Segmentation of Cardiac Structures via Successive Subspace Learning with Saab Transform from Cine MRI

TL;DR

This paper introduces a lightweight, interpretable segmentation method for cardiac MRI using successive subspace learning with the Saab transform, outperforming CNNs in accuracy and efficiency for clinical applications.

Contribution

The study proposes a novel segmentation framework based on Saab transform and successive subspace learning, reducing model complexity while maintaining high accuracy.

Findings

Outperforms state-of-the-art U-Net models in segmentation accuracy

Uses 200 times fewer parameters than CNN-based models

Demonstrates potential for clinical application in cardiac MRI analysis

Abstract

Assessment of cardiovascular disease (CVD) with cine magnetic resonance imaging (MRI) has been used to non-invasively evaluate detailed cardiac structure and function. Accurate segmentation of cardiac structures from cine MRI is a crucial step for early diagnosis and prognosis of CVD, and has been greatly improved with convolutional neural networks (CNN). There, however, are a number of limitations identified in CNN models, such as limited interpretability and high complexity, thus limiting their use in clinical practice. In this work, to address the limitations, we propose a lightweight and interpretable machine learning model, successive subspace learning with the subspace approximation with adjusted bias (Saab) transform, for accurate and efficient segmentation from cine MRI. Specifically, our segmentation framework is comprised of the following steps: (1) sequential expansion of near-to-far neighborhood at different resolutions; (2) channel-wise subspace approximation using the Saab transform for unsupervised dimension reduction; (3) class-wise entropy guided feature selection for supervised dimension reduction; (4) concatenation of features and pixel-wise classification with gradient boost; and (5) conditional random field for post-processing. Experimental results on the ACDC 2017 segmentation database, showed that our framework performed better than state-of-the-art U-Net models with 200$\times$ fewer parameters in delineating the left ventricle, right ventricle, and myocardium, thus showing its potential to be used in clinical practice.