Semi-supervised motion flow and myocardial strain estimation in cardiac videos using distance maps and memory networks.
Authors
Affiliations (7)
Affiliations (7)
- Sorbonne Université, CNRS, INSERM, Institut des systèmes intelligents et de robotique, ISIR, Paris, 75005, France; Sorbonne Université, CNRS, INSERM, Laboratoire d'Imagerie Biomédicale, LIB, Paris, 75006, France. Electronic address: [email protected].
- Sorbonne Université, CNRS, INSERM, Laboratoire d'Imagerie Biomédicale, LIB, Paris, 75006, France; Institut de Cardiométabolisme et Nutrition (ICAN), Paris, 75013, France.
- Sorbonne Université, CNRS, INSERM, Laboratoire d'Imagerie Biomédicale, LIB, Paris, 75006, France.
- Institut de Cardiométabolisme et Nutrition (ICAN), Paris, 75013, France.
- Sorbonne Université, ACTION group, Pitié-Salpêtrière Hospital (AP-HP), Paris, 75013, France.
- Sorbonne Université, CNRS, INSERM, Laboratoire d'Imagerie Biomédicale, LIB, Paris, 75006, France; Institut de Cardiométabolisme et Nutrition (ICAN), Paris, 75013, France; Imagerie Cardio-Thoracique (ICT), Sorbonne Université, AP-HP, Groupe Hospitalier Pitié-Salpêtrière, Paris, 75013, France.
- Sorbonne Université, CNRS, INSERM, Institut des systèmes intelligents et de robotique, ISIR, Paris, 75005, France.
Abstract
Myocardial strain plays a crucial role in diagnosing heart failure and myocardial infarction. Its computation relies on assessing heart muscle motion throughout the cardiac cycle. This assessment can be performed by following key points on each frame of a cine Magnetic Resonance Imaging (MRI) sequence. The use of segmentation labels yields more accurate motion estimation near heart muscle boundaries. However, since few frames in a cardiac sequence usually have segmentation labels, most methods either rely on annotated pairs of frames/volumes, greatly reducing available data, or use all frames of the cardiac cycle without segmentation supervision. Moreover, these techniques rarely utilize more than two phases during training. In this work, a new semi-supervised motion estimation algorithm using all frames of the cardiac sequence is presented. The distance map generated from the end-diastolic segmentation label is used to weight loss functions. The method is tested on an in-house dataset containing 271 patients. Several deep learning image registration and tracking algorithms were retrained on our dataset and compared to our approach. The proposed approach achieves an average End Point Error (EPE) of 1.02mm, against 1.19mm for RAFT (Recurrent All-Pairs Field Transforms). Using the end-diastolic distance map further improves this metric to 0.95mm compared to 0.91 for the fully supervised version. Correlations in systolic peak were 0.83 and 0.90 for the left ventricular global radial and circumferential strain respectively, and 0.91 for the right ventricular circumferential strain.