SPASM: Segmentation of Sparse and Arbitrarily Oriented Cardiac MRI Data Using a 3D-ASM

Abstract

Abstract. In this paper, a new technique (SPASM) based on a 3D-ASM is presented for automatic segmentation of cardiac MRI image data sets consisting of multiple planes with different orientations, and with large undersampled regions. SPASM was applied to sparsely sampled and radially oriented cardiac LV image data.Performance of SPASM has been compared to results from other methods reported in literature. The accuracy of SPASM is comparable to these other methods, but SPASM uses considerably less image data.

Automatic Quantitative Analysis of Myocardial Wall Motion and Thickening from Long-and Short-Axis Cine MRI Studies

“…The 3D extension of the algorithm (3D-ASM) was put forward by van Assen et al [3], using an optimized fuzzy inference strategy to provide the appearance (grey-level) model necessary to deform the shape model, and conduct the segmentation. This methodology allows for using sparse and evenly sampled image slices in any 3D position and orientation [4].…”

Efficient Reconstruction of Cardiac LV Surfaces Using a 3D Sparse ASM

“…The PDM is trained from a population of typical examples of the target shape, and models shape variability as a linear combination of a mean shape, and a number of eigenshapes. Three dimensional ASMs for cardiac segmentation were described by Van Assen et al [2] and Kaus et al [4].…”

“…The second part of the SPASM, the matching algorithm, is based on a Takagi-Sugeno Fuzzy Inference System (FIS) using Fuzzy C-means (FCM) [2] …”

“…A recently proposed MRI technique, based on acquisition of the radially (RAD) oriented long-axis (LA) images, provides a clearer depiction of both the apical and basal part of the heart and can be alternatively utilized for quantification of LV volume and mass [1]. In this paper a new method for fully automated, rapid, accurate 3D reconstruction of the LV surfaces, based on a special purpose 3D-ASM coined SPASM and proposed in [2], is thoroughly evaluated. While in [2], this method was only applied to a four-slice radial LA data set, here multiple data sets comprising differently oriented slices are used.…”

“…In this paper a new method for fully automated, rapid, accurate 3D reconstruction of the LV surfaces, based on a special purpose 3D-ASM coined SPASM and proposed in [2], is thoroughly evaluated. While in [2], this method was only applied to a four-slice radial LA data set, here multiple data sets comprising differently oriented slices are used. Moreover, the sparsity limit of these data sets is sought.…”