Does an automated deep learning pipeline accurately localize and classify the configuration of a MitraClip device in 3D TEE images?
A deep learning pipeline can accurately segment and classify the configuration of a MitraClip device in 3D TEE images in an in vitro setting, offering a proof-of-concept for automated intraoperative guidance.
BACKGROUND AND OBJECTIVE: The MitraClip is the most widely used percutaneous treatment for mitral regurgitation, typically performed under the real-time guidance of 3D transesophageal echocardiography (TEE). However, artifacts and low image contrast in echocardiography hinder accurate clip visualization. This study presents a proof-of-concept of an automated pipeline for clip detection from 3D TEE images acquired in a controlled in vitro simulation environment. METHODS: An Attention UNet was employed to segment the device, while a DenseNet classifier predicted its configuration among ten possible states, ranging from fully closed to fully open. Based on the predicted configuration, a template model derived from computer-aided design (CAD) was automatically registered to refine the segmentation and enable quantitative characterization of the device. The pipeline was trained and validated on 196 3D TEE images acquired using a heart simulator, with ground-truth annotations refined through CAD-based templates. RESULTS: The Attention UNet achieved an average surface distance of 0.76 mm and a 95% Hausdorff distance of 2.44 mm for segmentation, while the DenseNet achieved an average weighted F1-score of 0.80 for classification. Post-refinement, segmentation accuracy improved, with average surface distance and 95% Hausdorff distance reduced to 0.69 mm and 1.83 mm, respectively. CONCLUSION: This pipeline enhanced clip visualization, providing fast and accurate detection with quantitative feedback, potentially improving procedural efficiency and reducing adverse outcomes.
Munafò et al. (Sat,) studied this question.