Abdominal Aortic Aneurysm Segmentation with a Small Number of Training Subjects

Zheng, Jian-Qing; Zhou, Xiao-Yun; Li, Qingbiao; Riga, Celia; Yang, Guang‐Zhong

doi:10.31256/hsmr2018.23

Cited by 9 publications

(6 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study describes a fully automatic and high-resolution algorithm able to extract the aortic volume from both CTA and non-contrast CT images at a level superior to that of other currently published methods 12,13 . High accuracy of our segmentation pipeline was supported by the DICE score metric between model predictions and ground truth annotations for both the thoracic and abdominal aorta.…”

Section: Discussionmentioning

confidence: 99%

“…Recently, DL methods on CTAs have been proposed to tackle this problem without encountering many of the limitations of their predecessors. Variations on Deep Belief and Unet based networks have been used to segment the infra-renal region of the aorta 13,22 . Unfortunately, many of these networks are limited to 2-D inputs (axial CT slices), which may fail to appropriately capture the aneurysm's 3D geometry.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

A Deep Learning Pipeline to Automate High-Resolution Arterial Segmentation With or Without Intravenous Contrast

et al. 2020

View full text Add to dashboard Cite

Pathological changes in the vasculature (ex. abdominal aortic aneurysms) prevent automatic vessel reconstruction and extraction of vital morphological/pathological information. In this study, a modified U-Net with attention-gating was implemented to establish a highthroughput and automated segmentation pipeline of pathological blood vessels in CT images acquired with/without contrast agent injection. The performance of our network for aneurysmal and entire aortic volume segmentation from both contrast and non-contrast CT images displayed strong clinical accuracy. This method can be used to simplify, expedite and standardize aneurysmal disease assessment, sets the foundation for complex geometric analysis and can be extended to other vascular pathologies.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

A Deep Learning Pipeline to Automate High-Resolution Arterial Segmentation With or Without Intravenous Contrast

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Recently, a variety of machine/ deep-learning methods on contrast-enhanced CTAs have been proposed to tackle this problem without encountering many of the limitations of their predecessors. Variations on Deep Belief and U-net based networks have been used to segment the infra-renal region of the aorta [16,17]. Unfortunately, many of these networks are limited to 2-D inputs (axial CT slices), which may fail to appropriately capture the 3D geometry of the aneurysm.…”

Section: Discussionmentioning

confidence: 99%

A deep learning approach to automate high-resolution blood vessel reconstruction on computerised tomography images with or without the use of contrast agents

Chandrashekar

Handa

Shivakumar

et al. 2020

European Heart Journal

View full text Add to dashboard Cite

Background Existing methods to reconstruct vascular structures from a computed tomography (CT) angiogram rely on injection of intravenous contrast to enhance the radio-density within the vessel lumen. Pathological changes present within the blood lumen, vessel wall or a combination of both prevent accurate 3D reconstruction. In the example of aortic aneurysmal (AAA) disease, a blood clot or thrombus adherent to the aortic wall within the expanding aneurysmal sac is present in 95% of cases. These deformations prevent the automatic extraction of vital clinically relevant information by current methods. Objectives In this study, we utilised deep learning segmentation methods to establish a high-throughput and automated segmentation pipeline for pathological blood vessels (ex. Aortic Aneurysm) in CT images acquired with or without the use of a contrast agent. Methods Twenty-six patients with paired non-contrast and contrast-enhanced CT images were randomly selected from an ethically-approved ongoing study (Ethics Ref 13/SC/0250), manually annotated and used for model training and evaluation (13/13). Data augmentation methods were implemented to diversify the training data set in a ratio of 10:1. We utilised a 3D U-Net with attention gating for both the aortic region-of-interest (ROI) detection and segmentation tasks. Trained architectures were evaluated using the DICE similarity score. Results Inter- and Intra- observer analysis supports the accuracy of the manual segmentations used for model training (intra-class correlation coefficient, “ICC” = 0.995 and 1.00, respective. P<0.001 for both). The performance of our Attention-based U-Net (DICE score: 94.8±0.5%) in extracting both the inner lumen and the outer wall of the aortic aneurysm from CT angiograms (CTA) was compared against a generic 3-D U-Net (DICE score: 89.5±0.6%) and displayed superior results (p<0.01). Fig 1A depicts the implementation of this network architecture within the aortic segmentation pipeline (automated ROI detection and aortic segmentation). This pipeline has allowed accurate and efficient extraction of the entire aortic volume from both contrast-enhanced CTA (DICE score: 95.3±0.6%) and non-contrast CT (DICE score: 93.2±0.7%) images. Fig 1B illustrates the model output alongside the labelled ground truth segmentation for the pathological aneurysmal region; only minor differences are visually discernible (coloured boxes). Conclusion We developed a novel automated pipeline for high resolution reconstruction of blood vessels using deep learning approaches. This pipeline enables automatic extraction of morphologic features of blood vessels and can be applied for research and potentially for clinical use. Automated Segmentation of Blood Vessels Funding Acknowledgement Type of funding source: Foundation. Main funding source(s): University of Oxford Medical Research Fund, John Fell Fund

show abstract

“…Lopez-Linares et al [13] propose a pipeline that includes adapted DetectNet for AAA region of interest (ROI) detection, followed by a novel segmentation network based on holistically-nested edge detection and fully convolutional network (FCN). Zheng et al [14] investigated the impact of using an extremely small number of datasets for AAA segmentation using U-Net. They trained and evaluated network performance on just two CT datasets, obtaining successful results by employing strong data augmentation.…”

Section: A Related Workmentioning

confidence: 99%

Abdominal Aortic Aneurysm Segmentation from CT Images using Modified 3D U-Net with Deep Supervision

Habijan

Galić

Leventić

et al. 2020

2020 International Symposium ELMAR

View full text Add to dashboard Cite

An abdominal aortic aneurysm (AAA) is a dangerous cardiovascular disease that can cause serious health complications and death. Methods that can provide automatic and accurate segmentation of the AAA can significantly help in preoperative planning and postoperative follow-ups. Therefore, in this work, we present an automatic method for AAA segmentation from CT images using a modified 3D U-Net network with deep supervision. We compare obtained results for AAA segmentation using original 3D U-Net, and modified 3D U-Net with deep supervision. The trained network is evaluated on 19 volumetric CT images from the publicly available dataset provided by the University Hospitals Leuven, Belgium, using four-fold cross-validation. We obtained DSC of 91.03% using modified 3D U-Net with deep supervision. Additionally, we provide a discussion of the effects of using up-sampling versus deconvolution layers and its influence on the performance of both networks for this specific clinical application.

show abstract

Abdominal Aortic Aneurysm Segmentation with a Small Number of Training Subjects

Cited by 9 publications

References 4 publications

A Deep Learning Pipeline to Automate High-Resolution Arterial Segmentation With or Without Intravenous Contrast

A Deep Learning Pipeline to Automate High-Resolution Arterial Segmentation With or Without Intravenous Contrast

A deep learning approach to automate high-resolution blood vessel reconstruction on computerised tomography images with or without the use of contrast agents

Abdominal Aortic Aneurysm Segmentation from CT Images using Modified 3D U-Net with Deep Supervision

Contact Info

Product

Resources

About