Search

Objectives: We compared the interobserver agreement for the recently introduced contrast-enhanced ultrasound (CEUS)-based algorithm CEUS-LI-RADS (Liver Imaging Reporting and Data System) versus the well-established magnetic resonance imaging (MRI)-LI-RADS for non-invasive diagnosis of hepatocellular carcinoma (HCC) in high-risk patients., Methods: Focal liver lesions in 50 high-risk patients (mean age 66.2 ± 11.8 years; 39 male) were assessed retrospectively with CEUS and MRI. Two independent observers reviewed CEUS and MRI examinations, separately, classifying observations according to CEUS-LI-RADSv.2016 and MRI-LI-RADSv.2014. Interobserver agreement was assessed with Cohen's kappa., Results: Forty-three lesions were HCCs; two were intrahepatic cholangiocarcinomas; five were benign lesions. Arterial phase hyperenhancement was perceived less frequently with CEUS than with MRI (37/50 / 38/50 lesions = 74%/78% [CEUS; observer 1/observer 2] versus 46/50 / 44/50 lesions = 92%/88% [MRI; observer 1/observer 2]). Washout appearance was observed in 34/50 / 20/50 lesions = 68%/40% with CEUS and 31/50 / 31/50 lesions = 62%/62%) with MRI. Interobserver agreement was moderate for arterial hyperenhancement (ĸ = 0.511/0.565 [CEUS/MRI]) and "washout" (ĸ = 0.490/0.582 [CEUS/MRI]), fair for CEUS-LI-RADS category (ĸ = 0.309) and substantial for MRI-LI-RADS category (ĸ = 0.609). Intermodality agreement was fair for arterial hyperenhancement (ĸ = 0.329), slight to fair for "washout" (ĸ = 0.202) and LI-RADS category (ĸ = 0.218) CONCLUSION: Interobserver agreement is substantial for MRI-LI-RADS and only fair for CEUS-LI-RADS. This is mostly because interobserver agreement in the perception of washout appearance is better in MRI than in CEUS. Further refinement of the LI-RADS algorithms and increasing education and practice may be necessary to improve the concordance between CEUS and MRI for the final LI-RADS categorization., Key Points: • CEUS-LI-RADS and MRI-LIRADS enable standardized non-invasive diagnosis of HCC in high-risk patients. • With CEUS, interobserver agreement is better for arterial hyperenhancement than for "washout". • Interobserver agreement for major features is moderate for both CEUS and MRI. • Interobserver agreement for LI-RADS category is substantial for MRI, and fair for CEUS. • Interobserver-agreement for CEUS-LI-RADS will presumably improve with ongoing use of the algorithm.

Being able to segment the esophagus without user interaction from 3-D CT data is of high value to radiologists during oncological examinations of the mediastinum. The segmentation can serve as a guideline and prevent confusion with pathological tissue. However, limited contrast to surrounding structures and versatile shape and appearance make segmentation a challenging problem. This paper presents a multistep method. First, a detector that is trained to learn a discriminative model of the appearance is combined with an explicit model of the distribution of respiratory and esophageal air. In the next step, prior shape knowledge is incorporated using a Markov chain model. We follow a "detect and connect" approach to obtain the maximum a posteriori estimate of the approximate esophagus shape from hypothesis about the esophagus contour in axial image slices. Finally, the surface of this approximation is nonrigidly deformed to better fit the boundary of the organ. The method is compared to an alternative approach that uses a particle filter instead of a Markov chain to infer the approximate esophagus shape, to the performance of a human observer and also to state of the art methods, which are all semiautomatic. Cross-validation on 144 CT scans showed that the Markov chain based approach clearly outperforms the particle filter. It segments the esophagus with a mean error of 1.80 mm in less than 16 s on a standard PC. This is only 1 mm above the interobserver variability and can compete with the results of previously published semiautomatic methods.

In this paper, a method is described to automatically estimate the visible body region of a computed tomography (CT) volume image. In order to quantify the body region, a body coordinate (BC) axis is used that runs in longitudinal direction. Its origin and unit length are patient-specific and depend on anatomical landmarks. The body region of a test volume is estimated by registering it only along the longitudinal axis to a set of reference CT volume images with known body coordinates. During these 1D registrations, an axial image slice of the test volume is compared to an axial slice of a reference volume by extracting a descriptor from both slices and measuring the similarity of the descriptors. A slice descriptor consists of histograms of visual words. Visual words are code words of a quantized feature space and can be thought of as classes of image patches with similar appearance. A slice descriptor is formed by sampling a slice on a regular 2D grid and extracting a Speeded Up Robust Features (SURF) descriptor at each sample point. The codebook, or visual vocabulary, is generated in a training step by clustering SURF descriptors. Each SURF descriptor extracted from a slice is classified into the closest visual word (or cluster center) and counted in a histogram. A slice is finally described by a spatial pyramid of such histograms. We introduce an extension of the SURF descriptors to an arbitrary number of dimensions (N-SURF). Here, we make use of 2-SURF and 3-SURF descriptors. Cross-validation on 84 datasets shows the robustness of the results. The body portion can be estimated with an average error of 15.5mm within 9s. Possible applications of this method are automatic labeling of medical image databases and initialization of subsequent image analysis algorithms., (Copyright © 2010 Elsevier Ltd. All rights reserved.)

Automatic segmentation of the esophagus from CT data is a challenging problem. Its wall consists of muscle tissue, which has low contrast in CT. Sometimes it is filled with air or remains of orally given contrast agent. While air holes are a clear hint to a human when searching for the esophagus, we found that they are rather distracting to discriminative models of the appearance because of their similarity to the trachea and to lung tissue. However, air inside the respiratory organs can be segmented easily. In this paper, we propose to combine a model based segmentation algorithm of the esophagus with a spatial probability map generated from detected air. Threefold cross-validation on 144 datasets showed that this probability map, combined with a technique that puts more focus on hard cases, increases accuracy by 22%. In contrast to prior work, our method is not only automatic on a manually selected region of interest, but on a whole thoracic CT scan, while our mean segmentation error of 1.80mm is even better.

Automated segmentation of the esophagus in CT images is of high value to radiologists for oncological examinations of the mediastinum. It can serve as a guideline and prevent confusion with pathological tissue. However, segmentation is a challenging problem due to low contrast and versatile appearance of the esophagus. In this paper, a two step method is proposed which first finds the approximate shape using a "detect and connect" approach. A classifier is trained to find short segments of the esophagus which are approximated by an elliptical model. Recently developed techniques in discriminative learning and pruning of the search space enable a rapid detection of possible esophagus segments. Prior shape knowledge of the complete esophagus is modeled using a Markov chain framework, which allows efficient inferrence of the approximate shape from the detected candidate segments. In a refinement step, the surface of the detected shape is non-rigidly deformed to better fit the organ boundaries. In contrast to previously proposed methods, no user interaction is required. It was evaluated on 117 datasets and achieves a mean segmentation error of 2.28mm with less than 9s computation time.

Volumetric analysis of the kidney parenchyma provides additional information for the detection and monitoring of various renal diseases. Therefore the purposes of the study were to develop and evaluate a semi-automated segmentation tool and a modified ellipsoid formula for volumetric analysis of the kidney in non-contrast T2-weighted magnetic resonance (MR)-images. Three readers performed semi-automated segmentation of the total kidney volume (TKV) in axial, non-contrast-enhanced T2-weighted MR-images of 24 healthy volunteers (48 kidneys) twice. A semi-automated threshold-based segmentation tool was developed to segment the kidney parenchyma. Furthermore, the three readers measured renal dimensions (length, width, depth) and applied different formulas to calculate the TKV. Manual segmentation served as a reference volume. Volumes of the different methods were compared and time required was recorded. There was no significant difference between the semi-automatically and manually segmented TKV ( p = 0.31). The difference in mean volumes was 0.3 ml (95% confidence interval (CI), −10.1 to 10.7 ml). Semi-automated segmentation was significantly faster than manual segmentation, with a mean difference = 188 s (220 vs. 408 s); p < 0.05. Volumes did not differ significantly comparing the results of different readers. Calculation of TKV with a modified ellipsoid formula (ellipsoid volume × 0.85) did not differ significantly from the reference volume; however, the mean error was three times higher (difference of mean volumes −0.1 ml; CI −31.1 to 30.9 ml; p = 0.95). Applying the modified ellipsoid formula was the fastest way to get an estimation of the renal volume (41 s). Semi-automated segmentation and volumetric analysis of the kidney in native T2-weighted MR data delivers accurate and reproducible results and was significantly faster than manual segmentation. Applying a modified ellipsoid formula quickly provides an accurate kidney volume. [ABSTRACT FROM AUTHOR]