Automatic left ventricular segmentation in 4D interventional ultrasound data using a patient-specific temporal synchronized shape prior

摘要

The fusion of pre-operative 3D magnetic resonance (MR) images with real-time 3D ultrasound (US) images can be themost beneficial way to guide minimally invasive cardiovascular interventions without radiation. Previously, weaddressed this topic through a strategy to segment the left ventricle (LV) on interventional 3D US data using apersonalized shape prior obtained from a pre-operative MR scan. Nevertheless, this approach was semi-automatic,requiring a manual alignment between US and MR image coordinate systems.In this paper, we present a novel solution to automate the abovementioned pipeline. In this sense, a method toautomatically detect the right ventricular (RV) insertion point on the US data was developed, which is subsequentlycombined with pre-operative annotations of the RV position in the MR volume, therefore allowing an automaticalignment of their coordinate systems. Moreover, a novel strategy to ensure a correct temporal synchronization of the USand MR models is applied. Finally, a full evaluation of the proposed automatic pipeline is performed.The proposed automatic framework was tested in a clinical database with 24 patients containing both MR and USscans. A similar performance between the proposed and the previous semi-automatic version was found in terms ofrelevant clinical measurements. Additionally, the automatic strategy to detect the RV insertion point showed itseffectiveness, with a good agreement against manually identified landmarks.Overall, the proposed automatic method showed high feasibility and a performance similar to the semi-automaticversion, reinforcing its potential for normal clinical routine.