Dataset Information

Automated Generation of Reliable Blood Velocity Parameter Maps from Contrast-Enhanced Ultrasound Data.

ABSTRACT: The purpose of this study was the automated generation and validation of parametric blood flow velocity maps, based on contrast-enhanced ultrasound (CEUS) scans.Ethical approval for animal experiments was obtained. CEUS destruction-replenishment sequences were recorded in phantoms and three different tumor xenograft mouse models. Systematic pixel binning and intensity averaging was performed to generate parameter maps of blood flow velocities with different pixel resolution. The 95% confidence interval of the mean velocity, calculated on the basis of the whole tumor segmentation, served as ground truth for the different parameter maps.In flow phantoms the measured mean velocity values were only weakly influenced by the pixel resolution and correlated with real velocities (r2 ? 0.94, p < 0.01). In tumor xenografts, however, calculated mean velocities varied significantly (p < 0.0001), depending on the parameter maps' resolution. Pixel binning was required for all in vivo measurements to obtain reliable parameter maps and its degree depended on the tumor model.Systematic pixel binning allows the automated identification of optimal pixel resolutions for parametric maps, supporting textural analysis of CEUS data. This approach is independent from the ultrasound setup and can be implemented in the software of other (clinical) ultrasound devices.

SUBMITTER: Theek B

PROVIDER: S-EPMC5612675 | biostudies-other | 2017

REPOSITORIES: biostudies-other

ACCESS DATA

Publications

Automated Generation of Reliable Blood Velocity Parameter Maps from Contrast-Enhanced Ultrasound Data.

Theek Benjamin B Opacic Tatjana T Möckel Diana D Schmitz Georg G Lammers Twan T Kiessling Fabian F

Contrast media & molecular imaging 20170530

<h4>Objectives</h4>The purpose of this study was the automated generation and validation of parametric blood flow velocity maps, based on contrast-enhanced ultrasound (CEUS) scans.<h4>Materials and methods</h4>Ethical approval for animal experiments was obtained. CEUS destruction-replenishment sequences were recorded in phantoms and three different tumor xenograft mouse models. Systematic pixel binning and intensity averaging was performed to generate parameter maps of blood flow velocities with ...[more]

PMID: 29097912

Similar Datasets

Project description:BackgroundThe aim of modern medicine is to safely classify diseases for successful therapy without invasive measures. Sonography, computed tomography (CT), and magnetic resonance imaging (MRI) are potent imaging techniques. However, without contrast medium, the informative value of the 3 native methods is limited. The advantages of sonography are: no radiation exposure or previously known physically harmful interactions with tissue, proportionate disappearance of a contrast agent risk, no (probably irreversible) contrast agent deposits, and no risk of renal insufficiency. But, is that enough to compete with of even exceed CT and MRI?SummaryIn this review, the state of the art of contrast-enhanced ultrasound (CEUS) in the abdominal cavity is presented. The remarkable diagnostic possibilities can unfortunately only be demonstrated here in a small number of impressive, typical case studies underpinned by the literature, so that, from one's own perspective, the full spectrum of CEUS can be used by oneself or initiated. Within the limits of physics, the real-time dynamics of CEUS enable conclusions to be drawn, so that with the current technology, sonography, including expansion by contrast, can be considered superior to other imaging methods. It is not uncommon for CEUS to have the value of a control and reference method.Key messagesSonography very often enables reliable diagnostics. The introduction of a contrast agent in sonography has led to a quantum leap similar to that of other imaging techniques. Already natively, the real-time representation of dynamic events leads to a certain superiority, i.e., complete observation of the inflow and outflow phases of the contrast medium and the resulting diagnostic; tissue-specific differentiation options provide a unique selling point. Further advantages of the first-choice imaging diagnostic method are: a lack of radiation exposure, repeatability of the examination at any time, local independence, a negligible allergy rate compared to the contrast agents of other methods, and a lack of kidney and thyroid exposure or excluded deposits.

Project description:New findingsWhat is the central question of this study? Contrast-enhanced ultrasound (CEUS) can be used to directly assess skeletal muscle perfusion but its day-to-day repeatability over time has not yet been validated: is CEUS a repeatable method for the measurement of skeletal muscle microvascular blood flow (MBF) at rest and in response to exercise, across independent assessment sessions? What is the main finding and its importance? A strong agreement between CEUS MBF measures across sessions suggests it is a repeatable method for assessing skeletal muscle perfusion over time. This validation provides confidence for incorporating these measures into longitudinal studies such as a chronic intervention or disease progression to gain further knowledge of skeletal muscle microvascular function.AbstractContrast-enhanced ultrasound (CEUS) can be used to directly assess skeletal muscle perfusion. However, its repeatability over time has not yet been validated and therefore its use in longitudinal measures (i.e., exploring the impact of a chronic intervention or disease progression) is limited. This study aimed to determine the repeatability of CEUS for the measurement of skeletal muscle microvascular blood flow (MBF) at baseline and in response to exercise, across independent assessment sessions. Ten healthy volunteers (five female; 30 ± 6 years) had CEUS of the right vastus lateralis recorded in two separate sessions, 14 days apart. Measurements were taken at baseline, during an isometric leg extension and during recovery. Acoustic intensity data from a region of interest were plotted as a replenishment curve to obtain blood volume (A) and flow velocity (β) values from a one-phase association non-linear regression of mean tissue echogenicity. Linear regression and Bland-Altman analyses of A and β values were performed, with significance assumed as P < 0.05. Strong positive correlations were observed across sessions for all A and β values (both P < 0.0001). Bland-Altman analysis showed a bias (SD) of -0.013 ± 1.24 for A and -0.014 ± 0.31 for β. A bias of 0.201 ± 0.770 at baseline, 0.527 ± 1.29 during contraction and -0.203 ± 1.29 at recovery was observed for A, and -0.0328 ± 0.0853 (baseline), -0.0446 ± 0.206 (contraction) and 0.0382 ± 0.233 (recovery) for β. A strong agreement between CEUS MBF measures across independent sessions suggests it to be a repeatable method for assessing skeletal muscle perfusion over time, and therefore facilitates wider use in longitudinal studies.

Project description:ObjectivesThe aim of this study was to assess the potential of machine learning based on B-mode, shear-wave elastography (SWE), and dynamic contrast-enhanced ultrasound (DCE-US) radiomics for the localization of prostate cancer (PCa) lesions using transrectal ultrasound.MethodsThis study was approved by the institutional review board and comprised 50 men with biopsy-confirmed PCa that were referred for radical prostatectomy. Prior to surgery, patients received transrectal ultrasound (TRUS), SWE, and DCE-US for three imaging planes. The images were automatically segmented and registered. First, model-based features related to contrast perfusion and dispersion were extracted from the DCE-US videos. Subsequently, radiomics were retrieved from all modalities. Machine learning was applied through a random forest classification algorithm, using the co-registered histopathology from the radical prostatectomy specimens as a reference to draw benign and malignant regions of interest. To avoid overfitting, the performance of the multiparametric classifier was assessed through leave-one-patient-out cross-validation.ResultsThe multiparametric classifier reached a region-wise area under the receiver operating characteristics curve (ROC-AUC) of 0.75 and 0.90 for PCa and Gleason > 3 + 4 significant PCa, respectively, thereby outperforming the best-performing single parameter (i.e., contrast velocity) yielding ROC-AUCs of 0.69 and 0.76, respectively. Machine learning revealed that combinations between perfusion-, dispersion-, and elasticity-related features were favored.ConclusionsIn this paper, technical feasibility of multiparametric machine learning to improve upon single US modalities for the localization of PCa has been demonstrated. Extended datasets for training and testing may establish the clinical value of automatic multiparametric US classification in the early diagnosis of PCa.Key points• Combination of B-mode ultrasound, shear-wave elastography, and contrast ultrasound radiomics through machine learning is technically feasible. • Multiparametric ultrasound demonstrated a higher prostate cancer localization ability than single ultrasound modalities. • Computer-aided multiparametric ultrasound could help clinicians in biopsy targeting.

Dataset Information

Automated Generation of Reliable Blood Velocity Parameter Maps from Contrast-Enhanced Ultrasound Data.

Publications

Automated Generation of Reliable Blood Velocity Parameter Maps from Contrast-Enhanced Ultrasound Data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets