, 85.8-99.9); for advanced adenomas these values were 75% (95% CI, 50.9-91.3) and 100% (95% CI, 83.2-100.0), respectively. MRC identified 87.1% (95% CI, 70.2-96.4), colonoscopy 96.8% (95% CI, 83.3-99.9), and FOBT 10.0% (95% CI, 2.1-26.5) of individuals with adenomas ≥6 mm and 83.8% (95% CI, 58.6-96.4), 100% (95% CI, 81.5-100.0), and 17.6% (95% CI, 3.8-43.4) of individuals with advanced neoplasia. Specificities of MRC, colonoscopy, and FOBT for individuals with adenomas ≥6 mm were 95.3% (95% CI, 91.9-97.5), 96.9% (95% CI, 93.9-98.6), and 91.8% (95% CI, 87.6-94.9), respectively. CONCLUSIONS: 3 Tesla MRC detects colorectal adenomas ≥6 mm and advanced neoplasia with high levels of sensitivity and specificity. Although MRC detects colorectal neoplasia with lower levels of sensitivity than colonoscopy, it strongly outperforms one-time FOBT.

Complete pathological response to neoadjuvant pemetrexed/cisplatin in combination with regional hyperthermia in a patient with sarcomatoid peritoneal mesothelioma.

Diffuse malignant peritoneal mesothelioma (DMPM) is a rare disease. Although most patients eligible for surgery undergo cytoreductive surgery in combination with hyperthermic intraperitoneal chemotherapy, the role of perioperative systemic chemotherapy still remains undefined. Here we report the case of a 52-year-old female patient with advanced sarcomatoid DMPM. After five cycles of systemic pemetrexed and cisplatin, along with two cycles of regional hyperthermia, tumor resection with histomorphological examination showed a complete pathological response. We therefore conclude that there is a subgroup of DMPM patients that might benefit from systemic neoadjuvant chemotherapy with pemetrexed and cisplatin.

The role of functional imaging in the era of targeted therapy of renal cell carcinoma.

Antiangiogenic therapies interacting with tumor-specific pathways have been established for targeted therapy of renal cell carcinoma (RCC). However, evaluation of tumor response based on morphologic tumor diameter measurements has limitations, as tumor shrinkage may lag behind pathophysiological response. Functional imaging techniques such as dynamic contrast-enhanced (DCE) ultrasound (US), computed tomography (CT) and magnetic resonance imaging (MRI), unenhanced diffusion-weighted MRI (DW-MRI), and also metabolic imaging with positron emission tomography (PET) have the ability to assess physiological parameters and to predict and monitor therapy response. Assessment of changes in vascularity, cellularity, oxygenation, and glucose uptake with functional imaging during targeted therapy may correlate with progression-free survival and can predict tumor response or progression. In this review, we explore the potential of functional imaging techniques for assessing the effects of targeted therapy of RCC and as well review the reproducibility and limitations.

Evaluating the agreement between tumour volumetry and the estimated volumes of tumour lesions using an algorithm.

OBJECTIVES: To evaluate the agreement between tumour volume derived from semiautomated volumetry (SaV) and tumor volume defined by spherical volume using longest lesion diameter (LD) according to Response Evaluation Criteria In Solid Tumors (RECIST) or ellipsoid volume using LD and longest orthogonal diameter (LOD) according to World Health Organization (WHO) criteria. MATERIALS AND METHODS: Twenty patients with metastatic colorectal cancer from the CIOX trial were included. A total of 151 target lesions were defined by baseline computed tomography and followed until disease progression. All assessments were performed by a single reader. A variance component model was used to compare the three volume versions. RESULTS: There was a significant difference between the SaV and RECIST-based tumour volumes. The same model showed no significant difference between the SaV and WHO-based volumes. Scatter plots showed that the RECIST-based volumes overestimate lesion volume. The agreement between the SaV and WHO-based relative changes in tumour volume, evaluated by intraclass correlation, showed nearly perfect agreement. CONCLUSIONS: Estimating the volume of metastatic lesions using both the LD and LOD (WHO) is more accurate than those based on LD only (RECIST), which overestimates lesion volume. The good agreement between the SaV and WHO-based relative changes in tumour volume enables a reasonable approximation of three-dimensional tumour burden. KEY POINTS: • Tumour response in patients undergoing chemotherapy is assessed using CT images • Measurements are based on RECIST (unidimensional)-based or WHO (bidimensional)-based criteria • We calculated tumour volume from bidimensional target lesion measurements • This formula provides good tumour volume approximation, based on semiautomated volumetry.

Risk factors for advanced neoplasia within subcentimetric polyps: implications for diagnostic imaging.

OBJECTIVE: Diagnostic imaging by CT colonography and capsule endoscopy is used to detect colonic lesions. Controversy exists regarding the work-up of subcentimetric lesions. The aim of this study was to identify risk indicators for advanced neoplasia (AN) in subcentimetric polyps. DESIGN: Colonoscopies were classified on the basis of the largest lesion found. AN was defined as high-grade dysplasia, villous histology, or cancer. Logistic regression models were developed to identify risk factors for AN, and validated on separate datasets. A risk index based on the logistic regression was generated, and the number needed to screen (NNS) to detect AN was determined. RESULTS: 1,077,956 colonoscopies identified 106,270 intermediate (5-9 mm) and 198,954 diminutive (≤ 4 mm) lesions; 13% of intermediate and 3.7% of diminutive lesions contained AN. The risk of AN was higher in intermediate than in diminutive lesions (OR 3.1; 95% CI 3.0 to 3.3). Age ≥ 85 versus <45 years was associated with ORs of 2.4 (95% CI 1.8 to 3.1) for intermediate polyps and 3.2 (95% CI 2.3 to 4.5) for diminutive polyps. Pedunculated versus sessile morphology was associated with a higher risk of AN in intermediate (OR 2.0; 95% CI 1.9 to 2.2) and diminutive (OR 3.5; 95% CI 2.9 to 4.1) lesions. In the combined analysis for subcentimetric lesions, ORs were 2.7 (95% CI 2.2 to 3.3) for age ≥ 85 versus <45 years, 1.1 (95% CI 1.1 to 1.2) for male sex, 1.6 (95% CI 1.4 to 1.7) for occult blood, 1.3 (95% CI 1.2 to 1.5) for overt blood in stool, 1.3 (95% CI 1.2 to 1.4) for more than four lesions, and 2.2 (95% CI 2.1 to 2.3) for pedunculated versus sessile lesions. At median risk index values, the NNS was 9.3 (95% CI 9.1 to 9.5) in individuals with intermediate lesions and 29.4 (95% CI 28.5 to 30.2) in those with diminutive lesions. Compared with the NNS of 15 of the whole cohort, the majority of intermediate, but a minority of diminutive, lesions were deemed at high risk of AN. CONCLUSION: This study successfully identified risk factors and established a risk index for subcentimetric lesions. This has implications for the work-up of patients with subcentimetric lesions identified on diagnostic imaging.

Staging of muscle-invasive bladder cancer: can computerized tomography help us to decide on local treatment?

OBJECTIVES: To assess the power of multi-detector row computerized tomography (MDCT) in daily routine as a basic staging procedure for the decision on local treatment of patients with bladder cancer. PATIENTS AND METHODS: We retrospectively analysed 276 patients who had undergone radical cystectomy between 2004 and 2008 and correlated the MDCT findings with pathological findings, number of removed lymph nodes and type of urinary diversion. RESULTS: Accuracy of MDCT in predicting pathological tumour stage was 49% (kappa coefficient, 0.23; P < 0.001). Overstaging occurred in 23.4%, and understaging occurred in 24.7%. Accuracy in predicting lymph node metastases was 54% (kappa coefficient, 0.04; P = 0.297). Overstaging and understaging occurred in 8.3 and 29.4%, respectively. Significantly more ileal conduits were performed in patients with high postoperative pathological tumour stages (P = 0.04) and positive lymph nodes (P = 0.013). In contrast, there was no correlation between preoperative CT tumour/nodal stage and the number of removed lymph nodes (P = 0.44 and P = 0.732, respectively), and between preoperative tumour stage and type of urinary diversion (P = 0.126). CONCLUSIONS: MDCT as a preoperative staging procedure has a low accuracy in predicting the correct tumour and nodal stage, and therefore, it has little impact on decision-making for local treatment of muscle-invasive bladder cancer during radical cystectomy.

Dual energy CT in patients with polycystic kidney disease.

OBJECTIVES: To evaluate the diagnostic efficacy of dual source-dual energy CT (DECT) in the detection of neoplasia in patients with polycystic kidney disease (PKD). METHODS: A total of 21 patients with PKD underwent DECT on a dual source system, using kVp settings of Sn140/100 or 140/80. Colour-coded iodine maps and virtual unenhanced images were used to determine enhancement within cysts and to differentiate haemorrhagic from simple cysts. A cut-off of 15 HU was used as a threshold for malignancy. In patients with malignancy, histopathology was the gold standard; otherwise, patients underwent follow-up imaging for 150-908 days. RESULTS: On the basis of measured enhancement, 13 enhancing masses were seen in 4 patients (12 renal cell cancers and 1 adenoma); follow-up imaging showed no malignancy in 18 patients. Cysts did not enhance by more than 15 HU, whereas masses showed a mean enhancement of 45 (25-123) HU. Average radiation exposure was 9.6 mSv for the biphasic protocol and 5.8 mSv for DECT only. CONCLUSION: DECT greatly facilitates the detection of malignancy in patients with polycystic kidney disease, at the same time reducing radiation exposure by omission of a true unenhanced phase. KEY POINTS : • Identification of tumours within polycystic kidneys can be difficult. • Dual energy computed tomography (DECT) provides two separate sets of images. • Iodine maps and virtual non-enhanced (VNE) images can then be calculated. • DECT facilitates screening for potential renal tumours in polycystic kidneys.

Time-efficient CT colonography interpretation using an advanced image-gallery-based, computer-aided "first-reader" workflow for the detection of colorectal adenomas.

OBJECTIVES: To assess the performance of an advanced "first-reader" workflow for computer-aided detection (CAD) of colorectal adenomas ≥ 6 mm at computed tomographic colonography (CTC) in a low-prevalence cohort. METHODS: A total of 616 colonoscopy-validated CTC patient-datasets were retrospectively reviewed by a radiologist using a "first-reader" CAD workflow. CAD detections were presented as galleries of six automatically generated two-dimensional (2D) and three-dimensional (3D) images together with interactive 3D target views and 2D multiplanar views of the complete dataset. Each patient-dataset was interpreted by initially using CAD image-galleries followed by a fast 2D review to address unprompted colonic areas. Per-patient, per-polyp, and per-adenoma sensitivities were calculated for lesions ≥ 6 mm. Statistical testing employed Fisher's exact and McNemar tests. RESULTS: In 91/616 patients, 131 polyps (92 adenomas, 39 non-adenomas) ≥ 6 mm and two cancers were identified by reference standard. Using the CAD gallery-based first-reader workflow, the radiologist detected all adenomas ≥ 10 mm (34/34) and cancers. Per-patient and polyp sensitivities for lesions ≥ 6 mm were 84.3 % (75/89), and 83.2 % (109/131), respectively, with 89.1 % (57/64) and 85.9 % (79/92) for adenomas. Overall specificity was 95.6 % (504/527). Mean interpretation time was 3.1 min per patient. CONCLUSIONS: A CAD algorithm, applied in an image-gallery-based first-reader workflow, can substantially decrease reading times while enabling accurate detection of colorectal adenomas in a low-prevalence population. KEY POINTS: Computer-aided detection (CAD) is increasingly used to help interpret CT colonography (CTC). An image-gallery first-reader CAD-workflow is feasible for detection of colorectal adenomas ≥ 6 mm. Image-gallery first-reader CAD yields per-patient sensitivity of 89.1 % and specificity of 95.6 %. The mean reading time for CTC was 3.1 min, making screening feasible. No large adenoma was missed by the radiologist who reviewed with CAD galleries.

Dynamic contrast-enhanced CT-derived blood flow measurements enable early prediction of long term outcome in metastatic renal cell cancer patients on antiangiogenic treatment.

PURPOSE: To evaluate the role of dynamic contrast-enhanced CT (DCE-CT) as an independent non-invasive biomarker in predicting long term outcome in patients with metastatic renal cell carcinoma (mRCC) on antiangiogenic treatment. MATERIAL AND METHODS: Eighty two mRCC patients were prospectively enrolled from 09/2011 to 04/2015, out of which 71 were included in the final data analysis; the population was observed until 12/2020 to obtain complete overall survival data. DCE-CT imaging was performed at baseline and 10 to 12 weeks after start of treatment with targeted therapy. DCE-CT included a dynamic acquisition after injection of 50 ml of nonionic contrast agent at 6 ml/s using a 4D spiral mode (10 cm z-axis coverage, acquisition time 43 sec, 100 kVp (abdomen), 80 kVp (chest), 80-100 mAs) on a dual source scanner (Definition FLASH, Siemens). Blood flow (BF) was calculated for target tumor volumes using a deconvolution model. Progression free survival (PFS) and overall survival (OS) were analyzed using Kaplan-Meier statistics (SPSS version 24). RESULTS: Patients were treated with either sunitinib, pazopanib, sorafenib, tivozanib, axitinib, or cabozantinib. A cut-off value of 50% blood flow reduction at follow-up allowed for identification of patients with favorable long-term outcome: Median OS in n = 42 patients with an average blood flow reduction of >50% (mean, 79%) was 34 (range, 14-54) months, while n = 21 patients with an average reduction of less than 50% (mean, 28%) showed a median OS of 12 (range, 6-18) months, and n = 8 patients with an increase in blood flow survived for a median of 7 (range, 3-11) months. CONCLUSION: Blood flow in metastases measured with DCE-CT at first follow-up is a strong predictor of overall survival in mRCC patients on antiangiogenic treatment.

Comparison of diagnostic accuracy and interpretation times for a standard and an advanced 3D visualisation technique in CT colonography.

OBJECTIVE: To compare the diagnostic accuracy of a standard bi-directional, three-dimensional (3D) CT colonography (CTC) fly-through (standard view, SV) with a unidirectional, 3D unfolding technique (panoramic view, PV). METHODS: 150 consecutive endoscopically-validated CTC patient datasets were retrospectively reviewed twice by two expert radiologists: first, with bidirectional SV, second, after 6-15 months, with unidirectional PV. Per-polyp sensitivities, percentage of visualised colonic mucosa, and reading times were calculated for both 3D visualisations. Results were tested for statistical significance by equivalence analysis for paired proportions and Student's paired t-test. RESULTS: In 81 patients, 236 polyps (101 adenomas, 135 non-adenomas) were detected. Sensitivities for polyps ≤5 mm, 6-9 mm and ≥10 mm were 60.1% (113/188), 92.9% (26/28) and 95.0% (19/20) with bidirectional SV, and 60.6% (114/188), 96.4% (27/28) and 95.0% (19/20) with unidirectional PV. Overall sensitivity for adenomas was 86.1% and 84.2% for SV and PV. Both methods provided equivalent polyp detection, with an equivalence limit set at 5%. PV and SV visualised 98.9 ± 1.1% (97.0-99.9%) and 96.2 ± 2.3% (91.4-98.8%) of the colonic mucosa (p > 0.05). Mean interpretation time decreased from 14.6 ± 2.5 (9.2-22.8) minutes with SV to 7.5 ± 3.2 (5.0-14.4) using PV (p < 0.0001). CONCLUSION: 3D CTC interpretation using unidirectional PV is equally as accurate, but significantly faster than an interpretation based on bidirectional SV.

Diagnostic value of CT-colonography as compared to colonoscopy in an asymptomatic screening population: a meta-analysis.

OBJECTIVES: Previous meta-analyses on CT-colonography included both average and high risk individuals, which may overestimate the diagnostic value in screening. A meta-analysis was performed to obtain the value of CT-colonography for screening. METHODS: A search was performed using PubMed, Embase and Cochrane. Article selection and critical appraisal was done by two reviewers. INCLUSION CRITERIA: prospective, randomized trials or cohort studies comparing CT-colonography with colonoscopy (≥50 participants), ≥95% average risk participants ≥50 years. Study characteristics and 2 × 2 contingency Tables were recorded. Sensitivity and specificity estimates were calculated per patient and per polyp (≥6 mm, ≥10 mm), using univariate and bivariate analyses. RESULTS: Five of 1,021 studies identified were included, including 4,086 participants (<1% high risk). I(2)-values showed substantial heterogeneity, especially for 6-9 mm polyps and adenomas: 68.1% vs. 78.6% (sensitivity per patient). Estimated sensitivities for patients with polyps or adenomas ≥ 6 mm were 75.9% and 82.9%, corresponding specificities 94.6% and 91.4%. Estimated sensitivities for patients with polyps or adenomas ≥ 10 mm were 83.3% and 87.9%, corresponding specificities 98.7% and 97.6%. Estimated sensitivities per polyp for advanced adenomas ≥ 6 mm and ≥ 10 mm were 83.9% and 83.8%. CONCLUSION: Compared to colonoscopy, CT-colonography has a high sensitivity for adenomas ≥ 10 mm. For (advanced) adenomas ≥ 6 mm sensitivity is somewhat lower.

Image quality of virtual noncontrast images derived from dual-energy CT angiography after endovascular aneurysm repair.

PURPOSE: To compare true and "virtual" noncontrast images derived from dual-energy CT examinations in patients after endovascular repair of aortic aneurysms. MATERIALS AND METHODS: Seventy dual-energy CT examinations were performed on a dual-source CT scanner with a single-energy noncontrast scan and a dual-energy acquisition in venous phase. True and virtual noncontrast images were compared regarding image quality, calcifications in true noncontrast images, subtraction of calcification in virtual noncontrast images, and acceptance levels by two radiologists. Presence of endoleaks was assessed on venous-phase images and on virtual or true noncontrast images. In addition, the acceptance of color-coded images, in which iodine information is colored, was assessed. Possible dose reduction of a single-phase dual-energy examination protocol was compared with a standard biphasic examination protocol. RESULTS: Twenty-four endoleaks were detected and correctly classified with both approaches. Mean image quality was rated good for virtual noncontrast images (1.97 +/- 0.99) and excellent for true noncontrast images (1.16 +/- 0.37; P< .0001). Ninety-four percent of virtual noncontrast images were rated as diagnostic, and 80% of all true noncontrast images showed calcifications within the aneurysm. Subtraction of calcification in virtual noncontrast images was classified as none (30%), minimal (40%), moderate (24%), or severe (6%). Eighty-three percent of color-coded images were rated as fully diagnostic, 11% were accepted with restrictions, and 6% were nondiagnostic. Possible dose reduction of a single-phase dual-energy protocol, compared with a standard biphasic protocol, was 44%. CONCLUSIONS: Dual-energy CT makes a reliable detection of endoleaks feasible in a single acquisition. This provides a potential dose reduction for patients who have to undergo lifelong follow-up examinations after endovascular aneurysm repair.

Dual-source dual-energy MDCT of pancreatic adenocarcinoma: initial observations with data generated at 80 kVp and at simulated weighted-average 120 kVp.

OBJECTIVE: The purpose of this study was to determine whether the conspicuity of malignant tumors of the pancreas at dual-source dual-energy CT is better with 80-kVp acquisition than with 120-kVp acquisition simulated with a weighted average. MATERIALS AND METHODS: Fifteen patients with pancreatic adenocarcinoma underwent contrast-enhanced dual-source dual-energy CT. The abdominal diameter of all patients was 35 cm or less. Data were reconstructed as a weighted average of the 140- and 80-kVp acquisitions, simulating 120 kVp, and as a pure 80-kVp data set. A region-of-interest cursor was placed within the tumor and the adjacent normal parenchyma, and attenuation differences and contrast-to-noise ratios were calculated for pancreatic tumors at 80 kVp and with the weighted-average acquisition. The 80-kVp and weighted-average images were subjectively compared in terms of lesion conspicuity, image quality, and duct visualization. An exact Wilcoxon's matched pairs signed rank test was used to test whether differences in attenuation, contrast-to-noise ratio, and subjective assessment were greater at 80 kVp. RESULTS: The mean difference in attenuation for each pancreatic tumor and adjacent portion of normal pancreas was 83.27+/-29.56 (SD) HU at 80 kVp and 49.40+/-23.00 HU at weighted-average 120 kVp. Adenocarcinoma attenuation differences were significantly greater at 80 kVp than at 120 kVp (p=0.00006). Contrast-to-noise ratio was significantly higher at 80 kVp than at 120 kVp (p=0.00147). Subjective analysis showed lesion conspicuity (p=0.001) and duct visualization (p=0.0156) were significantly better on the 80-kVp images. CONCLUSION: At portal venous phase dual-source dual-energy CT, the conspicuity of malignant tumors of the pancreas is greater at 80 kVp than with weighted-average acquisition.