3D US Data Research Articles

Optimization of Thyroid Volume Determination by Stitched 3D-Ultrasound Data Sets in Patients with Structural Thyroid Disease.

Ultrasound (US) is the most important imaging method for the assessment of structural disorders of the thyroid. A precise volume determination is relevant for therapy planning and outcome monitoring. However, the accuracy of 2D-US is limited, especially in cases of organ enlargements and deformations. Software-based "stitching" of separately acquired 3D-US data revealed precise volume determination in thyroid phantoms. The purpose of this study is to investigate the feasibility and accuracy of 3D-US stitching in patients with structural thyroid disease. A total of 31 patients from the clinical routine were involved, receiving conventional 2D-US (conUS), sensor-navigated 3D-US (3DsnUS), mechanically-swept 3D-US (3DmsUS), and I-124-PET/CT as reference standard. Regarding 3DsnUS and 3DmsUS, separately acquired 3D-US images (per thyroid lobe) were merged to one comprehensive data set. Subsequently, anatomical correctness of the stitching process was analysed via secondary image fusion with the I-124-PET images. Volumetric determinations were conducted by the ellipsoid model (EM) on conUS and CT, and manually drawn segmental contouring (MC) on 3DsnUS, 3DmsUS, CT, and I-124-PET/CT. Mean volume of the thyroid glands was 44.1 ± 25.8 mL (I-124-PET-MC = reference). Highly significant correlations (all p < 0.0001) were observed for conUS-EM (r = 0.892), 3DsnUS-MC (r = 0.988), 3DmsUS-MC (r = 0.978), CT-EM (0.956), and CT-MC (0.986), respectively. The mean volume differences (standard deviations, limits of agreement) in comparison with the reference were -10.50 mL (±11.56 mL, -33.62 to 12.24), -3.74 mL (±3.74 mL, -11.39 to 3.78), and 0.62 mL (±4.79 mL, -8.78 to 10.01) for conUS-EM, 3DsnUS-MC, and 3DmsUS-MC, respectively. Stitched 3D-US data sets of the thyroid enable accurate volumetric determination even in enlarged and deformed organs. The main limitation of high time expenditure may be overcome by artificial intelligence approaches.

Preablation three-dimensional ultrasonography can predict therapeutic effect and local tumor progression after radiofrequency ablation for hepatocellular carcinoma

PurposeTo assess the agreement between ablative margin (AM) predicted by preablation three-dimensional ultrasonography (3D-US) and AM measured on postablation computed tomography (CT)/magnetic resonance (MR) images. MethodsSixty patients with 73 hepatocellular carcinoma nodules were enrolled. 3D-US data were collected immediately after puncture by the electrode before ablation. The maximum distance from the electrode to the edge of the tumor in the plane perpendicular to the electrode (C-plane) was defined as “a” and the diameter of the ablation zone as “b”. We classified predicted AM into “0.5b − a” ≥0 mm as AM(+) or <0 mm as AM(−), and “0.5b − a” ≥3 mm or <3 mm. ResultsForty-eight nodules (66 %) were visualized in the C-plane. There was an agreement between the predicted and measured AMs for 39 (81 %) of the 48 nodules. Local tumor progression was observed in 3 (7%) of 43 nodules with predicted AM(+) and in 2 (40 %) of 5 nodules with predicted AM(−) but was not observed in any of 21 nodules with predicted AM ≥ 3 mm. The local tumor progression rate was significantly lower for nodules with predicted AM(+) compared with predicted AM(−)(p = 0.03), and for nodules with predicted AM ≥ 3 mm compared with predicted AM < 3 mm (p = 0.04). Local progression was detected in 2 (4.7 %) of 42 nodules with a sufficient AM (≥0 mm) on postablation CT/MR images and in 5 (83.3 %) of 6 nodules with an insufficient AM (<0 mm); the difference in progression rate was significant (p = 0.0008). Conclusion3D-US allows prediction of the AM before radiofrequency ablation.

Magnetic Resonance Imaging and Freehand 3-D Ultrasound Provide Similar Estimates of Free Achilles Tendon Shape and 3-D Geometry

The purpose of this study was to assess the similarity of free Achilles tendon shape and 3-D geometry between magnetic resonance imaging (MRI) and freehand 3-D ultrasound (3-DUS) imaging methods. Fourteen elite/sub-elite middle-distance runners participated in the study. MRI and 3-DUS scans of the Achilles tendon were acquired on two separate imaging sessions, and all 3-D reconstructions were performed using identical methods. Shape similarity of free Achilles tendon reconstructed from MRI and 3-DUS data was assessed using Jaccard index, Hausdorff distance and root mean square error (RMSE). The Jaccard index, Hausdorff distance and RMSE values were 0.76 ± 0.05, 2.70 ± 0.70 and 0.61 ± 0.10 mm, respectively. The level of agreement between MRI and 3-DUS for free Achilles tendon volume, length and average cross-sectional area (CSA) was assessed using Bland-Altman analysis. Compared to MRI, freehand 3-DUS overestimated volume, length and average CSA by 30.6 ± 15.8 mm3 (1.1% ± 0.6%), 0.3 ± 0.7 mm (0.6% ± 1.9%) and 0.3 ± 1.42 mm2 (0.4% ± 2.0%), respectively. The upper and lower limits of agreement between MRI and 3-DUS for volume, length and average CSA were −0.4 to 61.7 mm3 (−0.2% to 2.3%), −1.0 to 1.5 mm (−3.2% to 4.5%) and −2.5 to 3.1 mm2 (−3.5% to 4.3%), respectively. There were no significant differences between imaging methods in CSA along the length of the tendon. In conclusion, MRI and freehand 3-DUS may be considered equivalent methods for estimating shape and 3-D geometry of the free Achilles tendon. These findings, together with the practical benefits of being able to assess 3-D Achilles tendon shape and geometry in a laboratory environment and under isometric loading, make 3-DUS an attractive alternative to MRI for assessing 3-D free Achilles tendon macro-structure in future studies.

Three-dimensional ultrasound diagnosis of T-shaped uterus is associated with adverse pregnancy outcomes after embryo transfer.

Is T-shaped uterine cavity morphology associated with adverse pregnancy outcomes after transfer of a single thawed euploid blastocyst? In this secondary analysis of a prospective cohort study, 648 patients with three-dimensional ultrasound (3D-US) data obtained on the day before embryo transfer were categorized into three groups according to uterine cavity morphology: normal (n = 472), intermediate (n = 166) and T-shaped (n = 10). Quantitative uterine cavity dimensions were used to evaluate uterine cavity morphology. Pregnancy outcomes, including live birth, clinical miscarriage and ectopic pregnancy, were compared among the groups. The prevalence of a T-shaped uterus in this cohort was 1.5%. Uterine cavity morphology was strongly associated with the ratio of interostial distance and isthmic diameter (P < 0.01). Live birth rates were 66.5% for normal, 65.7% for intermediate and 40.0% for T-shaped cavity morphology. Women with a T-shaped uterus had an increased risk of clinical miscarriage (40.0% versus 7.0% for normal and 9.0% for intermediate cavity morphology, P < 0.01) and ectopic pregnancy (10.0% versus 1.1% for normal and 1.9% for intermediate cavity morphology, P = 0.05). When evaluating interostial distance and isthmic diameter ratio to determine pregnancy outcomes, a cut-off value of 2 was noted to have weak predictive value for live birth, but not clinical miscarriage or ectopic pregnancy. T-shaped uterine cavity morphology is associated with adverse pregnancy outcomes after transfer of a single thawed euploid blastocyst. Given the low prevalence of this condition, quantifying the magnitude of risk will require a larger cohort of patients.

Quantification of aortic stiffness and wall stress in healthy volunteers and abdominal aortic aneurysm patients using time-resolved 3D ultrasound: a comparison study

Using non-invasive 3D ultrasound, peak wall stress (PWS) and aortic stiffness can be evaluated, which may provide additional criteria in abdominal aortic aneurysm (AAA) risk assessment. In this study, these measures were determined in both young and age-matched individuals, and AAA patients while its relation to age, maximum diameter, and growth was assessed statistically. Time-resolved 3D-US data were acquired for 30 volunteers and 65 AAA patients. The aortic geometry was segmented, and tracked over the cardiac cycle using 3D speckle tracking to characterize the wall motion. Wall stress analysis was performed using finite element analysis. Model parameters were optimized until the model output matched the measured 3D displacements. A significant increase in aortic stiffness was measured between the age-matched volunteers [median 0.58, interquartile range (IQR) 0.48-0.71 kPa⋅m] and the small AAA patients (median 1.84, IQR 1.38-2.46 kPa⋅m; P < 0.001). In addition, an increase in aortic stiffness was evaluated between the small (30-39 mm) and large (≥50 mm) AAAs (median 2.72, IQR 1.99-3.14 kPa⋅m; P = 0.01). The 99th percentile wall stress showed a positive correlation with diameter (ρ = 0.73, P < 0.001), and significant differences between age-matched volunteers and AAA patients. The AAA pathology causes an early and significant increase in aortic stiffness of the abdominal aorta, even after correcting for the expected effect of ageing and differences in arterial pressure. Moreover, some AAAs revealed relatively high PWS, although the maximum diameter was below the threshold for surgical repair. Using the current method, these measures become available during follow-up, which could improve AAA rupture risk assessment.

Semiautomatic quantification of carotid plaque volume with three-dimensional ultrasound imaging

ObjectiveVessel wall volume (VWV) assessed by three-dimensional duplex ultrasound (3DUS) imaging provides a more comprehensive measure of plaque burden than conventional two-dimensional measures of diameter stenosis. We previously demonstrated that manual outlining of the arterial lumen-intima boundary and outer wall boundary can be performed reliably on images obtained with a commercially available 3D-DUS transducer. Manual segmentation, however, is time consuming (∼45 minutes), limiting its clinical translation. We have developed a semiautomatic algorithm (manual selection of the carotid bifurcation image with subsequent automatic plaque outlining) to outline carotid plaques on 3DUS data sets. In this study, we investigated the accuracy, reproducibility, reliability, and time taken by this algorithm. Methods3DUS data sets from 30 patients with asymptomatic ≥50% carotid stenosis underwent manual outlining of lumen-intima boundary and outer wall boundary to measure VWV. Two observers implemented a semiautomatic segmentation algorithm. The algorithm's accuracy was compared with manual outlining using the Pearson correlation coefficient. The Dice similarity coefficient (DSC) and modified-Hausdorff distance (MHD) were used to quantify the geometric similarity of the outlines. We also compared results after an intermediate stage of the algorithm vs the complete algorithm. Reproducibility and the least amount of detectable change in plaque volume were computed for each method. Intraobserver and interobserver metrics for each method were computed using the intraclass correlation coefficient (ICC), coefficient of variability (CV), minimum detectable change (MDC), and standard error of measurement (SEM) of the VWV. ResultsPlaque volume estimates obtained from the semiautomatic algorithm were accurate compared with manual outlining. The Pearson correlation coefficient was 0.76 (P < .001), and measurements were geometrically similar (DSC, 0.85; MHD, 0.48 mm). The algorithm was more reproducible and reliable and could detect smaller changes in plaque volume on repeat imaging (low interobserver variability: ICC, 0.9; CV, 8.22%; MDC, 5.57%; SEM, 1.45%; DSC, 0.88; MHD, 0.43 mm). Intraobserver variability was even lower (ICC, 0.9; CV, 8%; MDC, 3.62%; SEM, 1.31%; DSC, 0.89; MHD, 0.37 mm). Plaque volume estimates at the intermediate stage of the algorithm matched results from the full algorithm (Pearson correlation coefficient, 0.76; DSC, 0.84; MHD, 0.52 mm). The intermediate approach, however, was less reliable than the full algorithm (interobserver: ICC, 0.81; CV, 11.7%; MDC, 9.58%; SEM, 3.46%; DSC, 0.88; MHD, 0.42 mm; intraobserver: ICC, 0.87; CV, 8.6%; MDC, 4.55%; SEM, 1.64%; DSC, 0.89; MHD, 0.38 mm). The full algorithm required ∼14 minutes to implement. However, a quick (7 minutes) and accurate assessment of VWV can be obtained by running only the intermediate stage of the algorithm, although with a loss in repeatability and reliability. ConclusionsWe present a unique algorithm to perform semiautomatic quantification of carotid plaque volume using 3DUS imaging. It is quick (mean time, 14 minutes), accurate, repeatable, and implementable in a clinical environment and in longitudinal studies tracking plaque progression. It reliably detects plaque volume changes as low as 4% to 6% with 95% confidence.

Use of 3D Transvaginal Ultrasound in Prediction of the Need to Morcellate in Women Undergoing Laparoscopic Hysterectomy

In light of recent statements from the United States Food and Drug Administration warning against the use of power morcellation of uterine leiomyomas during laparoscopy, we have evaluated the use of preoperative two- (2D) and three- (3D) dimensional transvaginal ultrasound (US) assessment of uterine volume to predict the need for morcellation in women undergoing laparoscopic hysterectomy (LH). This was a prospective observational study performed between October 2008 and November 2011 in a tertiary referral laparoscopic unit. All women scheduled to undergo LH were included and underwent detailed preoperative transvaginal US. Uterine volumes were calculated using 2D-US measurements (ellipsoid formula), and using Virtual Organ Computer-aided AnaLysis (VOCAL™) having acquired 3D-US volumes of the uterus. Age, parity, need to morcellate and final uterine dry weight at histology were recorded. The estimated uterine volumes were then incorporated into a previously published logistic regression model to predict the need to morcellate for both nulliparous and parous women. The probability threshold cut-off of 0.14 (95% sensitivity) was evaluated in terms of sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and likelihood ratios (LRs). The performance of the models incorporating 2D- and 3D-US calculations were compared with 2D- and 3D-US-generated volumes alone, using receiver-operating characteristics (ROC) curves. Of 76 women who underwent LH during the study period, 79% (n = 60) had complete background and 3D-US data. Their mean age was 43.7 years, 91.7% were parous and 35% underwent morcellation. The greatest uterine volume that did not require morcellation was 404 mL estimated using 3D-US, which corresponded to a uterine volume of 688.8 mL using 2D-US. The smallest uterine volume that required morcellation was 118.9 mL using 3D-US, which corresponded to a uterine volume of 123.4 mL using 2D-US. The 3D-US uterine volume for parous women with a sensitivity of 95% based on ROC-curve analysis was approximately 120 mL, which equated to a predicted probability of morcellation cut-off of 0.14. For this cut-off, specificity was 55.00%, PPV was 51.35%, NPV was 95.65%, LR+ was 2.11 and LR- was 0.09. Areas under the ROC curves for the morcellation logistic regression model were 0.769 (95% CI, 0.653-0.886) and 0.586 (95% CI, 0.419-0.753) using uterine volumes obtained by 3D-US and by 2D-US, respectively, and they were 0.938 (95% CI, 0.879-0.996) and 0.815 (95% CI, 0.681-0.948) using 3D-US and 2D-US volumes alone. In conclusion, the need to morcellate can be predicted preoperatively using 3D-US uterine volumes obtained by transvaginal US with a fair degree of accuracy. Uteri with volumes smaller than 120 mL at 3D-US are very unlikely to require morcellation. The incorporation of 3D-US-estimated uterine volume into the previously published logistic regression model does not seem to confer any significant improvement when compared with 3D-US uterine volume alone to predict the need to morcellate in women undergoing total LH.

Multi-scale RANSAC algorithm for needle localization in 3D ultrasound guided puncture surgery.

Correct localization of the needle is of vital importance to guarantee successful puncture. The complexity of real US data increases the difficulties. A multi-scale random sample consensus (MS-RANSAC) algorithm is proposed in this paper to locate the needle in complicated 3D US data. The algorithm uses the radius difference between needle and other tubular human tissues to extract the correct needle location. The performance of classic RANSAC and MS-RANSAC are compared using three different datasets. Results show that MS-RANSAC can locate needle correctly in complicated condition where classic RANSAC cannot. A parallel framework of the algorithm is designed and implemented using CUDA, making it usable in real time and online.

MRI and three dimensional ultrasonography in the assessment of pulmonary hypoplasia in fetuses with urinary tract anomalies

PurposeTo analyze the correlation and agreement between three dimensional (3D) US and MRI in the assessment of pulmonary volumes of fetuses with different types of urinary tract malformations (UTM) and high-risk of pulmonary hypoplasia (PH). Patients and methodsThirty-nine fetuses with various UTM, at risk for PH were involved in this cross-sectional study. 3D volume US data sets of the fetal lungs were acquired. The right, left and total lung volumes were calculated separately using the virtual organ computer-aided analysis (VOCAL) method with a 30° rotation. MRI of fetal lung was obtained with assessment of signal intensity and lung volumetry. Comparison between mean lung volumes was performed using unpaired t test. Agreement between the 3D-US and MRI methods was done using Cohen kappa test. ResultsGood agreement was detected between the two methods (Kappa=0.629, p=0.001). The measured lung volumes by 3D-US were smaller than those measured by MRI (p>0.05, non-significant). MRI showed greater specificity, PPV and diagnostic accuracy (100% each) than 3D-US (50%, 88.9% and 90% respectively). ConclusionThere is a good concordance between 3D-US and MRI in the evaluation of PH in fetuses with UTM. MRI could be reserved for borderline cases of pulmonary hypoplasia and the difficult diagnostic situations.

The periconception maternal cardiovascular risk profile influences human embryonic growth trajectories in IVF/ICSI pregnancies.

Is the maternal cardiovascular (CV) risk profile associated with human embryonic growth trajectories and does the mode of conception affect this association? This small study suggests that the maternal CV risk profile is inversely associated with first trimester embryonic growth trajectories in in vitro fertilization (IVF)/intra-cytoplasmic sperm injection (ICSI) pregnancies, but not in spontaneously conceived pregnancies. Maternal high-blood pressure and smoking affect placental function, accompanied by increased risk of fetal growth restriction and low-birthweight. Mothers who experience pregnancies complicated by fetal growth restriction are at increased risk of CV disease in later life. In a prospective periconception birth cohort conducted in a tertiary hospital, 111 singleton ongoing pregnancies with reliable pregnancy dating, no pre-existing maternal disease and no malformed live borns were investigated. Spontaneously conceived pregnancies with a reliable first day of the last menstrual period and a regular menstrual cycle of 25-31 days only (n = 66) and IVF/ICSI pregnancies (n = 45) were included. Women underwent weekly three-dimensional ultrasound scans (3D US) from 6- to 13-week gestational age. To estimate embryonic growth, serial crown-rump length (CRL) measurements were performed using the V-Scope software in a BARCO I-Space. Maternal characteristics and CV risk factors were collected by self-administered questionnaires. The CV risk profile was created based on a score of risk factors, including maternal age, body-mass index, CV disease in the family, diet and smoking. Quartiles of the CV risk score were calculated. Associations between the CV risk score and embryonic growth were assessed using square root transformed CRL in multivariable linear mixed model analyses. From the 111 included pregnancies, 696 3D US data sets were obtained of which 637 (91.5%) CRLs could be measured. In the total group, The CV risk score was inversely, but not significantly associated with embryonic growth (-0.03√mm; P = 0.291). Stratified by mode of conception, the CV risk score was inversely and significantly associated with embryonic growth (β = -0.04√mm; P = 0.025, adjusted for possible confounders) in the IVF/ICSI group. Compared with the first quartile, embryos in the upper quartile were 10.4% smaller at 6(+0) weeks (4.4 versus 4.9 mm) and 3.1% smaller at 12(+0) weeks (56.5 versus 58.4 mm) of gestation. Although the CV risk score was slightly, but significantly, higher in women conceiving spontaneously compared with those undergoing IVF/ICSI treatment [CV risk score = 2.06 (SD: 1.23) and 1.60 (SD: 1.15), respectively], no association was established with embryonic growth in that particular group. Participants included in the present cohort are women with a singleton ongoing pregnancy without any pre-existing disease and selected from a tertiary hospital. Hence, they represent a selected group of women. Larger and population-based periconception birth cohort studies are recommended to demonstrate external validity. Differences in embryonic growth between pregnancies conceived spontaneously and after IVF/ICSI treatment in relation with CV risk factors substantiate the importance of more investigation into differences in sensitivity of endometrial, endothelial, placental and embryonic tissues. Funded by the Department of Obstetrics and Gynaecology, Erasmus MC, University Medical Centre, Rotterdam, the Netherlands. The authors declare no conflict of interest.

Visualization of lenticulostriate arteries during insular low-grade glioma surgeries by navigated 3D ultrasound power Doppler: technical note.

OBJECTIVE Resection of insular gliomas is challenging. In cases of intraoperative injury to the lenticulostriate arteries (LSAs), the usual result is a dense hemiplegia. LSAs are usually localized just behind the medial tumor border but they can also be encased by the tumor. Thus, exact localization of these perforators is important. However, intraoperative localization of LSAs using conventional neuronavigation can be difficult due to brain shift. In this paper, the authors present a novel method of intraoperative LSA visualization by navigated 3D ultrasound (3DUS) power Doppler. This technique enables almost real-time imaging of LSAs and evaluation of their shift during insular tumor resections. METHODS Six patients harboring insular Grade II gliomas were consecutively operated on at the Department of Neurosurgery in Bratislava using visualization of LSAs by navigated 3DUS power Doppler. In all cases, the 3DUS data were repeatedly updated to compensate for the brain shift and display the actual position of LSAs and residual tumor. RESULTS Successful visualization of LSAs was achieved in all cases. During all surgeries, the distance between the bottom of the resection cavity and LSAs could be accurately evaluated; in all tumors the resection approached the LSAs and only a minimal amount of tissue covering these perforators was intentionally left in place to avoid injury to them. CONCLUSIONS Visualization of LSAs by navigated 3DUS power Doppler is a useful tool that may help to prevent injury of LSAs during removal of insular low-grade gliomas. However, reliability of this method has to be carefully evaluated in further studies.

Quantification of kidneys from 3D ultrasound in pediatric hydronephrosis.

This paper introduces a complete framework for the quantification of renal structures (parenchyma, and collecting system) in 3D ultrasound (US) images. First, the segmentation of the kidney is performed using Gabor-based appearance models (GAM), a variant of the popular active shape models, properly tailored to the imaging physics of US image data. The framework also includes a new graph-cut based method for the segmentation of the collecting system, including brightness and contrast normalization, and positional prior information. The significant advantage (p = 0.03) of the new method over previous approaches in terms of segmentation accuracy has been successfully verified on clinical 3DUS data from pediatric cases with hydronephrosis. The promising results obtained in the estimation of the volumetric hydronephrosis index demonstrate the potential of our new framework to quantify anatomy in US and asses the severity of hydronephrosis.

The Potential of Transabdominal 3D Color Doppler Ultrasonography for Diagnosis of Gastric Varices

To examine the potential of transabdominal 3-dimensional (3D) color Doppler ultrasonography (3D-US) as a noninvasive tool to characterize gastric varices. This was a prospective study in which endoscopy was performed on 107 patients with chronic liver disease. Among these patients, 70 (42 males, 28 females) had gastric varices (46 fundal varices, 24 cardia varices; 30 small, 28 medium, and 12 large), and the 37 patients (25 males, 12 females) without gastric varices served as controls. The 3D-US data and endoscopic findings were compared with respect to grade, location, and similarity of varices. The sensitivity and specificity of the 3D-US technique to detect gastric varices were 88.6% (62/70) and 100% (37/37), respectively. Although all fundal varices appeared adjacent to the posterior gastric wall, cardia varices were detected separately from the wall with a mean distance of 21.2 mm. The vascular volumes (mL) were 0.84±0.71 in small varices, 5.52±3.81 in medium varices, and 10.9±6.3 in large varices, with significant differences between different grades. The best cutoff value to detect medium-grade/large-grade gastric varices was 2.0 mL, with 83.3% sensitivity and 95.8% specificity. Seventy-nine percent (55/70) of patients showed partial resemblance or better between the 3D images and the endoscopic findings with good interreviewer agreement. 3D-US can quantitatively characterize gastric varices noninvasively in terms of grade, location, and appearance. This approach has the potential to improve objectivity and reduce invasiveness in the management of gastric varices.

Measuring Fetal Volume During Late First Trimester by Three-Dimensional Ultrasonography Using Virtual Organ Computer-Aided Analysis

Our aim was to determine whether rotating the fetus over its largest axis and reducing the rotational step angle can improve reliability/agreement of fetal volume measurements obtained with three-dimensional ultrasonography (3-DUS). Two observers acquired three 3-DUS data sets for a fetus with a crown–rump length between 45 and 84 mm. These observers determined the fetal volume using virtual organ computer-aided analysis (VOCAL), by three different methods, with a rotational step angle of 30°: (1) minimal manipulation of the 3-DUS data set, fetus rotated over any axis; (2) manipulation of the 3-DUS data set until the fetus could be seen in a standardized manner, fetus rotated over its anteroposterior axis; (3) same 3-DUS data set manipulation, fetus rotated over its longitudinal axis. Intra- and inter-observer reliability/agreement was determined with intra-class correlation coefficients and limits of agreement. In addition, we tested the method that provided the best reliability/agreement results using 15° and 9° of rotational step angles. The time taken to manipulate 3-DUS and determine fetal volume was recorded. The best intra- and inter-observer reliability/agreement results were observed when the fetus was rotated over its longitudinal axis. Reducing rotational step angle to 15° or 9° did not further improve reliability/agreement. The observer took approximately 1 min to determine fetal volume using this method. Our findings indicate that fetal volume should be determined by rotating the fetus over its longitudinal axis, at a rotational step angle of 30°, which is relatively fast and allows analysis of fetal volume with good reliability and agreement.

Registration of 3D Ultrasound Through an Air–Tissue Boundary

In this study we evaluated a new method for registering three-dimensional ultrasound (3DUS) data to external coordinate systems. First, 3DUS was registered to the stereo endoscope of a da Vinci Surgical System by placing a registration tool against an air-tissue boundary so that the 3DUS could image ultrasound fiducials while the stereo endoscope could image camera markers on the same tool. The common points were used to solve the registration between the 3DUS and camera coordinate systems. The target registration error (TRE) when imaging through a PVC tissue phantom ranged from 3.85 1.76 mm to 1.82 1.03 mm using one to four registration tool positions. TRE when imaging through an ex-vivo liver tissue sample ranged from 2.36 1.01 mm to 1.51 0.70 mm using one to four registration tool positions. Second, using a similar method, 3DUS was registered to the kinematic coordinate system of a da Vinci Surgical System by using the da Vinci surgical manipulators to identify common points on an air-tissue boundary. TRE when imaging through a PVC tissue phantom was 0.95 0.38 mm. This registration method is simpler and potentially more accurate than methods using commercial motion tracking systems. This method may be useful in the future in augmented reality systems for laparoscopic and robotic-assisted surgery.

Comparison of different reconstruction algorithms for three‐dimensional ultrasound imaging in a neurosurgical setting

Freehand three-dimensional ultrasound imaging (3D-US) is increasingly used in image-guided surgery. During image acquisition, a set of B-scans is acquired that is distributed in a non-parallel manner over the area of interest. Reconstructing these images into a regular array allows 3D visualization. However, the reconstruction process may introduce artefacts and may therefore reduce image quality. The aim of the study is to compare different algorithms with respect to image quality and diagnostic value for image guidance in neurosurgery. 3D-US data sets were acquired during surgery of various intracerebral lesions using an integrated ultrasound-navigation device. They were stored for post-hoc evaluation. Five different reconstruction algorithms, a standard multiplanar reconstruction with interpolation (MPR), a pixel nearest neighbour method (PNN), a voxel nearest neighbour method (VNN) and two voxel based distance-weighted algorithms (VNN2 and DW) were tested with respect to image quality and artefact formation. The capability of the algorithm to fill gaps within the sample volume was investigated and a clinical evaluation with respect to the diagnostic value of the reconstructed images was performed. MPR was significantly worse than the other algorithms in filling gaps. In an image subtraction test, VNN2 and DW reliably reconstructed images even if large amounts of data were missing. However, the quality of the reconstruction improved, if data acquisition was performed in a structured manner. When evaluating the diagnostic value of reconstructed axial, sagittal and coronal views, VNN2 and DW were judged to be significantly better than MPR and VNN. VNN2 and DW could be identified as robust algorithms that generate reconstructed US images with a high diagnostic value. These algorithms improve the utility and reliability of 3D-US imaging during intraoperative navigation.

Development of a semi-automated method for mitral valve modeling with medial axis representation using 3D ultrasound

Precise 3D modeling of the mitral valve has the potential to improve our understanding of valve morphology, particularly in the setting of mitral regurgitation (MR). Toward this goal, the authors have developed a user-initialized algorithm for reconstructing valve geometry from transesophageal 3D ultrasound (3D US) image data. Semi-automated image analysis was performed on transesophageal 3D US images obtained from 14 subjects with MR ranging from trace to severe. Image analysis of the mitral valve at midsystole had two stages: user-initialized segmentation and 3D deformable modeling with continuous medial representation (cm-rep). Semi-automated segmentation began with user-identification of valve location in 2D projection images generated from 3D US data. The mitral leaflets were then automatically segmented in 3D using the level set method. Second, a bileaflet deformable medial model was fitted to the binary valve segmentation by Bayesian optimization. The resulting cm-rep provided a visual reconstruction of the mitral valve, from which localized measurements of valve morphology were automatically derived. The features extracted from the fitted cm-rep included annular area, annular circumference, annular height, intercommissural width, septolateral length, total tenting volume, and percent anterior tenting volume. These measurements were compared to those obtained by expert manual tracing. Regurgitant orifice area (ROA) measurements were compared to qualitative assessments of MR severity. The accuracy of valve shape representation with cm-rep was evaluated in terms of the Dice overlap between the fitted cm-rep and its target segmentation. The morphological features and anatomic ROA derived from semi-automated image analysis were consistent with manual tracing of 3D US image data and with qualitative assessments of MR severity made on clinical radiology. The fitted cm-reps accurately captured valve shape and demonstrated patient-specific differences in valve morphology among subjects with varying degrees of MR severity. Minimal variation in the Dice overlap and morphological measurements was observed when different cm-rep templates were used to initialize model fitting. This study demonstrates the use of deformable medial modeling for semi-automated 3D reconstruction of mitral valve geometry using transesophageal 3D US. The proposed algorithm provides a parametric geometrical representation of the mitral leaflets, which can be used to evaluate valve morphology in clinical ultrasound images.

Quantitative assessment of changes in carotid plaques during cilostazol administration using three-dimensional ultrasonography and non-gated magnetic resonance plaque imaging

Cilostazol, an antiplatelet agent, is reported to induce the regression of atherosclerotic changes. However, its effects on carotid plaques are unknown. Hence, we quantitatively investigated the changes that occur within carotid plaques during cilostazol administration using three-dimensional (3D) ultrasonography (US) and non-gated magnetic resonance (MR) plaque imaging. We prospectively examined 16 consecutive patients with carotid stenosis. 3D-US and T1-weighted MR plaque imaging were performed at baseline and 6 months after initiating cilostazol therapy (200 mg/day). We measured the volume and grayscale median (GSM) of the plaques from 3D-US data. We also calculated the contrast ratio (CR) of the carotid plaque against the adjacent muscle and areas of the intraplaque components: fibrous tissue, lipid, and hemorrhage components. The plaque volume on US decreased significantly (median at baseline and 6 months, 0.23 and 0.21 cm(3), respectively; p = 0.03). In the group exhibiting a plaque volume reduction of more than 10%, GSM on US increased significantly (24.8 and 71.5, respectively; p = 0.04) and CR on MRI decreased significantly (1.13 and 1.04, respectively; p = 0.02). In this group, in addition, the percent area of the fibrous component on MRI increased significantly (68.6% and 79.4%, respectively; p = 0.02), while those of the lipid and hemorrhagic components decreased (24.9% and 20.5%, respectively; p = 0.12) (1.0% and 0.0%, respectively; p = 0.04). There were no substantial changes in intraplaque characteristics in either US or MRI in the other group. 3D-US and MR plaque imaging can quantitatively detect changes in the size and composition of carotid plaques during cilostazol therapy.

3D ultrasound DICOM data of the thyroid gland

SummaryPurpose: It has recently become possible to generate and archive three-dimensional ultrasound (3D-US) volume data with the DICOM standard Enhanced Ultrasound Volume Storage (EUVS). The objective of this study was to examine the application of the EUVS standard based on the example of thyroid ultrasound. Patients, methods: 32 patients, who were referred for thyroid diagnosis, were given a 3D-US examination of the thyroid gland (GE Voluson E8, convex 3D probe RAB4–8-D). The 3D data sets were exported to EUVS. Necessary additions to DICOM entries and transformation into an established DICOM standard were carried out. The visual assessment and volume measurements were performed by two experts on nuclear medicine using standard software in our hospital. Results: In 24/32 (75%) of the patients, the whole organ was successfully recorded in a single 3D scan; in 8/32 (25%), only part of organ could be covered. In all cases, 3D-US data could be exported and archived. After supplementing the DICOM entry Patient Orientation and transformation into the DICOM PET format, 3D-US data could be displayed in the correct orientation and size at any viewing workstation and any web browser-based PACS viewer. Afterwards, 3D processing such as multiplanar reformation, volumetric measurements and image fusion with data of other cross sectional modalities could be performed. The intraclass correlation of the volume measurements was 0,94 and the interobserver variability was 5.7%. Conclusion: EUVS allows the generation, distribution and archiving of 3D-US data of the thyroid, facilitates a second reading by another physician and creates conditions for advanced 3D processing using routine software

Stand-alone 3D-ultrasound navigation after failure of conventional image guidance for deep-seated lesions

Image guidance has proven to be an important tool in surgery for deep-seated or eloquently located cavernomas. However, neuronavigation depending on preoperative images can fail. Thus, the displayed anatomy might be distorted already during the approach. This report demonstrates the use of three-dimensional intraoperative ultrasound (3D-US) as a rescue tool, when conventional navigation is erroneous. Two patients with symptomatic cavernomas, the one located subcortically in the right peritrigonum, the other in the left thalamus, were operated in our clinic via an image-guided approach. An integrated ultrasound-navigation system was used for neuronavigation. In both cases, navigation based on preoperative MRI failed after the craniotomy because patient-to-image registration was lost. In both cases, a simple registration of the patient's orientation was performed. Then a 3D-US volume was acquired and navigation was performed using the 3D-US data set. This is accurate as image acquisition and navigation are done in the same system. The cavernoma was visualized without difficulties in both cases. It could be reached directly via the ultrasound-guided approach. Patients' symptoms improved postoperatively and a complete resection could be documented. Two cavernomas were successfully resected using 3D-US guidance. In our experience, stand-alone 3D-US navigation is an effective option if conventional MRI-based navigation fails.