Patient-tailored risk assessment of obstructive coronary artery disease using Rubidium-82 PET-based myocardial flow quantification with visual interpretation.

INTRODUCTION: Our aim was to estimate the probability of obstructive CAD (oCAD) for an individual patient as a function of the myocardial flow reserve (MFR) measured with Rubidium-82 (Rb-82) PET in patients with a visually normal or abnormal scan. MATERIALS AND METHODS: We included 1519 consecutive patients without a prior history of CAD referred for rest-stress Rb-82 PET/CT. All images were visually assessed by two experts and classified as normal or abnormal. We estimated the probability of oCAD for visually normal scans and scans with small (5%-10%) or larger defects (> 10%) as function of MFR. The primary endpoint was oCAD on invasive coronary angiography, when available. RESULTS: 1259 scans were classified as normal, 136 with a small defect and 136 with a larger defect. For the normal scans, the probability of oCAD increased exponentially from 1% to 10% when segmental MFR decreased from 2.1 to 1.3. For scans with small defects, the probability increased from 13% to 40% and for larger defects from 45% to > 70% when segmental MFR decreased from 2.1 to 0.7. CONCLUSION: Patients with > 10% risk of oCAD can be distinguished from patients with < 10% risk based on visual PET interpretation only. However, there is a strong dependence of MFR on patient's individual risk of oCAD. Hence, combining both visual interpretation and MFR results in a better individual risk assessment which may impact treatment strategy.

Quantitative Volumetric Assessment of Ablative Margins in Hepatocellular Carcinoma: Predicting Local Tumor Progression Using Nonrigid Registration Software.

PURPOSE: After radiofrequency ablation (RFA) of hepatocellular carcinoma (HCC), pre- and postinterventional contrast-enhanced CT (CECT) images are usually qualitatively interpreted to determine technical success, by eyeballing. The objective of this study was to evaluate the feasibility of quantitative assessment, using a nonrigid CT-CT coregistration algorithm. MATERIALS AND METHODS: 25 patients treated with RFA for HCC between 2009 and 2014 were retrospectively included. Semiautomated coregistration of pre- and posttreatment CECT was performed independently by two radiologists. In scans with a reliable registration, the tumor and ablation area were delineated to identify the side and size of narrowest RFA margin. In addition, qualitative assessment was performed independently by two other radiologists to determine technical success and the anatomical side and size of narrowest margin. Interobserver agreement rates were determined for both methods, and the outcomes were compared with occurrence of local tumor progression (LTP). RESULTS: CT-CT coregistration was technically feasible in 18/25 patients with almost perfect interobserver agreement for quantitative analysis (κ = 0.88). The interobserver agreement for qualitative RFA margin analysis was κ = 0.64. Using quantitative assessment, negative ablative margins were found in 12/18 patients, with LTP occurring in 8 of these patients. In the remaining 6 patients, quantitative analysis demonstrated complete tumor ablation and no LTP occurred. CONCLUSION: Feasibility of quantitative RFA margin assessment using nonrigid coregistration of pre- and postablation CT is limited, but appears to be a valuable tool in predicting LTP in HCC patients (p=0.013).

Deep learning enables automatic quantitative assessment of puborectalis muscle and urogenital hiatus in plane of minimal hiatal dimensions.

OBJECTIVES: To measure the length, width and area of the urogenital hiatus (UH), and the length and mean echogenicity (MEP) of the puborectalis muscle (PRM), automatically and observer-independently, in the plane of minimal hiatal dimensions on transperineal ultrasound (TPUS) images, by automatic segmentation of the UH and the PRM using deep learning. METHODS: In 1318 three- and four-dimensional (3D/4D) TPUS volume datasets from 253 nulliparae at 12 and 36 weeks' gestation, two-dimensional (2D) images in the plane of minimal hiatal dimensions with the PRM at rest, on maximum contraction and on maximum Valsalva maneuver, were obtained manually and the UH and PRM were segmented manually. In total, 713 of the images were used to train a convolutional neural network (CNN) to segment automatically the UH and PRM in the plane of minimal hiatal dimensions. In the remainder of the dataset (test set 1 (TS1); 601 images, four having been excluded), the performance of the CNN was evaluated by comparing automatic and manual segmentations. The performance of the CNN was also tested on 117 images from an independent dataset (test set 2 (TS2); two images having been excluded) from 40 nulliparae at 12 weeks' gestation, which were acquired and segmented manually by a different observer. The success of automatic segmentation was assessed visually. Based on the CNN segmentations, the following clinically relevant parameters were measured: the length, width and area of the UH, the length of the PRM and MEP. The overlap (Dice similarity index (DSI)) and surface distance (mean absolute distance (MAD) and Hausdorff distance (HDD)) between manual and CNN segmentations were measured to investigate their similarity. For the measured clinically relevant parameters, the intraclass correlation coefficients (ICCs) between manual and CNN results were determined. RESULTS: Fully automatic CNN segmentation was successful in 99.0% and 93.2% of images in TS1 and TS2, respectively. DSI, MAD and HDD showed good overlap and distance between manual and CNN segmentations in both test sets. This was reflected in the respective ICC values in TS1 and TS2 for the length (0.96 and 0.95), width (0.77 and 0.87) and area (0.96 and 0.91) of the UH, the length of the PRM (0.87 and 0.73) and MEP (0.95 and 0.97), which showed good to very good agreement. CONCLUSION: Deep learning can be used to segment automatically and reliably the PRM and UH on 2D ultrasound images of the nulliparous pelvic floor in the plane of minimal hiatal dimensions. These segmentations can be used to measure reliably UH dimensions as well as PRM length and MEP. © 2018 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.

Low-dose CT angiography of the abdominal aorta and reduced contrast medium volume: Assessment of image quality and radiation dose.

AIM: To determine the effect of using 80 kV tube voltage and a reduced amount of contrast medium on the image quality and radiation dose of computed tomography angiography (CTA) of the abdominal aorta. MATERIALS AND METHODS: Patients who were referred for a CTA examination of the abdominal aorta were included in this technical efficacy study. Thirty patients were divided randomly into two groups. Fifteen patients underwent a dual-energy CT (DECT) protocol (Group A). Fifteen patients were scanned with the use of an automated tube potential selection algorithm tool (Group B). In both protocols, a test bolus injection of 10 ml ioversol (350 mg iodine/ml) was used, followed by 20 ml of 1:1 saline-diluted contrast medium. Quantitative analysis comprised determination of the mean attenuation and contrast-to-noise ratio. Qualitative image analysis was performed independently by five radiologists. The estimated radiation dose in terms of CT dose index and effective dose was recorded and compared with a standard 120 kV protocol. RESULTS: In Group B, six patients underwent CTA at 80 kV, seven patients underwent CTA at 100 kV and two patients underwent CTA at 120 kV. The mean contrast-enhancement values of Group A (80 kV) and the 80 kV subgroup of Group B were 16.5% and 27.6% higher compared to the 100 kV subgroup of Group B, these differences were, however, not significant. There were no significant differences in mean image quality between groups. In patients undergoing CTA at 80 kV the effective dose decreased by up to 51.3% compared to a conventional 120 kV CTA protocol. CONCLUSIONS: The findings of this study support the hypothesis that 80 kV in CTA of the abdominal aorta can reliably be used with only 30 ml contrast medium in total and a 50% reduction in radiation dose. The overall image quality was diagnostically adequate; however, it appeared to be suboptimal in patients with a BMI above 28 kg/m(2).

Three-dimensional reconstruction of stenosed coronary artery segments with assessment of the flow impedance.

In this paper preliminary results of a study about the diagnostic benefits of 3D visualization and quantitation of stenosed coronary artery segments are presented. As is well known, even biplane angiographic images do not provide enough information for binary reconstruction. Therefore, a priori information about the slice to be reconstructed must be incorporated into the reconstruction algorithm. One approach is to assume a circular cross-section of the coronary artery. Hence, the diameter is estimated from the contours of the vessels in both projections. Another approach is to search for a solution of the reconstruction problem close to the previously reconstructed adjacent slice. In this paper we follow the first method based on contour information. The reconstructed coronary segment is visualized in three dimensions. Based on the obtained geometry of the obstruction the pertinent blood flow impedance is estimated on the basis of fluid dynamic principles. The results of applying the reconstruction algorithms to clinical coronary biplane exposure are presented with an indication of the assessed flow impedance.