Reidar Brekken

Tissue Motion and Strain in the Human Brain Assessed by Intraoperative Ultrasound in Glioma Patients

The objective of the study was to investigate tissue motion and strain imposed by cardiovascular pulsation in pathologic and normal brain parenchyma, as quantified from in vivo ultrasound data. Ultrasound acquired during surgery of 16 patients with glial tumors was retrospectively processed and analyzed. The tissue velocity was quantified at depths of 1cm, 2cm and 3cm from brain cortex to investigate spatial dependency with depth. Comparison of strain and velocity in tumor and adjacent normal parenchyma was performed by selecting two regions-of-interest in the hyperechoic tumor and two regions in the low-echogenic areas interpreted as mainly normal tissue with some degree of tumor cell infiltration. The absolute maximum tissue velocity is seen to increase with increasing depths in 14 of 16 cases (87.5%). The maximum tissue velocities in the four regions close to the ultrasound visible tumor border are not statistically different (p=0.163 to p=0.975). The strain magnitudes are significantly higher in the regions with expected normal brain parenchyma than in regions with expected glial tumor tissue, both for the two regions being closest to the tumor border (p=0.0004) and for the two regions further away from the tumor border (p=0.0009). We conclude that the velocity of the brain parenchyma imposed by arterial pulsation during a cardiac cycle is generally increasing with increasing depth from cortex. The maximum velocity appears to be similar in regions with expected normal brain and tumor tissue, thus, does not seem to be affected by pathology. Strain magnitude is, however, a suitable parameter for discrimination of glial tumor and normal brain parenchyma. (E-mail: Tormod.Selbekk@sintef.no).

Comparison of contrast in brightness mode and strain ultrasonography of glial brain tumours

BackgroundImage contrast between normal tissue and brain tumours may sometimes appear to be low in intraoperative ultrasound. Ultrasound imaging of strain is an image modality that has been recently explored for intraoperative imaging of the brain. This study aims to investigate differences in image contrast between ultrasound brightness mode (B-mode) images and ultrasound strain magnitude images of brain tumours.MethodsUltrasound radiofrequency (RF) data was acquired during surgery in 15 patients with glial tumours. The data were subsequently processed to provide strain magnitude images. The contrast in the B-mode images and the strain images was determined in assumed normal brain tissue and tumour tissue at selected regions of interest (ROI). Three measurements of contrast were done in the ultrasound data for each patient. The B-mode and strain contrasts measurements were compared using the paired samples t- test.ResultsThe statistical analysis of a total of 45 measurements shows that the contrasts in the strain magnitude images are significantly higher than in the conventional ultrasound B-mode images (P < 0.0001).ConclusionsThe results indicate that ultrasound strain imaging provides better discrimination between normal brain tissue and glial tumour tissue than conventional ultrasound B-mode imaging. Ultrasound imaging of tissue strain therefore holds the potential of becoming a valuable adjunct to conventional intraoperative ultrasound imaging in brain tumour surgery.

Simulation Model for Assessing Quality of Ultrasound Strain Estimation in Abdominal Aortic Aneurysm

The purpose of this study was to develop a simulation model for evaluating methods for ultrasound strain estimation in abdominal aortic aneurysms. Wall geometry was obtained from a real ultrasound image and wall motion was simulated applying realistic blood pressures to a nonlinear viscoelastic wall model. The ultrasound simulation included speckle, absorption and angle dependent reflection. Gaussian white noise was added to simulate various noise levels. Despite not fully replicating real ultrasound images, the model simulated realistic circumferential variations in intensity and realistic speckle patterns and has potential for initial evaluation of strain estimation methods.

Reduced strain in abdominal aortic aneurysms after endovascular repair.

Purpose: To compare in vivo strain in abdominal aortic aneurysms before and after endovascular aneurysm repair (EVAR), thereby obtaining a quantitative measure of changes in mechanical burden on the aneurysm wall. Method: Transabdominal ultrasound was acquired from 10 patients (9 men; median age 76 years, range 61–83) 1 day before and 2 days after elective EVAR. Strain was estimated as the relative cyclic elongation and contraction of the wall tissue in a number of connected segments along the aneurysm circumference. For each time instance of the cardiac cycle, the maximum and the average strain values along the circumference were recorded. The temporal maximums of these parameters (defined as the maximum strain and the peak average strain, respectively) were compared before and after EVAR. Results: Both maximum strain and peak average strain were reduced following EVAR by 41% (range 35%–63%) and 68% (range 41%–93%), respectively. Despite the reduction, cyclic strain was still evident after the stent-graft was placed, even when no evidence of endoleak was found. Further, the strain values were inhomogeneous along the circumference, both before and after treatment. In 2 cases, endoleak was proven by routine computed tomography; the relative reduction in maximum strain was slightly less in these cases (35% and 38%) compared to those without endoleak (45%, range 38%–63%). No difference was found in reduction of peak average strain. Conclusion: Strain is significantly reduced after EVAR, but there may still be a certain level of strain after the treatment. The strain values are inhomogeneous along the circumference both before and after treatment. These results encourage further investigation to evaluate the potential for using circumferential strain as an additional indicator of outcome after endovascular repair.

Real-time ultrasound simulation for low cost training simulators

Ultrasound imaging is used within numerous medical disciplines. Extensive and repeated training is needed for efficient use of the technology. Simulator training has been proposed as a complement to other training methods. Advantages of simulator training include access to a large number of normal and rare cases without the need for suitable volunteers and available ultrasound equipment. The imaging of soft tissue can be simulated by considering the interaction between the tissue and the ultrasound field. The objective of this study is to include these effects in real-time simulators. One previous approach has been to simulate a three-dimensional (3D) ultrasound volume off line, and then cross-section the volume in real time. This approach, however, does not take into account the anisotropic resolution of ultrasound imaging. If we assume that the average acoustical properties of tissues are slowly varying and that the speckle pattern is independent of the tissue, we show that ultrasound images can be simulated by multiplying a pre-simulated speckle image by an any-plane cross section of a 3D representation of an anatomy. Thus anisotropic resolution can be simulated in real time. The simulated images were compared to true ultrasound images of soft tissue. Since the speckle was simulated independently of the tissue, the most realistic results were obtained for still images, but the method was also satisfactory for moving images when speckle tracking between views was not important. The method is well applicable to ultrasound training simulators on low cost platforms.

Efficiency of ultrasound training simulators: method for assessing image realism.

Abstract Although ultrasound has become an important imaging modality within several medical professions, the benefit of ultrasound depends to some degree on the skills of the person operating the probe and interpreting the image. For some applications, the possibility to educate operators in a clinical setting is limited, and the use of training simulators is considered an alternative approach for learning basic skills. To ensure the quality of simulator-based training, it is important to produce simulated ultrasound images that resemble true images to a sufficient degree. This article describes a method that allows corresponding true and simulated ultrasound images to be generated and displayed side by side in real time, thus facilitating an interactive evaluation of ultrasound simulators in terms of image resemblance, real-time characteristics and man-machine interaction. The proposed method could be used to study the realism of ultrasound simulators and how this realism affects the quality of training, as well as being a valuable tool in the development of simulation algorithms.

Degree of Adhesions After Repair of Incisional Hernia

Results of this study suggest that ultrasound can be used to quantitatively estimate the degree of adhesions between intestine and the abdominal wall.

Ultrasound in Abdominal Aortic Aneurysm

Formation and growth of abdominal aortic aneurysms (AAA) may lead to rupture resulting in life threatening haemorrhage. Elective treatment of asymptomatic AAA, either as open surgery or endovascular repair, is recommended when the maximum diameter of the aneurysm exceeds 50-55mm or increases rapidly (Brewster et al., 2003), whereas smaller aneurysms are recommended kept under surveillance. Risk factor modification, such as cessation of smoking, treatment of hypertension and pharmaceutical inhibition of inflammation and protease, could reduce growth in aneurysms kept under surveillance (Baxter et al., 2008; Chaikof et al., 2009; Moll et al., 2011). The size and growth of the aneurysm is monitored using different radiological imaging modalities. Imaging is also important during image guided endovascular repair, and in follow-up examinations after treatment. In this chapter, we describe how ultrasound is currently used in management of abdominal aortic aneurysm, and discuss future potential and challenges of ultrasound for assisting in improved clinical management with regard to patient selection, treatment alternatives and follow-up.

Interactive development of a CT-based tissue model for ultrasound simulation

The objective of this study was to make an interactive method for development of a tissue model, based on anatomical information in computed tomography (CT) images, for use in an ultrasound simulator for training or surgical pre-planning. The method consisted of (1) comparison of true ultrasound B-mode images with corresponding ultrasound-like images, and (2) modification of tissue properties to decrease the difference between these images. Ultrasound-like images that reproduced many, but not all the properties of corresponding true ultrasound images were generated. The tissue model could be used for real-time simulation of ultrasound-like B-mode images on a moderately priced computer.

Manually Steerable Catheter With Improved Agility

Purpose: A prototype steerable catheter was designed for endovascular procedures. This technical pilot study reports the initial experience using the catheter for cannulation of visceral arteries. Technique: The 7F catheter was manually steerable with operator control handle for bending and rotation of the tip. The maximum bending angle was approximately 90° and full 360° rotation of the tip was supported. The study involved 1 pig with 4 designated target arteries: the left and right renal arteries, the superior mesenteric artery, and the celiac trunk. Fluoroscopy with 3-dimensional (3D) overlay showing the ostia from preoperative computed tomography angiography was used for image guidance. The cannulation was considered successful if the guidewire was placed well inside the target artery. In addition to evaluating cannulation success, procedure time and associated radiation doses were recorded. The procedure was performed twice with 2 different operators. Conclusions: Both operators successfully reached all 4 target arteries, demonstrating the feasibility of the steerable catheter for endovascular cannulation of visceral arteries. No contrast medium was used, and median radiation dose was 4.5 mGy per cannulation. An average of approximately 2 minutes was used per cannulation. This study motivates further testing in a more comprehensive study to evaluate reproducibility in several animals and with inclusion of more operators. Further development by integrating the new catheter tool in a navigation system is also an interesting next step, combining fine control of catheter tip movements and 3D image guidance without ionizing radiation.