Measurement Of Abdominal Aortic Aneurysms Research Articles

Computed tomography-based automated measurement of abdominal aortic aneurysm using semantic segmentation with active learning

Accurate measurement of abdominal aortic aneurysm is essential for selecting suitable stent-grafts to avoid complications of endovascular aneurysm repair. However, the conventional image-based measurements are inaccurate and time-consuming. We introduce the automated workflow including semantic segmentation with active learning (AL) and measurement using an application programming interface of computer-aided design. 300 patients underwent CT scans, and semantic segmentation for aorta, thrombus, calcification, and vessels was performed in 60–300 cases with AL across five stages using UNETR, SwinUNETR, and nnU-Net consisted of 2D, 3D U-Net, 2D-3D U-Net ensemble, and cascaded 3D U-Net. 7 clinical landmarks were automatically measured for 96 patients. In AL stage 5, 3D U-Net achieved the highest dice similarity coefficient (DSC) with statistically significant differences (p < 0.01) except from the 2D–3D U-Net ensemble and cascade 3D U-Net. SwinUNETR excelled in 95% Hausdorff distance (HD95) with significant differences (p < 0.01) except from UNETR and 3D U-Net. DSC of aorta and calcification were saturated at stage 1 and 4, whereas thrombus and vessels were continuously improved at stage 5. The segmentation time between the manual and AL-corrected segmentation using the best model (3D U-Net) was reduced to 9.51 ± 1.02, 2.09 ± 1.06, 1.07 ± 1.10, and 1.07 ± 0.97 min for the aorta, thrombus, calcification, and vessels, respectively (p < 0.001). All measurement and tortuosity ratio measured − 1.71 ± 6.53 mm and − 0.15 ± 0.25. We developed an automated workflow with semantic segmentation and measurement, demonstrating its efficiency compared to conventional methods.

Fully automatic detection and measurement of abdominal aortic aneurysm using artificial intelligence

Abstract Background Abdominal aortic aneurysm (AAA) is a life-threatening disease, and the decision to treat relies on the exact detection and size of AAA, which is time-consuming and prone to high inter-reader variability. Artificial intelligence (AI) has revealed new insights into the management of AAA, but it needs large amounts of contrast-enhanced CT data. Moreover, there has been few studies about the fully automatic detection and measurement of AAA. Purpose The aim of this study is to evaluate the accuracy of AI for the detection and measurement of AAA using small amounts of non-contrast CT. Methods We retrospectively collected 145 non-contrast CT scans of suspected AAA. The first step was to manually segment the 30 non-contrast CT scans to create training data for the AI, and the second step was to improve accuracy using 9 of non-contrast CT scans. Image processing is used to identify the AAA area and automatically measure the size. The AAA region was identified from the amount of change in the minor-axis, and the maximum minor-axis diameter was calculated using elliptical fitting. To evaluate the diagnostic utility of AI for the detection of AAA, the sensitivity and positive predictive value were calculated referred for the diagnostic report. The reproducibility in the size of AAA was assessed using intraclass correlation coefficients (ICCs) between the diagnostic report and AI. Results Among 145 of non-contrast CT scans, 111 had AAA. The sensitivity and positive predictive value of AI for the detection of AAA were 94.6% and 53.8%, respectively (Picture 1). The size of AAA calculated by the AI (42.5±8.8 mm) showed a strong correlation with the those of diagnostic reports (44.6±8.4 mm; ICCs = 0.97) (Picture 2). Conclusions AI represents a useful tool in the fully automatic detection and measurement of AAA using small amounts of non-contrast CT. Funding Acknowledgement Type of funding sources: None. Picture 1Picture 2

Abdominal aortic aneurysm measurement at CT/MRI: potential clinical ramifications of non-standardized measurement technique and importance of multiplanar reformation.

Accurate and reproducible measurement of abdominal aortic aneurysm (AAA) size is an essential component of patient management, and most reliably performed at CT using a multiplanar reformat (MPR) strategy. This approach is not universal, however. This study aims to characterize the measurement error present in routine clinical assessment of AAAs and the potential clinical ramifications. Patients were included if they had AAA assessed by CT and/or MRI at two time points at least 6 months apart. Clinical maximal AAA diameter, assessed by non-standardized methods, was abstracted from the radiology report at each time point and compared to the reference aneurysm diameter measured using a MPR strategy. Discrepancies between clinical and reference diameters, and associated aneurysm enlargement rates were analyzed. Two hundred thirty patients were included, with average follow-up 3.3±2.5 years. When compared to MPR-derived diameters, clinical aneurysm measurement inaccuracy was, on average, 3.3 mm. Broad limits of agreement were found for both clinical diameters [-6.7 to +6.5 mm] and aneurysm enlargement rates [-4.6 to +4.2 mm/year] when compared to MPR-based measures. Of 78 AAAs measuring 5-6 cm by the MPR method, 21 (26.9%) were misclassified by the clinical measurement with respect to a common repair threshold (5.5 cm), of which 5 were misclassified as below, and 16 were misclassified as above the threshold. The clinical use of non-standardized AAA measurement strategies can lead to incorrect classification of AAAs as larger or smaller than the commonly accepted repair threshold of 5.5 cm and can induce large errors in quantification of aneurysm enlargement rate.

Observational Study Comparing the Measurement of Abdominal Aortic Aneurysm (AAA) Diameter in Different Morphology with Ultrasound and CT Scan

Observational Study Comparing the Measurement of Abdominal Aortic Aneurysm (AAA) Diameter in Different Morphology with Ultrasound and CT Scan

4:51 PM Abstract No. 235 - Clinical validation of a semi-automated software for maximal diameter measurements and follow-up of complex AAA geometries

To compare automated measurement of abdominal aortic aneurysm (AAA) maximal diameter (D-max) orthogonal to luminal or thrombus centerline. The local ethic comity approved this retrospective study of 83 patients (74 men, 9 women; mean age, 75 years, range, 46-91 years) with AAA ≥ 35 mm followed by multidetector computer tomography (MDCT) with one scan pre-stenting and at least two scans three months post-EVAR. 3-D reconstruction of the AAA and its components (wall, lumen, thrombus and centerlines) was achieved with dedicated segmentation software. Performances of automatic calculation algorithms of D-max perpendicular to lumen or thrombus centerlines were then compared to manually measured D-max on double-oblique MPR projections. To assess agreement, mean difference, bland-altman plot, regression models and paired student’s t-test were performed at baseline, first follow-up (FU) scan after EVAR and on the progression interval (last FU scan versus baseline). There was no significant difference in D-max measurements between thrombus algorithm and manual method (at baseline, mean error: -0.07 ± 1.66 mm, p-value= 0.7; at FU, 0.24 ± 1.69 mm, p-value= 0.2) compared to the luminal algorithm (at baseline, mean error: -1.24 ± 2.01 mm, p-value < 0.01; at FU, -1.49 ± 3.30 mm, p-value < 0.01). The difference in D-max progression between manual and thrombus algorithm was not significant (p-value = 0.1). AAA thrombus segmentation is necessary to enable automated calculation of D-max and follow D-max of AAAs with complex geometries.

The role of diameter versus volume as the best prognostic measurement of abdominal aortic aneurysms

Accurate measurement of abdominal aortic aneurysms is necessary to predict rupture risk and, more recently, to follow aneurysm sac behavior following endovascular repair. Up until this point, aneurysm diameter has been the most common measurement utilized for these purposes. Although aneurysm diameter is predictive of rupture, accurate measurement is hindered by such factors as aortic tortuosity and interobserver variability, and it does not account for variations in morphology such as saccular aneurysms. Additionally, decreases in aneurysm diameter do not completely describe the somewhat complex remodeling seen following endovascular repair of aortic aneurysms. Measurement of aneurysm volume has the advantage of describing aneurysm morphology in a multidimensional fashion, but it has not been readily available or easily measured until recently. This has changed with the introduction of commercially available software tools that permit quicker and easier to perform volume measurements. Whether it is time for volume to replace, or compliment, diameter is the subject of the current debate.

Editors' commentary

The abdominal aortic aneurysm (AAA) diameter is a key component in the surveillance of AAAs for the assessment of aneurysm progression. AAA external diameter has been shown to be a reliable method for repeated measurements in cases near the threshold for surgical referral. Other measures such as volume or wall stress have had, until now, a limited value in clinical practice. This argument was well developed by Mastracci et al in this debate. But with reported mean annual growth rates of 2 to 3 mm in diameter, a high reproducibility is required to allow detection of small changes in AAA diameter. In a systematic review of ultrasound measurement of the abdominal aorta diameter, Beales et al1Beales L. Wostenhulme S. Evans J.A. West R. Scott D.J.A. Reproducibility of ultrasound measurement of the abdominal aorta.Br J Surg. 2011; 98: 1517-1525Crossref PubMed Scopus (87) Google Scholar reported intraobserver and interobserver values greater than the 5-mm level regarded as acceptable by the U.K. and U.S. screening programs. These differences may have had a significant clinical impact on screening and surveillance. In addition, even though ultrasound diameter imaging has been used for years, no standardized image acquisition exists. This limitation has been emphasized by Bredahl et al,2Bredahl K. Eldrup N. Meyer C. Eiberg J.E. Sillesen H. Reproducibility of ECG-gated ultrasound diameter assessment of small abdominal aortic aneurysms.Eur J Vasc Endovasc Surg. 2013; 45: 235-240Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar who showed the importance of a standardized protocol including electrocardiograph-gating and subsequent off-line reading with minute caliper placement to reduce variability. Grondal et al have also shown that measurement of the maximum external AAA diameter by ultrasound is influenced by the pulse wave propagation, with an average difference of 1.9 mm between diastole and systole and a wide range in variation (0-4.7 mm).3Grondal N. Bramsen M.B. Thomsen M.D. Rasmussen C.B. Lindholt J.S. The cardiac cycle is a major contributor to variability in size measurements of abdominal aortic aneurysms by ultrasound.Eur J Vasc Endovasc Surg. 2012; 43: 30-33Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar This explains why ultrasound has been supplanted by computed tomography angiography with the use of center-lumen of flow by postprocessing software to estimate the AAA diameter with greater accuracy. As discussed by our debaters, assessment of the AAA volume is another parameter beyond diameter. It allows measurement of contour changes of the AAA and intraluminal thrombus volume. Using segmentation software, it permits accurate measurements of the AAA volume even using non-contrast-enhanced computed tomography scans. Volumetric measurements also have a higher sensitivity for AAA growth than diameter measurements. In addition, three-dimensional ultrasound permits quantification of the intraluminal thrombus without any risk of contrast agent or radiation. As shown by van Keulen,4van Keulen J.W. van Prehn J. Prokop M. Moll F.L. van Herwaarden J.A. Potential value of aneurysm sac volume measurements in addition to diameter measurements after endovascular aneurysm repair.J Endovasc Ther. 2009; 16: 506-513Crossref PubMed Scopus (47) Google Scholar aortic volume measurement may be particularly useful for surveillance after endovascular aneurysm repair. In his study, sac expansion was detected by volumetry in 32 patients, although an increase in sac diameter was seen only in 14 of them. Despite ample evidence,5Prinssen M. Verhoeven E.L. Verhagen H.J. Blankensteijn J.D. Decision-making in follow-up after endovascular aneurysm repair based on diameter and volume measurements: a blinded comparison.Eur J Vasc Endovasc Surg. 2003; 26: 184-187Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar, 6Wever J.J. Blankensteijn J.D. Th M Mali W.P. Eikelboom B.C. Maximal aneurysm diameter follow-up is inadequate after endovascular abdominal aortic aneurysm repair.Eur J Vasc Endovasc Surg. 2000; 20: 177-182Abstract Full Text PDF PubMed Scopus (151) Google Scholar volume assessment is still not carried out in many institutions. The reasons are many: Volume assessment is time consuming and requires dedicated software and skilled technicians and may be difficult to organize in high-volume centers. Furthermore, observer variability still exists in multiplanar reconstructions. Finally, the ability of aortic volume to predict rupture has not been established.7Parr A. Jayaratne C. Buttner P. Golledge J. Comparison of volume and diameter measurement in assessing small abdominal aortic aneurysm expansion examined using computed tomographic angiography.Eur J Radiol. 2011; 79: 42-47Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar In conclusion, even if volumetric measurements are likely to be of value in assessing the efficacy of new therapies for small AAA, we need more evidence to revise our guidelines, based until now on diameter thresholds. The role of diameter versus volume as the best prognostic measurement of abdominal aortic aneurysmsJournal of Vascular SurgeryVol. 58Issue 1PreviewAccurate measurement of abdominal aortic aneurysms is necessary to predict rupture risk and, more recently, to follow aneurysm sac behavior following endovascular repair. Up until this point, aneurysm diameter has been the most common measurement utilized for these purposes. Although aneurysm diameter is predictive of rupture, accurate measurement is hindered by such factors as aortic tortuosity and interobserver variability, and it does not account for variations in morphology such as saccular aneurysms. Full-Text PDF Open Archive

WITHDRAWN: The impact of using inner-to-inner wall diameter measurement of abdominal aortic aneurysm

Stephen Wolstenhulme. The impact of using inner-to-inner wall diameter measurement of abdominal aortic aneurysm. Ultrasound. 2016; 21 (2): 105–105. Article withdrawn by publisher. This article was accidentally published twice in the online version of Volume 21 Issue 2, with different DOIs but the same page numbers. There was no duplication of the article in the printed version of Volume 21 Issue 2. The incorrect version of the article with DOI: 10.1177/1742271X1301200211 has been replaced with this correction notice. The correct and citable version of the article remains: Stephen Wolstenhulme. The Impact of using Inner-to-Inner Wall Diameter Measurement of Abdominal Aortic Aneurysm. Ultrasound. 2017; 21 (2): 105–105. 10.1177/1742271X1302100201

Strain Measurement of Abdominal Aortic Aneurysm with Real-time 3D Ultrasound Speckle Tracking

Abdominal aortic aneurysm rupture is caused by mechanical vascular tissue failure. Although mechanical properties within the aneurysm vary, currently available ultrasound methods assess only one cross-sectional segment of the aorta. This study aims to establish real-time 3-dimensional (3D) speckle tracking ultrasound to explore local displacement and strain parameters of the whole abdominal aortic aneurysm. Validation was performed on a silicone aneurysm model, perfused in a pulsatile artificial circulatory system. Wall motion of the silicone model was measured simultaneously with a commercial real-time 3D speckle tracking ultrasound system and either with laser-scan micrometry or with video photogrammetry. After validation, 3D ultrasound data were collected from abdominal aortic aneurysms of five patients and displacement and strain parameters were analysed. Displacement parameters measured in vitro by 3D ultrasound and laser scan micrometer or video analysis were significantly correlated at pulse pressures between 40 and 80 mmHg. Strong local differences in displacement and strain were identified within the aortic aneurysms of patients. Local wall strain of the whole abdominal aortic aneurysm can be analysed in vivo with real-time 3D ultrasound speckle tracking imaging, offering the prospect of individual non-invasive rupture risk analysis of abdominal aortic aneurysms.

Commentary on ‘Strain Measurement of Abdominal Aortic Aneurysm with Real-time 3D Ultrasound Speckle Tracking’

Commentary on ‘Strain Measurement of Abdominal Aortic Aneurysm with Real-time 3D Ultrasound Speckle Tracking’

WITHDRAWN: The Impact of Using Inner-to-Inner Wall Diameter Measurement of Abdominal Aortic Aneurysm

Y Sensier, Y Akil, P Khodabakhsh, M Naylor, R Sharpe, J Walker, T Hartshorne. The Impact of Using Inner-to-Inner Wall Diameter Measurement of Abdominal Aortic Aneurysm. Ultrasound. 2013; 21 (3): 8–11. Article withdrawn by publisher. This article was accidentally published twice in the online version of Volume 21 Issue 1, with different DOIs but the same page numbers. There was no duplication of the article in the printed version of Volume 21 Issue 1. The incorrect version of the article with DOI: 10.1177/1742271X1302100102 has been replaced with this correction notice. The correct and citable version of the article remains: Y Sensier, Y Akil, P Khodabakhsh, M Naylor, R Sharpe, J Walker, T Hartshorne. The Impact of using Inner-to-Inner Wall Diameter Measurement of Abdominal Aortic Aneurysm. Ultrasound. 2013; 21 (1): 8–11. 10.1177/1742271X12472849

Ultrasound Measurement of Aortic Diameter in a National Screening Programme

Objective Currently there is no universally accepted standard for ultrasound measurement of abdominal aortic aneurysm (AAA). The aim was to investigate the reliability and reproducibility of inner to inner (ITI) versus outer to outer (OTO) ultrasound measurement of AAA diameter. Methods A prospective study design was used to collect 60 random images of aorta (1.4–7.1 cm). Inner and outer wall diameter measurements were then performed by 13 qualified AAA screening technicians and 11 vascular sonographers. Results The mean (range) diameter for all 60 aortas by ITI was 3.91 cm (1.39–6.80) and by OTO was 4.18 cm (1.63–7.09), a significant mean difference of 0.27 cm (95% CI:0.23–0.32 cm). The reproducibility coefficients for differences between technicians were 0.30 cm (95% CI:0.24–0.36) for ITI and 0.42 cm (95% CI:0.35–0.49) for OTO indicating significantly better repeatability using ITI. Finally, 15 images were measured twice in random order by all screeners and sonographers. For AAAs >5 cm, repeatability was significantly better with ITI than OTO (0.14 vs. 0.21; p = 0.016). Conclusion There was the expected difference in AAA diameter between the two methods (0.27 cm). However, ITI wall method was measurably more reproducible.

Rapid Assessment of Abdominal Aortic Aneurysms by 3-Dimensional Ultrasonography

The purpose of this study was to assess the application and accuracy of 3-dimensional (3D) volume acquisition ultrasonography in the measurement of abdominal aortic aneurysms (AAAs). Thirty consecutive patients undergoing surveillance ultrasonographic examinations of known AAAs were scanned according to a conventional protocol, after which 3D volume data sets were acquired with a commercially available mechanical transducer. Maximum aortic diameters were measured with multiplanar reconstructions from the 3D volume data, and these were compared with those from the conventional technique. Abdominal aortic aneurysm diameters were between 2.9 and 6.75 cm. For both anteroposterior and transverse diameters, a paired t test showed a Pearson correlation coefficient of 0.98 (significant at the P = .01 level) and a coefficient of determination of 0.96. Bland-Altman analysis showed that the mean difference between the two sets of measurements was very close to 0 (P = .05). Thus, there was no significant difference between the conventional and 3D volume measurement methods. The scan acquisition time for the 3D volume data was only 3 seconds for each set (anteroposterior and transverse). Three-dimensional ultrasonography using volume acquisition offers a new opportunity to acquire fast and reliable AAA measurements. The reduced scan times can be used to allow greater patient throughput and will help cope with the increasing workload of AAA surveillance. By archiving a complete set of data, 3D ultrasonography allows subsequent analysis and comparison of measurements. This study also suggests that the technique could be used for other applications with similar efficiency gains.

Accuracy of ultrasonographic measurement of infrarenal abdominal aortic aneurysm.

Abstract The ultrasonographic measurement of abdominal aortic aneurysm (AAA) must be accurate as the decision on elective repair will usually be based on such results. To detect expansion, measurements must be comparable. The interobserver variation of ultrasonographic measurement of infrarenal aortic aneurysms under standardized conditions is described.

Noninvasive measurement of abdominal aortic aneurysms in intact mice by a high-frequency ultrasound imaging system

Mouse models of abdominal aortic aneurysm (AAA) have been commonly used in many laboratories for studying molecular mechanisms of AAA formation and development, as well as for testing novel therapeutic agents in the treatment of AAA. However, because of the small size of the animal, the quantification and characterization of AAA development and progress is difficult, time-consuming and requires the sacrifice of the experimental animals. We report here a noninvasive method to detect and measure AAA in mice using a high-frequency ultrasound (US) imaging system specifically designed for microimaging of the mice (Vevo 660; VisualSonics, Toronto, ONT, Canada). A total of 21 male apolipoprotein-E-deficient mice were chronically infused with angiotensin II (1.44 mg/kg daily) for 28 days to induce AAA formation. A 2-D echo image of the abdominal aorta was acquired at longitudinal and transverse planes, followed immediately by post mortem dissection of the abdominal aorta for direct measurements. The US images clearly showed a bulge-like expansion localized specifically in the suprarenal region of the abdominal aorta, with a shape strikingly similar to that of the aorta dissected post mortem. In addition, the US images can also provide measurements of the luminal diameter and wall thickness of the abdominal aorta. The average dimensions of the abdominal aorta were not significantly different between the US and post mortem measurements, nor between the transverse and longitudinal US images. The different types of the measurements are also highly correlated with each other, with a linear correlation (r) between 0.7 and 0.9. Thus, we have established and validated a novel application to noninvasively measure AAA development and progress in a mouse model using a high-frequency US imaging system that has the advantages of low cost, rapid imaging speed, reproducibility and high resolution, and makes repeated monitoring of the progress of AAA development over a time-course possible. (E-mail: jim_wang@berlex.com)

Aneurysmal sizing after endovascular repair in patients with abdominal aortic aneurysm: interobserver variability of various measurement protocols and its clinical relevance.

The aim of this study was to determine the variability of various measurement protocols for measurement of abdominal aortic aneurysm (AAA) and the clinical relevance of variability. Three radiologists performed computed tomographic angiography measurements of both the aorta and the largest portion of the aneurysm on selected axial slices. Then measurements of the largest portion of the aneurysm were performed on unselected axial slices, sagittal and coronal reformatted. Finally, aortic volume was calculated. Measurements and volume calculation were performed before and after endovascular repair and assessed: Part 1: interobserver variability for maximum anteroposterior (MAP) and maximum transverse (MTR) diameters on selected slices; part 2: interobserver variability for unselected slices considering MAP and MTR; part 3: interobserver variability considering maximum diameter in any direction (MAD); part 4: interobserver variability for sagittal (SAG) and coronal (COR) free curved multiplanar reformation (MPR); and part 5: volume calculations. We then determined which technique of measurement was the most clinically relevant for detecting changes in aneurysm size or aortic volume. Parts 1 and 2: interobserver variability was 4.1 mm for both MAP and MTR; part 3: interobserver variability was 7 mm for MAD; part 4: interobserver variability was 5.5 mm (COR) and 4.9 mm (SAG); part 5: interobserver variability for volume was 5.5 ml. A combination of MAP and MTR was the most useful for detecting aortic modification. Volume calculation was needed in only a few cases. We recommend avoiding MAD and MPR measurements and suggest instead measuring both maximum anteroposterior and maximum transverse diameters. If aneurysm size remains stable after endovascular repair, aneurysm volume should be measured.

Measurement of abdominal aortic aneurysms with three-dimensional ultrasound imaging: Preliminary report

Objective: Accurate measurements of abdominal aortic aneurysms (AAAs) are required for surgical planning and monitoring over time. We have examined the feasibility of using a three-dimensional (3-D) ultrasound imaging system to derive quantitative measurements of interest from AAAs. Methods: A normal aorta, a small AAA, and an AAA repaired with an endovascular stent graft were scanned with a 3-D ultrasound imaging system. For each case, a 3-D surface reconstruction was generated from manual outlines of a sequence of two-dimensional ultrasound images, registered in 3-D space with a magnetic tracking system. The surfaces were resampled in planes perpendicular to the vessel center axis to calculate cross-sectional area and maximum diameter as a function of distance along the length of the aorta. Results: Cross-sectional area and maximum diameter were plotted along the length of the aneurysmal aortas from the renal arteries to the aortic bifurcation. The overall maximum diameter was found for both aneurysms. For the small AAA, the distances of the aneurysm from the renal arteries and the bifurcation were measured. For the repaired AAA, the location of the stent graft relative to the renal arteries was measured. Conclusions: 3-D surface reconstructions from ultrasound images show promise for quantitatively characterizing the geometry of AAAs both before surgery and after endovascular repair. (J Vasc Surg 2001;33:700-7.)