A new pipeline generates ultra-realistic 2D synthetic echocardiographic sequences controlled by a computational heart model to facilitate benchmarking of cardiac deformation and strain algorithms.
The development of ultra-realistic synthetic echocardiographic sequences provides a standardized quality assurance framework to benchmark and improve the reproducibility of cardiac deformation and strain imaging algorithms.
Despite the overwhelming availability of techniques for computation of cardiac deformation and strain with 2D echocardiography, their widespread dissemination in clinical practice is still held back by the reported low reproducibility between different solutions. This can in part be attributed to the lack of a solid and open quality assurance framework to assess and compare their performance. Building upon our previous work, we present here a new pipeline for the benchmarking of such algorithms making use of realistic 2D synthetic sequences. While the synthetic motion is fully controlled by a computational heart model, the visual appearance is extremely similar to a real ultrasound recording. The pipeline is used to generate an initial library of four sequences.
Alessandrini et al. (Wed,) conducted a other in Cardiac deformation and strain. Synthetic echocardiographic sequences pipeline was evaluated. A new pipeline generates ultra-realistic 2D synthetic echocardiographic sequences controlled by a computational heart model to facilitate benchmarking of cardiac deformation and strain algorithms.