Abstract
Accurate cardiac motion estimation from medical images such as ultrasound is important for clinical evaluation. We present a novel regularisation layer for cardiac motion estimation that will be applied after image registration and demonstrate its effectiveness. The regularisation utilises a spatio-temporal model of motion, b-splines of Fourier, to fit to displacement fields from pairwise image registration. In the process, it enforces spatial and temporal smoothness and consistency, cyclic nature of cardiac motion, and better adherence to the stroke volume of the heart. Flexibility is further given for inclusion of any set of registration displacement fields. The approach gave high accuracy. When applied to human adult Ultrasound data from a Cardiac Motion Analysis Challenge (CMAC), the proposed method is found to have 10% lower tracking error over CMAC participants. Satisfactory cardiac motion estimation is also demonstrated on other data sets, including human fetal echocardiography, chick embryonic heart ultrasound images, and zebrafish embryonic microscope images, with the average Dice coefficient between estimation motion and manual segmentation at 0.82–0.87. The approach of performing regularisation as an add-on layer after the completion of image registration is thus a viable option for cardiac motion estimation that can still have good accuracy. Since motion estimation algorithms are complex, dividing up regularisation and registration can simplify the process and provide flexibility. Further, owing to a large variety of existing registration algorithms, such an approach that is usable on any algorithm may be useful.
Highlights
Cyclic constraint dictates that the transformation returns to its original state smoothly after one cardiac cycle, just as the myocardium should return to its initial state after its contraction
Three of the four participants of the Cardiac motion Analysis Challenge (CMAC) performed cardiac motion tracking from the ultrasound data: Fraunhofer MEVIS (MEVIS) from Germany, the Inria-Asclepios project (INRIA) from France, and Universitat Pompeu Fabra (UPF) from Spain
At the initialisation step, tracking errors are comparable with results from CMAC participants, with an Euclidean distance (Eu) of 3.80 ± 0.42 mm
Summary
Cyclic constraint dictates that the transformation returns to its original state smoothly after one cardiac cycle, just as the myocardium should return to its initial state after its contraction. The transformation across a large time step should be similar to the transformations of its smaller constituent time steps compounded together Many authors mitigated this issue by using information from various registration paths, such as the combination of sequential registration and registration against a fixed reference time[14,15,16]. This study aims to improve the tracking accuracy of myocardial motion by an add-on regularisation layer of pairwise image registrations. The proposed framework is designed to enforce spatio-temporal smoothness, cyclic-nature of cardiac motion, and temporal consistency discussed previously. Temporal consistency and cyclic constraint can be satisfied by having a spatial temporal model of the motion. We further demonstrate the feasibility of the proposed framework by comparison of tracking accuracy against data from Cardiac motion Analysis Challenge (CMAC)[19]. Each of the groups used their own unique algorithm, which would be described below
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call