Dual Active-Shape Modeling for Efficient Right Ventricular Segmentation from MRI Images
Hossam El-Rewaidy1, El-Sayed H. Ibrahim2, and Ahmed Fahmy1

1Nile University, Cairo, Egypt, 2University of Michigan, Ann Arbor, MI, United States


Active-shape modeling (ASM) has potential for segmenting the right ventricle (RV) from MRI images. Nevertheless, the complexity of the RV shape does not allow for concisely capturing all possible shape variations. In this work, we propose a new ASM framework in which the RV contour is split into two simpler segments, septal and free-wall, whose shape variations are independently modeled using two ASM models. Further, the RV contours are aligned using the Bookstein coordinate-transformation. The results from a dataset of 10 patients show that the proposed framework can efficiently model complex RV shape variation with high accuracy in few iterations.


Assessment of the right-ventricular (RV) structure and function plays an important role in diagnosing and monitoring a number of cardiovascular diseases. A necessary step for such analysis is the delineation of the RV boundaries in the acquired images at different cardiac phases. Active-shape models (ASM) showed to have potential for segmenting the RV from MRI images. Basically, ASM’s detect the cardiac contour by minimizing an energy function that measures the difference between the model and image data.1,2 Nevertheless, the large variability and complexity of the RV shape do not allow for concisely capturing all possible shape variations among different patients and anatomical cross-sections. Noticeably, the latter increases the number of iterations required to converge to a proper solution and reduces the segmentation accuracy. In this work, we propose a modified ASM framework that can be used to efficiently capture the RV shape variations.


The developed technique includes two contributions. The first contribution involves splitting each RV contour in two segments: septal and free-wall (Figure 1(a)). In the training phase, this process is performed semi-automatically by selecting the two RV insertion points into the septal wall. Each contour segment is modeled separately, yielding a dual-ASM model. The two segments are then merged together using a third-order B-Spline algorithm to obtain a smooth RV contour. The second contribution of the developed technique involves using the Bookstein algorithm,3 instead of the conventional Procrustes method4 used in conventional ASM, to align the generated contours. In the Bookstein alignment method, a linear space of shape variations is used to represent the RV shape (Figure 1(b)). Given a contour from the training dataset, the two RV insertion points are manually selected and transformed into points (0,0) and (1,0) in the Bookstein Coordinates. The remaining contour points are then transformed to their corresponding points in the Bookstein Coordinates.

The developed technique has been tested on a dataset from 10 patients imaged with cine MRI (total of 546 short-axis images covering the whole cardiac cycle at the basal, mid-ventricular, and apical locations). The dataset was randomly divided into two subsets: a training set of 162 images and a testing set of 384 images. The ground truth was defined by manually delineating the RV boundaries in the dataset. It is worth noting that only one RV model was built from the three cross-sectional slices (basal, mid-cavity and apical) of each patient, rather than building three models, for improved segmentation efficiency. The principal component analysis technique was applied to the aligned shapes to estimate the mean shape for every segment. The first 8 and 15 modes of variation were selected to represent about 98% and 95% of the variance in the training set for the free-wall and septal segments, respectively.


Table 1 shows the mean±SD errors between the contours produced by the proposed dual-ASM and conventional ASM models with respect to the manually delineated contours at different cross-sectional slices. As can be seen in Table 1, the performance of the proposed ASM framework is better than that of the conventional model. This is evident by the lower value of the Mean Absolute Distance (MAD) (which measures the average absolute distance between each point on the estimated contour and the corresponding point on the manually delineated contour) and Hausdorf measure (which calculates the maximum distance between the two contours) and higher value of the Dice index (which measures the similarity between the set of points enclosed by the estimated contour and those enclosed by the ground truth contour). Figure 2 shows the evolution of the ASM models in two patients from the initial contour to the contours at iterations number 5 and 20. The figure shows that the initial contour of the proposed ASM framework is much better than that of the conventional ASM model. The figure also shows that the proposed ASM framework converges after about 5 iterations, whereas the conventional ASM model needs 15-20 iterations to correctly delineate the RV contour. The average computation times for segmenting one slice using a personal computer were 0.09 s and 0.17 s for the conventional and proposed ASM models, respectively. Nevertheless, the parallelized nature of the problem renders this difference insignificant.


The developed dual-ASM RV segmentation technique outperforms the conventional ASM framework and can efficiently model complex RV shape variation with more accuracy and in fewer iteration steps. Although the proposed framework extracts only the RV endocardium, the RV epicardium can be segmented through dilating the endocardium contour generated by the proposed technique.


Funding from ITAC program, CFP #59, ITIDA Agency, Ministry of Communication and Information Technology, Egypt.


1. Ginneken, B., Alejandro, F., Joes, S., et al. Active Shape Model Segmentation With Optimal Features. IEEE Transactions on Medical Imaging, 2002; 21(8):924-933.

2. ElBaz, M., Fahmy, A. Active shape model with inter-profile modeling paradigm for cardiac right ventricle segmentation. MICCAI, 2012; 15(1):691-698.

3. Bookstein, F. Size and Shape Spaces for Landmark Data in Two Dimensions. Statistical Science Journal, 1986; 1(2):181-242.

4. Ordas, S., Boisrobert, L., et al. Active shape models with invariant optimal features (IOF-ASM) application to cardiac MRI segmentation. Computers in Cardiology, 2003; 633-636.


Figure 1. (a)The RV contour is divided into septal and free-wall segments. (b)Transformation of the RV shape into Bookstein-Coordinates in 2 steps: 1)registering the two RV insertion points to points 0 and 1; and 2)normalizing each point on the original RV shape with-respect-to the distance between the insertion points.

Figure 2. The RV segmentation results in 2 subjects using the proposed and conventional ASM models at the initial, 5th, and 20th iterations. The figure shows 3 cross-sections at the basal, mid-cavity, and apical levels.

Table 1. Mean±SD of the mean absolute distance (MAD), Hausdorff measure, and Dice Index of the segmented contours at basal, mid-cavity, and apical levels using the proposed dual-ASM and conventional ASM models with respect to the ground-truth of manual segmentation. +p-value < 0.005; *p-value < 0.05; ^p-value < 0.07.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)