DeepAAA: Clinically Applicable and Generalizable Detection of Abdominal Aortic Aneurysm Using Deep Learning

Lu, Jen-Tang; Brooks, Rupert; Hahn, Stefan; Chen, Jin; Buch, Varun; Kotecha, Gopal; Andriole, Katherine P.; Ghoshhajra, Brian B.; Pinto, Joel; Vozila, Paul; Michalski, Mark; Tenenholtz, Neil A.

doi:10.1007/978-3-030-32245-8_80

Cited by 22 publications

(27 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gray value variation and translation have shown to successfully overcomes the influence of overfitting. Furthermore, Lu et al [15] propose DeepAAA, a modified 3D U-Net architecture combined with ellipse fitting. Wang et al [16] propose a novel network, based on U-Net architecture, that fuses the high-level part of the MR and CT images together, allowing successful multi-modal segmentation of the AAA.…”

Section: A Related Workmentioning

confidence: 99%

See 1 more Smart Citation

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

Section: A Related Workmentioning

confidence: 99%

“…Training is done using the Adam optimizer with an initial learning rate lr init = 5 * 10 4 , and the learning rate schedule: l2 weight decay of 10 5 and lr init = 0.985 epoch. For loss function, we employ a smoothed negative dice score [15], defined with :…”

Section: B Modified 3d U-net With Deep Supervisionmentioning

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

show abstract

“…We compared the regression performance of the DeepAAA [ 35 ] and DetectNet [ 34 ] methods using our dataset. Table 3 shows that our proposed loss function outperformed the DeepAAA method using the smooth negative Dice coefficient and the DetectNet method using the

loss in terms of recall.…”

Section: Resultsmentioning

confidence: 99%

“…DetectNet is only used for the detection of AAA thrombi, while segmentation is performed using a modified holistically nested edge detection (HED) network. Even more relevant to this paper, Lu et al [ 35 ] presented a three-dimensional algorithm for AAA segmentation for the first time. The detection and segmentation are performed by applying ellipse fitting that is based on a variant of the 3D UNet architecture.…”

Section: Related Workmentioning

confidence: 99%

Automatic Detection and Segmentation of Thrombi in Abdominal Aortic Aneurysms Using a Mask Region-Based Convolutional Neural Network with Optimized Loss Functions

Hwang

Kim

Lee

et al. 2022

Sensors

View full text Add to dashboard Cite

The detection and segmentation of thrombi are essential for monitoring the disease progression of abdominal aortic aneurysms (AAAs) and for patient care and management. As they have inherent capabilities to learn complex features, deep convolutional neural networks (CNNs) have been recently introduced to improve thrombus detection and segmentation. However, investigations into the use of CNN methods is in the early stages and most of the existing methods are heavily concerned with the segmentation of thrombi, which only works after they have been detected. In this work, we propose a fully automated method for the whole process of the detection and segmentation of thrombi, which is based on a well-established mask region-based convolutional neural network (Mask R-CNN) framework that we improve with optimized loss functions. The combined use of complete intersection over union (CIoU) and smooth L1 loss was designed for accurate thrombus detection and then thrombus segmentation was improved with a modified focal loss. We evaluated our method against 60 clinically approved patient studies (i.e., computed tomography angiography (CTA) image volume data) by conducting 4-fold cross-validation. The results of comparisons to multiple other state-of-the-art methods suggested the superior performance of our method, which achieved the highest F1 score for thrombus detection (0.9197) and outperformed most metrics for thrombus segmentation.

show abstract

“…There are limited studies describing tools for automatic aortic segmentation/measurements that, for example, detect abdominal aortic aneurysms, measure the descending aortic diameter prior to stent graft planning, segment and measure aortic diameters in native scans of the thoracic aorta in CT scans, or help to improve reading follow-up CT scans according to guidelines (16)(17)(18)(19).…”

Section: Introductionmentioning

confidence: 99%

Fully automated guideline-compliant diameter measurements of the thoracic aorta on ECG-gated CT angiography using deep learning

Pradella¹,

Weikert²,

Sperl³

et al. 2021

Quant Imaging Med Surg

View full text Add to dashboard Cite

Background: Manually performed diameter measurements on ECG-gated CT-angiography (CTA) represent the gold standard for diagnosis of thoracic aortic dilatation. However, they are time-consuming and show high inter-reader variability. Therefore, we aimed to evaluate the accuracy of measurements of a deep learning-(DL)-algorithm in comparison to those of radiologists and evaluated measurement times (MT). Methods:We retrospectively analyzed 405 ECG-gated CTA exams of 371 consecutive patients with suspected aortic dilatation between May 2010 and June 2019. The DL-algorithm prototype detected aortic landmarks (deep reinforcement learning) and segmented the lumen of the thoracic aorta (multilayer convolutional neural network). It performed measurements according to AHA-guidelines and created visual outputs. Manual measurements were performed by radiologists using centerline technique. Human performance variability (HPV), MT and DL-performance were analyzed in a research setting using a linear mixed model based on 21 randomly selected, repeatedly measured cases. DL-algorithm results were then evaluated in a clinical setting using matched differences. If the differences were within 5 mm for all locations, the cases was regarded as coherent; if there was a discrepancy >5 mm at least at one location (incl. missing values), the case was completely reviewed.Results: HPV ranged up to ±3.4 mm in repeated measurements under research conditions. In the clinical setting, 2778/3192 (87.0%) of DL-algorithm's measurements were coherent. Mean differences of paired measurements between DL-algorithm and radiologists at aortic sinus and ascending aorta were −0.45±5.52 and −0.02±3.36 mm. Detailed analysis revealed that measurements at the aortic root were over-/ underestimated due to a tilted measurement plane. In total, calculated time saved by DL-algorithm was 3:10 minutes/case.

show abstract

DeepAAA: Clinically Applicable and Generalizable Detection of Abdominal Aortic Aneurysm Using Deep Learning

Cited by 22 publications

References 11 publications

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

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

Automatic Detection and Segmentation of Thrombi in Abdominal Aortic Aneurysms Using a Mask Region-Based Convolutional Neural Network with Optimized Loss Functions

Fully automated guideline-compliant diameter measurements of the thoracic aorta on ECG-gated CT angiography using deep learning

Contact Info

Product

Resources

About