DeepStrain: A Deep Learning Workflow for the Automated Characterization of Cardiac Mechanics

Myocardial strain analysis from cinematic magnetic resonance imaging (cine-MRI) data could provide a more thorough characterization of cardiac mechanics than volumetric parameters such as left-ventricular ejection fraction, but sources of variation including segmentation and motion estimation have limited its wide clinical use. We designed and validated a deep learning (DL) workflow to generate both volumetric parameters and strain measures from cine-MRI data, including strain rate (SR) and regional strain polar maps, consisting of segmentation and motion estimation convolutional neural networks developed and trained using healthy and cardiovascular disease (CVD) subjects (n=150). DL-based volumetric parameters were correlated (>0.98) and without significant bias relative to parameters derived from manual segmentations in 50 healthy and CVD subjects. Compared to landmarks manually-tracked on tagging-MRI images from 15 healthy subjects, landmark deformation using DL-based motion estimates from paired cine-MRI data resulted in an endpoint-error of 2.9 (1.5) mm. Measures of end-systolic global strain from these cine-MRI data showed no significant biases relative to a tagging-MRI reference method. On 4 healthy subjects, intraclass correlation coefficient for intrascanner repeatability was excellent (>0.95) for strain, moderate to excellent for SR (0.690-0.963), and good to excellent (0.826-0.994) in most polar map segments. Absolute relative change was within ~5% for strain, within ~10% for SR, and <1% in half of polar map segments. In conclusion, we developed and evaluated a DL-based, end-to-end fully-automatic workflow for global and regional myocardial strain analysis to quantitatively characterize cardiac mechanics of healthy and CVD subjects based on ubiquitously acquired cine-MRI data.

Download Full-text

DeepStrain: A Deep Learning Workflow for the Automated Characterization of Cardiac Mechanics

Frontiers in Cardiovascular Medicine ◽

10.3389/fcvm.2021.730316 ◽

2021 ◽

Vol 8 ◽

Author(s):

Manuel A. Morales ◽

Maaike van den Boomen ◽

Christopher Nguyen ◽

Jayashree Kalpathy-Cramer ◽

Bruce R. Rosen ◽

...

Keyword(s):

Deep Learning ◽

Motion Estimation ◽

Healthy Subjects ◽

Strain Analysis ◽

Cardiac Mechanics ◽

Left Ventricular ◽

Left Ventricular Ejection ◽

Cine Mri ◽

Tagging Mri

Myocardial strain analysis from cinematic magnetic resonance imaging (cine-MRI) data provides a more thorough characterization of cardiac mechanics than volumetric parameters such as left-ventricular ejection fraction, but sources of variation including segmentation and motion estimation have limited its wider clinical use. We designed and validated a fast, fully-automatic deep learning (DL) workflow to generate both volumetric parameters and strain measures from cine-MRI data consisting of segmentation and motion estimation convolutional neural networks. The final motion network design, loss function, and associated hyperparameters are the result of a thorough ad hoc implementation that we carefully planned specific for strain quantification, tested, and compared to other potential alternatives. The optimal configuration was trained using healthy and cardiovascular disease (CVD) subjects (n = 150). DL-based volumetric parameters were correlated (>0.98) and without significant bias relative to parameters derived from manual segmentations in 50 healthy and CVD test subjects. Compared to landmarks manually-tracked on tagging-MRI images from 15 healthy subjects, landmark deformation using DL-based motion estimates from paired cine-MRI data resulted in an end-point-error of 2.9 ± 1.5 mm. Measures of end-systolic global strain from these cine-MRI data showed no significant biases relative to a tagging-MRI reference method. On 10 healthy subjects, intraclass correlation coefficient for intra-scanner repeatability was good to excellent (>0.75) for all global measures and most polar map segments. In conclusion, we developed and evaluated the first end-to-end learning-based workflow for automated strain analysis from cine-MRI data to quantitatively characterize cardiac mechanics of healthy and CVD subjects.

Download Full-text

Role of Echocardiography for the Perioperative Assessment of the Right Ventricle

Current Anesthesiology Reports ◽

10.1007/s40140-021-00474-1 ◽

2021 ◽

Author(s):

Jakob Labus ◽

Christopher Uhlig

Keyword(s):

Myocardial Strain ◽

Three Dimensional ◽

Strain Analysis ◽

Cardiac Mechanics ◽

Myocardial Strain Analysis ◽

Echocardiographic Evaluation ◽

Dimensional Echocardiography ◽

Three Dimensional Echocardiography ◽

The Right ◽

Rv Function

Abstract Purpose of Review This review aims to highlight the perioperative echocardiographic evaluation of right ventricular (RV) function with strengths and limitations of commonly used and evolving techniques. It explains the value of transthoracic echocardiography (TTE) and transesophageal echocardiography (TEE) and describes the perioperative changes of RV function echocardiographers should be aware of. Recent Findings RV dysfunction is an entity with strong influence on outcome. However, its definition and assessment in the perioperative interval are not well-defined. Moreover, values assessed by TTE and TEE are not interchangeable; while some parameters seem to correlate well, others do not. Myocardial strain analysis and three-dimensional echocardiography may overcome the limitations of conventional echocardiographic measures and provide further insight into perioperative cardiac mechanics. Summary Echocardiography has become an essential part of modern anesthesiology in patients with RV dysfunction. It offers the opportunity to evaluate not only global but also regional RV function and distinguish alterations of RV contraction.

Download Full-text

Echocardiographic Myocardial Strain Analysis Describes Subclinical Cardiac Dysfunction after Craniospinal Irradiation in Pediatric and Young Adult Patients with Central Nervous System Tumors

10.21203/rs.3.rs-121700/v1 ◽

2020 ◽

Author(s):

Thomas D Ryan ◽

Hugo R. Martinez ◽

Ralph Salloum ◽

Erin Wright ◽

Lauren Bueche ◽

...

Keyword(s):

Central Nervous System ◽

Cardiac Dysfunction ◽

Myocardial Strain ◽

Strain Analysis ◽

Late Time ◽

Systolic Strain ◽

Peak Systolic Strain ◽

Myocardial Strain Analysis ◽

Late Time Point ◽

Time Point

Abstract Background: Craniospinal irradiation (CSI) is part of the treatment of central nervous system (CNS) tumors and is associated with cardiovascular disease in adults. Global myocardial strain analysis including longitudinal peak systolic strain (GLS), circumferential peak systolic strain (GCS) and radial peak systolic strain (GRS) can reveal subclinical cardiac dysfunction.Methods: Retrospective, single-center study in patients managed with CSI vs. age-matched controls. Clinical data and echocardiography, including myocardial strain analysis, were collected at early (<12 months) and late ( 12 months) after completion of CSI.Results: Echocardiograms were available in 20 early and 34 late patients. Patients at the late time point were older (21.7±10.4 vs. 13.3 9.6 years), and further out from CSI (13.1±8.8 vs. 0.2±0.3 years). Standard echocardiographic parameters were normal for all subjects. For the early time, CSI vs. control: GLS was -16.8 3.6% vs. -21.3 4.0% (p=0.0002), GCS was -22.5 5.2% vs. -21.3 3.4% (p=0.28), and GRS was 21.8 11.0% vs. 26.9 7.7% (p=0.07). At the late time point, CSI vs. control: GLS was -16.2 5.4% vs. -21.6 3.7% (p<0.0001), GCS was -20.9 6.8% vs. -21.9 3.5% (p=0.42), and GRS was 22.5 10.0% vs. 27.3 8.3% (p=0.03). Radiation type (proton vs. photon), and radiation dose (<30 Gy vs. 30 Gy) did not impact any parameter.Conclusions: Subclinical cardiac systolic dysfunction by GLS is present both early and late after CSI. These results argue for inclusion of baseline cardiovascular assessment and early initiation of longitudinal follow-up post CSI.

Download Full-text