Measuring the aorta in the era of multimodality imaging: still to be agreed

Díaz-Peláez, Elena; Barreiro-Pérez, Manuel; Martín-García, Ana; Sánchez, Pedro L.

doi:10.21037/jtd.2017.03.96

Cited by 10 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Fourth, in our dataset, ascending aortic size tended to be more stable over time than segments of the aortic arch and descending thoracic aorta, which enabled the exclusion of fewer scans. Ground truth segmentations considered the vessel wall as part of the aorta, as is standard in patients with aneurysmal disease [23][24][25], and this was successfully reproduced by the deep learning model and reflected in all measurements (Fig. 2).…”

Section: Discussionmentioning

confidence: 99%

Deep Learning Improves the Temporal Reproducibility of Aortic Measurement

et al. 2021

View full text Add to dashboard Cite

Imaging-based measurements form the basis of surgical decision making in patients with aortic aneurysm. Unfortunately, manual measurement suffer from suboptimal temporal reproducibility, which can lead to delayed or unnecessary intervention. We tested the hypothesis that deep learning could improve upon the temporal reproducibility of CT angiography-derived thoracic aortic measurements in the setting of imperfect ground-truth training data. To this end, we trained a standard deep learning segmentation model from which measurements of aortic volume and diameter could be extracted. First, three blinded cardiothoracic radiologists visually confirmed non-inferiority of deep learning segmentation maps with respect to manual segmentation on a 50-patient hold-out test cohort, demonstrating a slight preference for the deep learning method (p < 1e-5). Next, reproducibility was assessed by evaluating measured change (coefficient of reproducibility and standard deviation) in volume and diameter values extracted from segmentation maps in patients for whom multiple scans were available and whose aortas had been deemed stable over time by visual assessment (n = 57 patients, 206 scans). Deep learning temporal reproducibility was superior for measures of both volume (p < 0.008) and diameter (p < 1e-5) and reproducibility metrics compared favorably with previously reported values of manual inter-rater variability. Our work motivates future efforts to apply deep learning to aortic evaluation.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep Learning Improves the Temporal Reproducibility of Aortic Measurement

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The inner edge-to-inner edge technique also correlates better with advanced imaging modalities such as computed tomography and cardiac magnetic resonance imaging, which also use the inner-edge technique to measure the aortic diameter. [ 11 ] Advances in echocardiography have also allowed for improved resolution with the ability to localize the blood-tissue interface. [ 12 ] However, this produces a limitation in this study in comparing the adult and pediatric guideline as the adult guideline currently calls for a leading edge-to-leading edge technique.…”

Section: Discussionmentioning

confidence: 99%

Aortic size in children

Al-mousily

Lopez

Muniz

et al. 2021

Annals of Pediatric Cardiology

View full text Add to dashboard Cite

Background: Current guidelines recommended aortic measurements during diastole in adults and during systole in children. Recent studies in adults have demonstrated noteworthy differences in aortic measurements during systole and diastole in the same subjects. In the present study, we aimed to characterize systolic and diastolic differences in aortic measurements in healthy children. Materials and Methods : This retrospective study included 272 children who had a complete echocardiogram and no heart disease. Aortic measurements at the annulus (ANN), aortic root (AOR), sinotubular junction (STJ), and ascending aorta (AAO) were performed. Systolic and diastolic values were compared by calculating the mean systolic to diastolic (SD) percent difference for each segment; if the SD difference was >5%, it was considered clinically important. Similar measurements were conducted by another observer in 18% of the subjects. Results: Systolic measurements were larger than diastolic measurements with mean SD percent differences >5% ( P < 0.001) for the AOR (7.3% ± 5.5%), STJ (10.24% ± 7.1%), and AAO (9.8% ± 7.4%). There was no clinically significant SD difference for the ANN. There was an excellent intraclass correlation coefficient between observers (0.982–0.995). Conclusions: Systolic measurements for the AOR, STJ, and AAO were larger than diastolic measurements. Normal reference values are utilized to design treatment for patients with abnormal aortic sizes, and the timing in the cardiovascular cycle used to decide the reference values should be equivalent to the timing used to make measurements in clinical practice. This is particularly imperative as patients transition their care from a pediatric to an adult cardiologist.

show abstract

“…After CTA acquisition, aortic diameters are measured in orthogonal views obtained through a reformation process (Kauffmann et al (2011);Nienaber et al (2016); Díaz-Peláez et al (2017)). Examples of reformation techniques are the manual, doubleoblique reformation (Kauffmann et al (2011)) and the curved multi-planar reformation (Kanitsar et al (2002); Hahn et al (2020)).…”

Section: Clinical Surveillancementioning

confidence: 99%

“…An accurate determination of aortic caliber requires that diameter measurements are obtained on cross-sections oriented orthogonal to the aortic flow channel (Erbel et al (2014); Lombardi et al (2020); Pepe et al (2020)). In clinical practice, the selection and orientation of these cross-sectional views, or planes, is defined manually by highly trained radiologists and radiology technicians, using free-hand interactive double-oblique reformations (Díaz-Peláez et al (2017); Bhave et al (2018)) or semiautomatic techniques that track the vessel centerline (Müller-Eschner et al (2013); Gamechi et al (2019)). Figure 1 and Figure 2 provide examples of these methodologies.…”

Section: Introductionmentioning

confidence: 99%

Automated cross-sectional view selection in CT angiography of aortic dissections with uncertainty awareness and retrospective clinical annotations

Pepe¹,

Egger²,

Codari³

et al. 2021

Preprint

View full text Add to dashboard Cite

Objective: Surveillance imaging of chronic aortic diseases, such as dissections, relies on obtaining and comparing cross-sectional diameter measurements at predefined aortic landmarks, over time. Due to a lack of robust tools, the orientation of the crosssectional planes is defined manually by highly trained operators. We show how manual annotations routinely collected in a clinic can be efficiently used to ease this task, despite the presence of a non-negligible interoperator variability in the measurements. Impact: Ill-posed but repetitive imaging tasks can be eased or automated by leveraging imperfect, retrospective clinical annotations. Methodology: In this work, we combine convolutional neural networks and uncertainty quantification methods to predict the orientation of such cross-sectional planes. We use clinical data randomly processed by 11 operators for training, and test on a smaller set processed by 3 independent operators to assess interoperator variability. Results: Our analysis shows that manual selection of cross-sectional planes is characterized by 95% limits of agreement (LOA) of 10.6 • and 21.4 • per angle. Our method showed to decrease static error by 3.57 • (40.2%) and 4.11 • (32.8%) against state of the art and LOA by 5.4 • (49.0%) and 16.0 • (74.6%) against manual processing. Conclusion: This suggests that pre-existing annotations can be an inexpensive resource in clinics to ease ill-posed and repetitive tasks like cross-section extraction for surveillance of aortic dissections.

show abstract

Measuring the aorta in the era of multimodality imaging: still to be agreed

Cited by 10 publications

References 11 publications

Deep Learning Improves the Temporal Reproducibility of Aortic Measurement

Deep Learning Improves the Temporal Reproducibility of Aortic Measurement

Aortic size in children

Automated cross-sectional view selection in CT angiography of aortic dissections with uncertainty awareness and retrospective clinical annotations

Contact Info

Product

Resources

About