Fully Automated Pipeline for Measurement of the Thoracic Aorta Using Joint Segmentation and Localization Neural Network
dc.contributor.author | Katakol, Sudeep | |
dc.contributor.author | Bian, Zhangxing | |
dc.contributor.author | Lu, Yanglong | |
dc.contributor.author | Spahlinger, Greg | |
dc.contributor.author | Hatt, Charles | |
dc.contributor.author | Burris, Nicholas | |
dc.date.accessioned | 2023-07-03T02:59:03Z | |
dc.date.available | 2023-07-03T02:59:03Z | |
dc.date.issued | 2023-07-02 | |
dc.identifier.uri | https://hdl.handle.net/2027.42/177150 | en |
dc.description.abstract | Purpose: Diagnosis and surveillance of thoracic aortic aneurysm (TAA) involves measuring the aortic diameter at various locations along the length of the aorta, often using computed tomography angiography (CTA). Currently, measurements are performed by human raters using specialized software for 3D analysis, a time-consuming process, requiring 15-45 minutes of focused effort. Thus we aimed to develop a convolutional neural network (CNN)-based algorithm for fully automated and accurate aortic measurements. Approach: Using 212 CTA scans, we trained a CNN to perform segmentation and localization of key landmarks jointly. Segmentation mask and landmarks are subsequently used to obtain the centerline and cross-sectional diameters of the aorta. Subsequently, a cubic spline is fit to the aortic boundary at the sinuses of Valsalva to avoid errors related inclusions of coronary artery origins. Performance was evaluated on test set of 57 scans, with automated measurements compared against expert manual raters. Result: Joint training of segmentation and landmark localization tasks yielded higher accuracy for both tasks compared to networks trained for each task individually. Mean absolute error between human and automated was ≤ 1 mm at 6 of 9 standard clinical measurement locations. However, higher errors were noted in the aortic root and arch regions, ranging between 1.7 and 2.1 mm, although agreement of manual raters was also lower in these regions. Conclusion: Fully-automated aortic diameter measurements in TAA are feasible using a CNN-based algorithm. Automated measurements demonstrated low errors that are comparable in magnitude to those with manual raters, however, measurement error were highest in the aortic root and arch. | en_US |
dc.language.iso | en_US | en_US |
dc.title | Fully Automated Pipeline for Measurement of the Thoracic Aorta Using Joint Segmentation and Localization Neural Network | en_US |
dc.type | Preprint | en_US |
dc.subject.hlbsecondlevel | Radiology | |
dc.subject.hlbtoplevel | Health Sciences | |
dc.contributor.affiliationumcampus | Ann Arbor | en_US |
dc.description.bitstreamurl | http://deepblue.lib.umich.edu/bitstream/2027.42/177150/1/JMI.TAA.Automated.Seg.LLoc.pdf | |
dc.identifier.doi | https://dx.doi.org/10.7302/7884 | |
dc.identifier.source | Greg Spahlinger | en_US |
dc.description.filedescription | Description of JMI.TAA.Automated.Seg.LLoc.pdf : Main Article. V1.7.2.23 | |
dc.description.depositor | SELF | en_US |
dc.working.doi | 10.7302/7884 | en_US |
dc.owningcollname | Radiology, Department of |
Files in this item
Remediation of Harmful Language
The University of Michigan Library aims to describe library materials in a way that respects the people and communities who create, use, and are represented in our collections. Report harmful or offensive language in catalog records, finding aids, or elsewhere in our collections anonymously through our metadata feedback form. More information at Remediation of Harmful Language.
Accessibility
If you are unable to use this file in its current format, please select the Contact Us link and we can modify it to make it more accessible to you.