A Normalized Framework for the Design of Feature Spaces Assessing the Left Ventricular Function

Abstract

A through description of the left ventricle functionality requires combining complementary regional scores. A main limitation is the lack of multiparametric normality models oriented to the assessment of regional wall motion abnormalities (RWMA). This paper covers two main topics involved in RWMA assessment. We propose a general framework allowing the fusion and comparison across subjects of different regional scores. Our framework is used to explore which combination of regional scores (including 2-D motion and strains) is better suited for RWMA detection. Our statistical analysis indicates that for a proper (within interobserver variability) identification of RWMA, models should consider motion and extreme strains.

Multi-scale deep networks and regression forests for direct bi-ventricular volume estimation

“…Cardiac image representation serves as a fundamental role in cardiac function analysis and is typically obtained by feature engineering in the existing methods (Montillo et al, 2004;Qian et al, 2006;Zheng et al, 2008;Chen et al, 2010;Garcia-Barnes et al, 2010;Afshin et al, 2012a;Punithakumar et al, 2013;Wang et al, 2014;Zhen et al, 2014d). Handcrafted features by feature engineering are labor-intensive and suffers from the weakness of being unable to extract and organize representative information from the data (Bengio et al, 2013;Tang et al, 2014).…”

Regional Heart Motion Abnormality Detection via Multiview Fusion

“…Accurate detection of motion abnormality of regional myocardial segments in MRI is essential in the diagnosis and treatment of coronary heart disease [7,11,15], the leading cause of death worldwide. The problem has attracted a recent research attention recently [7,11,15].…”

“…The problem has attracted a recent research attention recently [7,11,15]. Unfortunately, existing MRI-based methods rely only on short-axis sequences [7,11,15].…”

Assessment of Regional Myocardial Function via Statistical Features in MR Images

“…In clinical practice, the regional myocardial function is commonly scored by following American Heart Association (AHA) standards [5], where the LV is divided into 17 segments. Existing regional heart function analysis methods are based on information theoretic measures and unscented Kalman filter approaches [6], differentiable manifolds [7], independent component analysis classifier [8], pattern recognition technique based on intra-segment correlation [9], and tensor-based classification [10]. Most of the existing methods require extensive user interaction or computationally expensive segmentation algorithms.…”