Automatic segmentation of multiple cardiovascular structures from cardiac computed tomography angiography images using deep learning

Baskaran, Lohendran; Al’Aref, Subhi J.; Maliakal, Gabriel; Lee, Benjamin C.; Xu, Zhuoran; Choi, Jeong Won; Lee, Sang Eun; Sung, Ji Min; Lin, Fay Y.; Dunham, Simon; Mosadegh, Bobak; Kim, Yong Jin; Gottlieb, Ilan; Lee, Byoung Kwon; Chun, Eun Ju; Cademartiri, Filippo; Maffei, Erica; Marques, Hugo; Shin, Sanghoon; Choi, Jung Hyun; Chinnaiyan, Kavitha M.; Hadamitzky, Martin; Conte, Edoardo; Andreini, Daniele; Budoff, Matthew J.; Leipsic, Jonathon; Raff, Gilbert; Virmani, Renu; Samady, Habib; Stone, Peter H.; Berman, Daniel S.; Narula, Jagat; Bax, Jeroen J.; Chang, Hyuk Jae; Min, James K.; Shaw, Leslee J.

doi:10.1371/journal.pone.0232573

Cited by 32 publications

(21 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To the best of our knowledge, this is the first study that adopted ENet and ERFNet for the cardiovascular medical imaging analysis. These models are developed for real-time applications and are therefore smaller and faster than the UNet model used in other studies for cardiovascular segmentation [8,9,25,26]. The ENet model has an order of magnitude fewer parameters than both ERFNet and UNet while ERFNet has less than half the number of parameters compared to UNet.…”

Section: Discussionmentioning

confidence: 99%

“…Moreover, a five-fold cross-validation strategy using 2D slices from all patient cases as model input was used to overcome the limit of the small training dataset while the overfitting was reduced by six different types of data augmentation techniques. In other deep learning studies, the number of data for training was remarkable [6,8,9,26]. With regards to cardiovascular anatomies, Baskaran et al [9] applied a UNet-inspired deep learning model to segment cardiac structures and great vessels from 206 patients who underwent coronary CT angiography.…”

Section: Discussionmentioning

confidence: 99%

“…It is also important to remember that size is not the only important imaging marker; aortic shape matters as well as the loss of the normal "waist" of the aorta at the sinotubular junction. 1 3 Deep learning methods are emerging for vascular segmentation and remains a challenging area of research [6][7][8][9][10]. These techniques have shown tremendous success in the last 5 years for image classification and segmentation tasks in various fields, especially for neuroimaging for small vessel segmentation [11].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deep learning approach for the segmentation of aneurysmal ascending aorta

et al. 2020

View full text Add to dashboard Cite

Diagnosis of ascending thoracic aortic aneurysm (ATAA) is based on the measurement of the maximum aortic diameter, but size is not a good predictor of the risk of adverse events. There is growing interest in the development of novel imagederived risk strategies to improve patient risk management towards a highly individualized level. In this study, the feasibility and efficacy of deep learning for the automatic segmentation of ATAAs was investigated using UNet, ENet, and ERFNet techniques. Specifically, CT angiography done on 72 patients with ATAAs and different valve morphology (i.e., tricuspid aortic valve, TAV, and bicuspid aortic valve, BAV) were semi-automatically segmented with Mimics software (Materialize NV, Leuven, Belgium), and then used for training of the tested deep learning models. The segmentation performance in terms of accuracy and time inference were compared using several parameters. All deep learning models reported a dice score higher than 88%, suggesting a good agreement between predicted and manual ATAA segmentation. We found that the ENet and UNet are more accurate than ERFNet, with the ENet much faster than UNet. This study demonstrated that deep learning models can rapidly segment and quantify the 3D geometry of ATAAs with high accuracy, thereby facilitating the expansion into clinical workflow of personalized approach to the management of patients with ATAAs.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning approach for the segmentation of aneurysmal ascending aorta

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Fourth, while it is possible that the results may be marginally different depending on the ratio use for the training, validation, and testing tests, we chose the ratio of 7:2:1 based on prior literature using data science approaches 41 . Fifth, we focused our main analysis on precision at k from 1 to 10% given that this is a commonly used threshold in the literature to define high-need, high-cost populations, including in a recent report by the National Academy of Medicine 42 , and considered a clinically actionable range for a population health approaches to target individuals at highest risk. Although further analysis at higher precision cutoffs may yield different results, we think that is unlikely given the clear trends in Fig.…”

Section: Discussionmentioning

confidence: 99%

Comparison of deep learning with traditional models to predict preventable acute care use and spending among heart failure patients

Lewis¹,

Elad²,

Beladev³

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Recent health reforms have created incentives for cardiologists and accountable care organizations to participate in value-based care models for heart failure (HF). Accurate risk stratification of HF patients is critical to efficiently deploy interventions aimed at reducing preventable utilization. The goal of this paper was to compare deep learning approaches with traditional logistic regression (LR) to predict preventable utilization among HF patients. We conducted a prognostic study using data on 93,260 HF patients continuously enrolled for 2-years in a large U.S. commercial insurer to develop and validate prediction models for three outcomes of interest: preventable hospitalizations, preventable emergency department (ED) visits, and preventable costs. Patients were split into training, validation, and testing samples. Outcomes were modeled using traditional and enhanced LR and compared to gradient boosting model and deep learning models using sequential and non-sequential inputs. Evaluation metrics included precision (positive predictive value) at k, cost capture, and Area Under the Receiver operating characteristic (AUROC). Deep learning models consistently outperformed LR for all three outcomes with respect to the chosen evaluation metrics. Precision at 1% for preventable hospitalizations was 43% for deep learning compared to 30% for enhanced LR. Precision at 1% for preventable ED visits was 39% for deep learning compared to 33% for enhanced LR. For preventable cost, cost capture at 1% was 30% for sequential deep learning, compared to 18% for enhanced LR. The highest AUROCs for deep learning were 0.778, 0.681 and 0.727, respectively. These results offer a promising approach to identify patients for targeted interventions.

show abstract

“…This is mainly due to the lack of annotated datasets. Recently, Baskaran et al [31] used the UNet architecture to segment the proximal ascending and descending aorta (PAA, DA), superior and inferior vena cavae (SVC, IVC), pulmonary artery (PA), coronary sinus (CS), right ventricular wall (RVW), and left atrial wall (LAW) and made the dataset publicly available. This dataset is used in our work.…”

Section: Previous Methods For Cardiovascular Segmentationmentioning

confidence: 99%

AB-ResUNet+: Improving Multiple Cardiovascular Structure Segmentation from Computed Tomography Angiography Images

et al. 2022

View full text Add to dashboard Cite

Accurate segmentation of cardiovascular structures plays an important role in many clinical applications. Recently, fully convolutional networks (FCNs), led by the UNet architecture, have significantly improved the accuracy and speed of semantic segmentation tasks, greatly improving medical segmentation and analysis tasks. The UNet architecture makes heavy use of contextual information. However, useful channel features are not fully exploited. In this work, we present an improved UNet architecture that exploits residual learning, squeeze and excitation operations, Atrous Spatial Pyramid Pooling (ASPP), and the attention mechanism for accurate and effective segmentation of complex cardiovascular structures and name it AB-ResUNet+. The channel attention block is inserted into the skip connection to optimize the coding ability of each layer. The ASPP block is located at the bottom of the network and acts as a bridge between the encoder and decoder. This increases the field of view of the filters and allows them to include a wider context. The proposed AB-ResUNet+ is evaluated on eleven datasets of different cardiovascular structures, including coronary sinus (CS), descending aorta (DA), inferior vena cava (IVC), left atrial appendage (LAA), left atrial wall (LAW), papillary muscle (PM), posterior mitral leaflet (PML), proximal ascending aorta (PAA), pulmonary aorta (PA), right ventricular wall (RVW), and superior vena cava (SVC). Our experimental evaluations show that the proposed AB-ResUNet+ significantly outperforms the UNet, ResUNet, and ResUNet++ architecture by achieving higher values in terms of Dice coefficient and mIoU.

show abstract

Automatic segmentation of multiple cardiovascular structures from cardiac computed tomography angiography images using deep learning

Abstract: Objectives To develop, demonstrate and evaluate an automated deep learning method for multiple cardiovascular structure segmentation.

Cited by 32 publications

References 23 publications

Deep learning approach for the segmentation of aneurysmal ascending aorta

Deep learning approach for the segmentation of aneurysmal ascending aorta

Comparison of deep learning with traditional models to predict preventable acute care use and spending among heart failure patients

AB-ResUNet+: Improving Multiple Cardiovascular Structure Segmentation from Computed Tomography Angiography Images

Contact Info

Product

Resources

About