Michael J. Washburn

A new method to combine contrast-enhanced magnetic resonance imaging during live ultrasound of the breast using volume navigation technique: a study for evaluating feasibility, accuracy and reproducibility in healthy volunteers.

OBJECTIVES To evaluate feasibility, accuracy and reproducibility of combined US-MR of the breast using volume navigation technique. SUBJECTS AND METHODS Five healthy females underwent bilateral contrast-enhanced MR (CE-MR) of the breast in supine position, after positioning three couples of markers on the breast. After CE-MR data uploading in the ultrasound (US) database, manual co-registration was obtained during live US of the breast by means of an electromagnetic transmitter positioned near the subject under examination and two electromagnetic sensors were mounted on the transducer bracket. Transmitter and sensors were connected to a position-sensing unit embedded in the US equipment allowing to track probe position and orientation within the electromagnetic field. Live US image were co-registered to the previously loaded breast CE-MR volume by coupling markers. For each subject, two independent radiologists recorded the examination time and verified twice image alignment using five fixed checkpoints. Pair t Student test and Wilcoxon test were used for statistical analysis. RESULTS In all subjects US and CE-MR images were successfully combined. The examination time was 10±2 vs. 9±4 min, respectively (p=0.642; NS). A total of one hundred measurements of images misalignment were performed: the measurements recorded between the two operators were 0.42±0.32 cm and 0.58±0.41 cm (p=0.161; NS), and 0.50±0.32 cm and 0.56±0.52 cm (p=0.928; NS), respectively. DISCUSSION In our preliminary experience, volume navigation technique appears to be a accurate and reproducible method to combine CE-MR image during unilateral US of the breast.

Real-Time Co-Registration Using Novel Ultrasound Technology: Ex Vivo Validation and In Vivo Applications

OBJECTIVE The study objective was to evaluate whether a novel global position system (GPS)-like position-sensing technology will enable accurate co-registration of images between imaging modalities. Co-registration of images obtained by different imaging modalities will allow for comparison and fusion between imaging modalities, and therefore has significant clinical and research implications. We compared ultrasound (US) and magnetic resonance imaging (MRI) scans of carotid endarterectomy (CEA) specimens using a novel position-sensing technology that uses an electromagnetic (EM) transmitter and sensors mounted on a US transducer. We then evaluated in vivo US-US and US-MRI co-registration. METHODS Thirteen CEA specimens underwent 3.0 Tesla MRI, after which images were uploaded to a LOGIQ E9 3D (GE Healthcare, Wauwatosa, WI) US system and registered by identifying two to three common points. A similar method was used to evaluate US-MRI co-registration in patients with carotid atherosclerosis. For carotid intima-media thickness (C-IMT) measurements, 10 volunteers underwent bilateral carotid US scans co-registered to three-dimensional US maps created on the initial visit, with a repeat scan 2 days later. RESULTS For the CEA specimens, there was a mean of 20 (standard error [SE] 2.0) frames per MRI slice. The mean frame difference, over 33 registration markers, between MRI and US scans for readers 1 and 2 was -2.82 ± 19.32 and 2.09 ± 14.68 (mean ± 95% CI) frames, respectively. The US-MRI intraclass correlation coefficients (ICCs) for the first and second readers were 0.995 and 0.997, respectively. For patients with carotid atherosclerosis, the mean US frames per MRI slice (9 [SE 2.3]) was within range of that observed with CEA specimens. Inter-visit, intra-reader, and inter-reader reproducibility of C-IMT measurements were consistently high (side-averaged ICC >0.9). CONCLUSION Accurate co-registration between US and other modalities is feasible with a GPS-like technology, which has significant clinical and research applicability.

Automated kidney morphology measurements from ultrasound images using texture and edge analysis

In a typical ultrasound scan, a sonographer measures Kidney morphology to assess renal abnormalities. Kidney morphology can also help to discriminate between chronic and acute kidney failure. The caliper placements and volume measurements are often time consuming and an automated solution will help to improve accuracy, repeatability and throughput. In this work, we developed an automated Kidney morphology measurement solution from long axis Ultrasound scans. Automated kidney segmentation is challenging due to wide variability in kidney shape, size, weak contrast of the kidney boundaries and presence of strong edges like diaphragm, fat layers. To address the challenges and be able to accurately localize and detect kidney regions, we present a two-step algorithm that makes use of edge and texture information in combination with anatomical cues. First, we use an edge analysis technique to localize kidney region by matching the edge map with predefined templates. To accurately estimate the kidney morphology, we use textural information in a machine learning algorithm framework using Haar features and Gradient boosting classifier. We have tested the algorithm on 45 unseen cases and the performance against ground truth is measured by computing Dice overlap, % error in major and minor axis of kidney. The algorithm shows successful performance on 80% cases.

Development of an MRI fiducial marker prototype for automated MR-US fusion of abdominal images

External MRI fiducial marker devices are expected to facilitate robust, accurate, and efficient image fusion between MRI and other modalities. Automating of this process requires careful selection of a suitable marker size and material visible across a variety of pulse sequences, design of an appropriate fiducial device, and a robust segmentation algorithm. A set of routine clinical abdominal MRI pulse sequences was used to image a variety of marker materials and range of marker sizes. The most successfully detected marker was 12.7 mm diameter cylindrical reservoir filled with 1 g/L copper sulfate solution. A fiducial device was designed and fabricated from four such markers arranged in a tetrahedral orientation. MRI examinations were performed with the device attached to phantom and a volunteer, and custom developed algorithm was used to detect and segment the individual markers. The individual markers were accurately segmented in all sequences for both the phantom and volunteer. The measured intra-marker spacings matched well with the dimensions of the fiducial device. The average deviations from the actual physical spacings were 0.45± 0.40 mm and 0.52 ± 0.36 mm for the phantom and the volunteer data, respectively. These preliminary results suggest that this general fiducial design and detection algorithm could be used for MRI multimodality fusion applications.

Development of a robust MRI fiducial system for automated fusion of MR‐US abdominal images

Abstract We present the development of a two‐component magnetic resonance (MR) fiducial system, that is, a fiducial marker device combined with an auto‐segmentation algorithm, designed to be paired with existing ultrasound probe tracking and image fusion technology to automatically fuse MR and ultrasound (US) images. The fiducial device consisted of four ~6.4 mL cylindrical wells filled with 1 g/L copper sulfate solution. The algorithm was designed to automatically segment the device in clinical abdominal MR images. The algorithms detection rate and repeatability were investigated through a phantom study and in human volunteers. The detection rate was 100% in all phantom and human images. The center‐of‐mass of the fiducial device was robustly identified with maximum variations of 2.9 mm in position and 0.9° in angular orientation. In volunteer images, average differences between algorithm‐measured inter‐marker spacings and actual separation distances were 0.53 ± 0.36 mm. “Proof‐of‐concept” automatic MR‐US fusions were conducted with sets of images from both a phantom and volunteer using a commercial prototype system, which was built based on the above findings. Image fusion accuracy was measured to be within 5 mm for breath‐hold scanning. These results demonstrate the capability of this approach to automatically fuse US and MR images acquired across a wide range of clinical abdominal pulse sequences.

SU‐E‐I‐68: Preliminary Evaluation of Potential MRI Contrast Materials for the Purpose of MR‐Ultrasound Fusion Application in the Abdomen

PURPOSE To evaluate a variety of materials that might serve as fiducial markers for abdominal MRI-ultrasound fusion applications. METHODS Two experiments were performed: (1) in a phantom, a broad set of candidate materials were evaluated based on visibility in spin echo T1 and T2, and gradient echo T1 and T2* MRI pulse sequences; (2) the leading candidates were evaluated using standard clinical abdominal pulse sequences, both in a phantom and volunteer. Experiment 1 evaluated: two commercial fiducial MRI markers (IZI Medical Products and Beekley Medical); vitamin E and fish oil capsules; water; and copper sulfate solution. Experiment 2 evaluated fish oil capsules and copper sulfate solution. In experiment 2, ultrasound coupling gel was also evaluated. Liquids were poured in wells drilled in a plexiglass base. All scanning was performed with a clinical 1.5T GE Signa Excite scanner. For each pulse sequence, maximum intensity projection images (MIPs) were formed, after removing the phantom signals. Visibility was evaluated by rank ordering signal magnitude in the MIPs. RESULTS In experiment 1, copper sulfate solution and fish oil were superior to the other materials. In experiment 2, ultrasound gel and copper sulfate solution were clearly superior to the other materials, and were easily seen in the all of the phantom and volunteer images. CONCLUSION Based on these preliminary experiments, ultrasound gel and copper sulfate solution appear to be the most promising fiducial marker materials for MRI-ultrasound fusion applications in the abdomen.

Gallbladder quantification in ultrasound using GVF snakes

Accurate identification of the gallbladder contour on ultrasound images is a first step towards reporting quantitative measures of gallbladder function. We present a semi-automatic method for segmentation of gallbladder on ultrasound images. We propose a technique for computing feature images in ultrasound that is based on the hessian computed within a multi-scale framework. We show how the proposed feature image is suited for evolving an active contour subject to the Gradient Vector Flow (GVF) energy. Qualitative results on real-world data show that the proposed approach is capable of segmenting both longitudinal and transverse images of the gall-bladder. Quantitative analysis of segmentation accuracy demonstrates a target overlap of 0.93 ± 0.02 and dice coefficient of 0.94 ± 0.01 with the ground truth.