CTSL: Codebook-based Temporal-Spatial Learning for Accurate Non-Contrast Cardiac Risk Prediction Using Cine MRIs

TL;DR

This paper presents CTSL, a self-supervised, contrast-free framework that learns spatiotemporal features from Cine MRI sequences to accurately predict cardiac risk without needing segmentation masks or contrast agents.

Contribution

Introduces a novel self-supervised learning method, CTSL, that captures dynamic cardiac features from raw Cine MRI data without segmentation masks or contrast agents.

Findings

Outperforms traditional contrast-dependent methods in MACE risk prediction.

Effectively captures temporal dependencies and motion patterns.

Provides a rapid, non-invasive cardiac risk assessment tool.

Abstract

Accurate and contrast-free Major Adverse Cardiac Events (MACE) prediction from Cine MRI sequences remains a critical challenge. Existing methods typically necessitate supervised learning based on human-refined masks in the ventricular myocardium, which become impractical without contrast agents. We introduce a self-supervised framework, namely Codebook-based Temporal-Spatial Learning (CTSL), that learns dynamic, spatiotemporal representations from raw Cine data without requiring segmentation masks. CTSL decouples temporal and spatial features through a multi-view distillation strategy, where the teacher model processes multiple Cine views, and the student model learns from reduced-dimensional Cine-SA sequences. By leveraging codebook-based feature representations and dynamic lesion self-detection through motion cues, CTSL captures intricate temporal dependencies and motion patterns. High-confidence MACE risk predictions are achieved through our model, providing a rapid, non-invasive solution for cardiac risk assessment that outperforms traditional contrast-dependent methods, thereby enabling timely and accessible heart disease diagnosis in clinical settings.