A Registration-Based Propagation Framework for Automatic Whole Heart Segmentation of Cardiac MRI

TLDR

Magnetic resonance imaging is routinely used to assess cardiac morphology, and its extraction is critical for clinical applications and interventional planning, but automation is challenged by limited image quality and substantial inter‑subject heart shape variation. The study aims to develop a fully automatic whole heart segmentation method for cardiac MRI to eliminate manual delineation variability. We propose a framework that uses locally affine registration (LARM) to align anatomical substructures and free‑form deformations with adaptive control point status (ACPS FFD) to refine local details via constrained optimization. Validation on 37 diverse cardiac MR volumes yielded a mean surface distance error of 2.14 mm (±0.63 mm) and a maximum error of 4.31 mm.

Abstract

Magnetic resonance (MR) imaging has become a routine modality for the determination of patient cardiac morphology. The extraction of this information can be important for the development of new clinical applications as well as the planning and guidance of cardiac interventional procedures. To avoid inter- and intra-observer variability of manual delineation, it is highly desirable to develop an automatic technique for whole heart segmentation of cardiac magnetic resonance images. However, automating this process is complicated by the limited quality of acquired images and large shape variation of the heart between subjects. In this paper, we propose a fully automatic whole heart segmentation framework based on two new image registration algorithms: the locally affine registration method (LARM) and the free-form deformations with adaptive control point status (ACPS FFDs). LARM provides the correspondence of anatomical substructures such as the four chambers and great vessels of the heart, while the registration using ACPS FFDs refines the local details using a constrained optimization scheme. We validated our proposed segmentation framework on 37 cardiac MR volumes on the end-diastolic phase, displaying a wide diversity of morphology and pathology, and achieved a mean accuracy of 2.14 ± 0.63 mm (rms surface distance) and a maximal error of 4.31 mm.