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.

Endovascular image-guided navigation: validation of two volume-volume registration algorithms.

The limited volume covered by intraoperatively acquired CT scans makes the use of navigation systems difficult. Preoperative images cover a larger volume of interest. Hence, reliable registration of high quality preoperative to intraoperative CT will provide the necessary image information required for navigation. This study evaluates two algorithms (Siemens, CAMP) for volume-volume registration for usage during endovascular navigation. Twenty patients treated for abdominal aortic aneurysm were scanned with pre-, intra- and postoperative CT. Six data sets were excluded due to variations in image acquisition parameters and severe artifacts. Fourteen intra- and postoperative datasets were registered ten times with both algorithms, altogether 140 registrations for each program. In all data sets five specified landmarks placed by two radiologists were used to evaluate registration accuracy. The distance between the paired landmarks in the registered intra- and postoperative volumes was measured and the root mean square value calculated. Reference registrations were based on rigid body registration of the five landmarks in the intra- and postoperative volumes. Registration accuracy (mean ± SD) was for Siemens 5.05 ± 4.74 mm, for CAMP 4.02 ± 1.52 mm and for the reference registrations 2.72 ± 1.18 mm. The registration algorithms differed significantly, p < 0.001.

Freehand 3D ultrasound reconstruction algorithms--a review.

Three-dimensional (3D) ultrasound (US) is increasingly being introduced in the clinic, both for diagnostics and image guidance. Although dedicated 3D US probes exist, 3D US can also be acquired with the still frequently used two-dimensional (2D) US probes. Obtaining 3D volumes with 2D US probes is a two-step process. First, a positioning sensor must be attached to the probe; second, a reconstruction of a 3D volume can be performed into a regular voxel grid. Various algorithms have been used for performing 3D reconstruction based on 2D images. Up till now, a complete overview of the algorithms, the way they work and their benefits and drawbacks due to various applications has been missing. The lack of an overview is made clear by confusions about algorithm and group names in the existing literature. This article is a review aimed at explaining and categorizing the various algorithms into groups, according to algorithm implementation. The algorithms are compared based on published data and our own laboratory results. Positive and practical uses of the various algorithms for different applications are discussed, with a focus on image guidance.

Intraoperative navigated 3-dimensional ultrasound angiography in tumor surgery.

BACKGROUND: Avoiding damage to blood vessels is often the concern of the neurosurgeon during tumor surgery. Using angiographic image data in neuronavigation may be useful in cases where vascular anatomy is of special interest. Since 2003, we have routinely used 3D ultrasound angiography in tumor surgery, and between January 2003 and May 2005, 62 patients with different tumors have been operated using intraoperative 3D ultrasound angiography in neuronavigation. METHODS: An ultrasound-based neuronavigation system was used. In addition to 3D ultrasound tissue image data, 3D ultrasound angiography (power Doppler) image data were acquired at different stages of the operation. The value and role of navigated 3D ultrasound angiography as judged by the surgeon were recorded. RESULTS: We found that intraoperative ultrasound angiography was easy to acquire and interpret, and that image quality was sufficient for neuronavigation. In 26 of 62 cases, ultrasound angiography was found to be helpful by visualizing hidden vessels adjacent to and inside the tumor, facilitating tailored approaches and safe biopsy sampling. CONCLUSIONS: Intraoperative 3D ultrasound angiography is straightforward to use, image quality is sufficient for image guidance, and it adds valuable information about hidden vessels, increasing safety and facilitating tailored approaches. Furthermore, with updated 3D ultrasound angiography imaging, accuracy of neuronavigation may be maintained in cases of brain shift.

Computer-assisted 3D ultrasound-guided neurosurgery: technological contributions, including multimodal registration and advanced display, demonstrating future perspectives.

BACKGROUND: Navigation systems are now frequently being used for guiding surgical procedures. Existing neuronavigation systems suffer from the lack of updated images when tissue changes during surgery as well as from user-friendly displays of all essential images for accurate and safe surgery guidance. METHODS: We have developed various new technologies for improved neuronavigation. Using intraoperative 3D ultrasound (US) imaging, we have developed various registration algorithms for using and updating a complete multimodal and multivolume 3D map for navigation. RESULTS: We experienced that advanced multimodal visualization makes it easy to interpret information from several image volumes and modalities simultaneously. Using high quality intraoperative 3D ultrasound, essential preoperative information could be corrected due to brain shift. fMRI and other important preoperative data could then be used together with intraoperative ultrasound imaging for more accurate, safer and improved guidance of therapy. CONCLUSIONS: We claim that new features, as demonstrated in the present paper, using intraoperative 3D ultrasound in combination with advanced registration and display algorithms will represent important contributions towards more accurate, safer and more optimized future patient treatment.

A new method for analysis of motion of carotid plaques from RF ultrasound images.

Motion of carotid artery plaques during the cardiac cycle may contribute to plaque disruption and embolism. We have developed a computerized method that objectively analyzes such motion from a sequence of ultrasound (US) radiofrequency (RF) images. A displacement vector map is obtained by 2-D correlation of local areas in consecutive images. From this map, motion dynamics can be quantified and presented as function of time, spatial (image) coordinates or as single numbers. Correct functionality has been verified on laboratory data. Applied to patient data, the method gives temporal results that correlate well with ECG data and the calculated peak systolic velocities of typically 10 mm/s agree well with values reported in the literature. The spatial analysis demonstrates that different plaque regions may exhibit different motion patterns that may cause internal stress, leading to fissures and plaque disruption. Thus, the motion analysis method may provide new and important information about the plaque characteristics and the prospective risk of cerebrovascular events.

Accuracy evaluation of a 3D ultrasound-based neuronavigation system.

We have investigated the 3D navigation accuracy of a frameless ultrasound-based neuronavigation system (SonoWand) for surgical planning and intraoperative image guidance. In addition, we present a detailed description and review of the error sources associated with surgical neuronavigation based on preoperative MRI data and intraoperative ultrasound. A phantom with 27 precisely defined points was scanned with ultrasound by various translation and tilt movements of the ultrasound probe (180 3D scans in total), and the 27 image points in each volume were located using an automatic detection algorithm. These locations were compared to the physically measured locations of the same 27 points. The accuracy of the neuronavigation system and the effect of varying acquisition conditions were found through a thorough statistical analysis of the differences between the two point sets. The accuracy was found to be 1.40 +/- 0.45 mm (arithmetic mean) for the ultrasound-based neuronavigation system in our laboratory setting. Improper probe calibration was the major contributor to this figure. Based on our extensive data set and thorough evaluation, the accuracy found in the laboratory setting is expected to be close to the overall clinical accuracy for ultrasound-based neuronavigation. Our analysis indicates that the overall clinical accuracy may be as low as 2 mm when using intraoperative imaging to compensate for brain shift.

Brain operations guided by real-time two-dimensional ultrasound: new possibilities as a result of improved image quality.

OBJECTIVE: In 1995, a project was initiated in Trondheim, Norway, to investigate various possibilities for more frequent use of ultrasound in brain surgery. Since that time, the quality of ultrasonic images has improved considerably through technological adjustment of parameters. The objective of the present study was to explore essential clinical parameters required for the successful use of ultrasonic guidance in brain surgery. METHODS: During the study period, several surgical setups designed to optimize the use of intraoperative real-time two-dimensional ultrasonic imaging were explored. These included various positions of the ultrasound probe in relation to both the operation cavity and the lesion, as well as the position of the operation channel in relation to the gravity line. RESULTS: All lesions from the latest period (1997-2001; n = 114) were depicted well by ultrasound imaging, with the exception of two cases. High image quality and direct image guidance of the tool were maintained best throughout the operation by imaging through an intact dura and at an angle relative to a vertical operation channel. All tumor operations were performed without complications, and ultrasound imaging was found to be an important factor in the detection of remaining tumor tissue at the conclusion of surgery. For 14 low vascular tumors, the operation was guided only by ultrasound imaging. No bleeding complications occurred. A method of minimally invasive ultrasound-guided evacuation of hematomas was developed. In 19 patients, the method was found to be efficient (i.e., >90% of the hematoma was evacuated) and without complications, except for one patient who experienced rebleeding. CONCLUSION: With proper planning and surgical setup, ultrasound imaging may provide acceptable image quality for use in image-guided brain operations.

Neuronavigation by intraoperative three-dimensional ultrasound: initial experience during brain tumor resection.

OBJECTIVE: Three-dimensional (3-D) ultrasound is an intraoperative imaging modality used in neuronavigation as an alternative to magnetic resonance imaging (MRI). This article summarizes 4 years of clinical experience in the use of intraoperative 3-D ultrasound integrated into neuronavigation for guidance in brain tumor resection. METHODS: Patients were selected for inclusion in the study on the basis of the size and location of their lesion. Preoperative 3-D MRI data were registered and used for planning as in other conventional neuronavigation systems. Intraoperative 3-D ultrasound images were acquired three to six times, and tumor resection was guided on the basis of these updated 3-D images. RESULTS: Intraoperative 3-D ultrasound represents a good solution to the problem of brain shift in neuronavigation because it easily provides an updated, and hence more accurate, map of the patient's true anatomy in all phases of the operation. Ultrasound makes it possible to follow the progression of the operation, and it improves the radicality of tumor resection by detecting tumor tissue that would remain if the imaging technology had not been used (in 53% of the cases). Integration of 3-D ultrasound with navigation technology solves the orientation problem experienced previously with two-dimensional ultrasound in neurosurgery. The technology makes it possible to directly compare intraoperative ultrasound and MRI data regarding visualization of the lesion. Ultrasound image quality is useful for guiding surgical procedures. CONCLUSION: Intraoperative 3-D ultrasound seems to provide a time- and cost-effective way to update high-quality 3-D maps used in neuronavigation.