Robust left ventricle segmentation from ultrasound data using deep neural networks and efficient search methods

Carneiro, Gustavo; Nascimento, Jacinto C.; Freitas, António

doi:10.1109/isbi.2010.5490181

Cited by 25 publications

(26 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This complexity has been reduced in several ways, such as with the branch and bound framework [19], which allows a reduction to O(K R/2 + S) or the marginal space learning (MSL) [4] which partitions the search space into subspaces of increasing complexity, achieving a complexity reduction of O(K + (R − 1) × ♯scales × K fine + S), where ♯scales accounts for the number of scales and K fine represents a reduced number of promising samples (note that in [1,3,4,20], ♯scales = 3 and K fine = O(10 1 )). Coarse-to-fine based derivative search has also been proposed in [1,3,20], which uses GA approach in the space of R dimensions [1,20] achieving a complexity of O(K + ♯scales × K fine × R + S). The use of sparse manifolds in the rigid detection has been implemented in [3] achieving a complexity of O((…”

Section: Running-time Complexity Comparisonsmentioning

confidence: 99%

Non-rigid Segmentation Using Sparse Low Dimensional Manifolds and Deep Belief Networks

Nascimento¹,

Carneiro

2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

Self Cite

View full text Add to dashboard Cite

Section: Running-time Complexity Comparisonsmentioning

confidence: 99%

Non-rigid Segmentation Using Sparse Low Dimensional Manifolds and Deep Belief Networks

Nascimento¹,

Carneiro

2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

Self Cite

View full text Add to dashboard Cite

“…Table I shows the general characteristics of the following methods: 1) bottom-up approaches [32], [33]; 2) active contours methods [7]; 3) active shape models (ASMs) [22]; 4) deformable templates [14]- [18], [36]; 5) active appearance models (AAMs) [3], [23], [25]; 6) level-set approaches [4]- [6], [8]- [13], [34], [35]; and 7) database-guided (DB-guided) segmentation [19]- [21], [24], [26], [27], [37]. In this table, five properties are used to define each method, where mark indicates the presence of the property and symbol means that, although the property is present in latest developments, it was not part of the original formulation.…”

Section: Literature Reviewmentioning

confidence: 99%

“…The smallest point to contour distance is (19) which is the distance to the closest point (DCP). The AV between and is (20) The HDF is defined as the maximum DCP between and , as in the following:…”

Section: Error Measuresmentioning

confidence: 99%

“…The most successful LV segmentation systems are based on the following techniques: active contours [4]- [13], deformable templates [14]- [18], and supervised learning methods [3], [19]- [27]. Although excellent results have been achieved by active contours and deformable templates, these methods are effective only to the extent of the prior knowledge about the LV shape and appearance present in the method [24].…”

mentioning

confidence: 99%

“…The complexity issue is addressed with the use of optimization algorithms of first (gradient descent) and second (Newton's method) orders [29]. This paper is an extension of the approach presented in [19], with more complete literature review, methodology derivations, and experiments (including a new comparison with interuser variations). Moreover, this paper is focused on the LV segmentation in still images, which is a different goal compared with [20], which addresses the problem of LV tracking.…”

mentioning

confidence: 99%

See 2 more Smart Citations

The Segmentation of the Left Ventricle of the Heart From Ultrasound Data Using Deep Learning Architectures and Derivative-Based Search Methods

Carneiro

Nascimento

Freitas

2012

IEEE Trans. on Image Process.

181

View full text Add to dashboard Cite

Abstract-We present a new supervised learning model designed for the automatic segmentation of the left ventricle (LV) of the heart in ultrasound images. We address the following problems inherent to supervised learning models: 1) the need of a large set of training images; 2) robustness to imaging conditions not present in the training data; and 3) complex search process. The innovations of our approach reside in a formulation that decouples the rigid and nonrigid detections, deep learning methods that model the appearance of the LV, and efficient derivative-based search algorithms. The functionality of our approach is evaluated using a data set of diseased cases containing 400 annotated images (from 12 sequences) and another data set of normal cases comprising 80 annotated images (from two sequences), where both sets present long axis views of the LV. Using several error measures to compute the degree of similarity between the manual and automatic segmentations, we show that our method not only has high sensitivity and specificity but also presents variations with respect to a gold standard (computed from the manual annotations of two experts) within interuser variability on a subset of the diseased cases. We also compare the segmentations produced by our approach and by two state-of-the-art LV segmentation models on the data set of normal cases, and the results show that our approach produces segmentations that are comparable to these two approaches using only 20 training images and increasing the training set to 400 images causes our approach to be generally more accurate. Finally, we show that efficient search methods reduce up to tenfold the complexity of the method while still producing competitive segmentations. In the future, we plan to include a dynamical model to improve the performance of the algorithm, to use semisupervised learning methods to reduce even more the dependence on rich and large training sets, and to design a shape model less dependent on the training set.

show abstract

Automated left ventricular myocardium segmentation using 3D deeply supervised attention U‐net for coronary computed tomography angiography; CT myocardium segmentation

Guo

Lei

et al. 2020

Medical Physics

View full text Add to dashboard Cite

Purpose Segmentation of left ventricular myocardium (LVM) in coronary computed tomography angiography (CCTA) is important for diagnosis of cardiovascular diseases. Due to poor image contrast and large variation in intensity and shapes, LVM segmentation for CCTA is a challenging task. The purpose of this work is to develop a region‐based deep learning method to automatically detect and segment the LVM solely based on CCTA images. Methods We developed a 3D deeply supervised U‐Net, which incorporates attention gates (AGs) to focus on the myocardial boundary structures, to segment LVM contours from CCTA. The deep attention U‐Net (DAU‐Net) was trained on the patients’ CCTA images, with a manual contour‐derived binary mask used as the learning‐based target. The network was supervised by a hybrid loss function, which combined logistic loss and Dice loss to simultaneously measure the similarities and discrepancies between the prediction and training datasets. To evaluate the accuracy of the segmentation, we retrospectively investigated 100 patients with suspected or confirmed coronary artery disease (CAD). The LVM volume was segmented by the proposed method and compared with physician‐approved clinical contours. Quantitative metrics used were Dice similarity coefficient (DSC), Hausdorff distance (HD), mean surface distance (MSD), residual mean square distance (RMSD), the center of mass distance (CMD), and volume difference (VOD). Results The proposed method created contours with very good agreement to the ground truth contours. Our proposed segmentation approach is benchmarked primarily using fivefold cross validation. Model prediction correlated and agreed well with manual contour. The mean DSC of the contours delineated by our method was 91.6% among all patients. The resultant HD was 6.840 ± 4.410 mm. The proposed method also resulted in a small CMD (1.058 ± 1.245 mm) and VOD (1.640 ± 1.777 cc). Among all patients, the MSD and RMSD were 0.433 ± 0.209 mm and 0.724 ± 0.375 mm, respectively, between ground truth and LVM volume resulting from the proposed method. Conclusions We developed a novel deep learning‐based approach for the automated segmentation of the LVM on CCTA images. We demonstrated the high accuracy of the proposed learning‐based segmentation method through comparison with ground truth contour of 100 clinical patient cases using six quantitative metrics. These results show the potential of using automated LVM segmentation for computer‐aided delineation of CADs in the clinical setting.

show abstract

Robust left ventricle segmentation from ultrasound data using deep neural networks and efficient search methods

Cited by 25 publications

References 17 publications

Non-rigid Segmentation Using Sparse Low Dimensional Manifolds and Deep Belief Networks

Non-rigid Segmentation Using Sparse Low Dimensional Manifolds and Deep Belief Networks

The Segmentation of the Left Ventricle of the Heart From Ultrasound Data Using Deep Learning Architectures and Derivative-Based Search Methods

Automated left ventricular myocardium segmentation using 3D deeply supervised attention U‐net for coronary computed tomography angiography; CT myocardium segmentation

Contact Info

Product

Resources

About