AI Virtual Native Enhancement for Gadolinium-Free CMR Tissue Characterization in Hypertrophic Cardiomyopathy
TAKE-HOME MESSAGE
-
The authors of this study evaluated a deep learning model of cardiovascular magnetic resonance (CMR) imaging without gadolinium in patients with hypertrophic cardiomyopathy. The CMR Virtual Native Enhancement (VNE) imaging technology had significantly better image quality and ability to detect cardiac lesions than late gadolinium enhancement (LGE). In addition, the VNE studies were carried out more quickly.
- New deep learning models such as VNE may improve CMR technology by providing higher-quality and more accurate studies with a shorter acquisition time.
abstract
This abstract is available on the publisher's site.
Access this abstract nowBackground
Late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) imaging is the gold standard for non-invasive myocardial tissue characterization, but requires intravenous contrast agent administration. It is highly desired to develop a contrast-agent-free technology to replace LGE for faster and cheaper CMR scans.
Methods
A CMR Virtual Native Enhancement (VNE) imaging technology was developed using artificial intelligence. The deep learning model for generating VNE uses multiple streams of convolutional neural networks to exploit and enhance the existing signals in native T1-maps (pixel-wise maps of tissue T1 relaxation times) and cine imaging of cardiac structure and function, presenting them as LGE-equivalent images. The VNE generator was trained using generative adversarial networks. This technology was first developed on CMR datasets from the multi-center Hypertrophic Cardiomyopathy Registry (HCMR), using HCM as an exemplar. The datasets were randomized into two independent groups for deep learning training and testing. The test data of VNE and LGE were scored and contoured by experienced human operators to assess image quality, visuospatial agreement and myocardial lesion burden quantification. Image quality was compared using nonparametric Wilcoxon test. Intra- and inter-observer agreement was analyzed using intraclass correlation coefficients (ICC). Lesion quantification by VNE and LGE were compared using linear regression and ICC.
Results
1348 HCM patients provided 4093 triplets of matched T1-maps, cines, and LGE datasets. After randomization and data quality control, 2695 datasets were used for VNE method development, and 345 for independent testing. VNE had significantly better image quality than LGE, as assessed by 4 operators (n=345 datasets, p<0.001, Wilcoxon test). VNE revealed characteristic HCM lesions in high visuospatial agreement with LGE. In 121 patients (n=326 datasets), VNE correlated with LGE in detecting and quantifying both hyper-intensity myocardial lesions (r=0.77-0.79, ICC=0.77-0.87; p<0.001) and intermediate-intensity lesions (r=0.70-0.76, ICC=0.82-0.85; p<0.001). The native CMR images (cine plus T1-map) required for VNE can be acquired within 15 minutes. Producing a VNE image takes less than one second.
Conclusions
VNE is a new CMR technology that resembles conventional LGE, without the need for contrast administration. VNE achieved high agreement with LGE in the distribution and quantification of lesions, with significantly better image quality.
Click on any of these tags to subscribe to Topic Alerts. Once subscribed, you can get a single, daily email any time PracticeUpdate publishes content on the topics that interest you.
Visit your Preferences and Settings section to Manage All Topic Alerts
Additional Info
Disclosure statements are available on the authors' profiles:
Towards Replacing Late Gadolinium Enhancement With Artificial Intelligence Virtual Native Enhancement for Gadolinium-Free Cardiovascular Magnetic Resonance Tissue Characterization in Hypertrophic Cardiomyopathy
Circulation 2021 Jul 07;[EPub Ahead of Print], Q Zhang, MK Burrage, E Lukaschuk, M Shanmuganathan, IA Popescu, C Nikolaidou, R Mills, K Werys, E Hann, A Barutcu, SD Polat, M Salerno, M Jerosch-Herold, RY Kwong, HC Watkins, CM Kramer, S Neubauer, VM Ferreira, SK Piechnik,From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.
The current paper is of great interest. Contemporary practice of cardiac MRI often uses gadolinium-based contrast agents (GBCA) to assess late gadolinium enhancement (LGE), which reflects fibrosis and scarring. An enormous literature over the past 20 years has shown the clinical utility of LGE in identifying and quantifying myocardial infarction, but also in illuminating the importance of fibrosis and scarring in syndromes such as nonischemic cardiomyopathy or hypertrophic cardiomyopathy (HCM). The process of performing LGE imaging with GBCA requires an IV line for injection, some patients are allergic, and these contrast agents cannot be used below a certain threshold of renal function. A good amount of post-processing is required to obtain quality images.
The authors have developed a technique using the CMR images without contrast, which they call Virtual Native Enhancement (VNE) imaging technology, using artificial intelligence. Leveraging the large database of the NIH-funded HCM Registry (1348 HCM patients), they created 2 groups, one for deep learning training and another for validation testing compared to the LGE data sets. They found high levels of agreement by multiple analytic techniques, and of interest was that the VNE images were rated as higher quality overall compared to the LGE images. There were no “false positive” VNE studies, ie, any patients whose LGE was negative but VNE was positive.
This is a potentially important development which, if generalizable beyond this world-class MRI center, would facilitate the analysis of fibrosis in CMR imaging. Substantial data support the importance of LGE imaging, and indeed in patients with HCM decisions regarding sudden death risk and ICD implantation are sometimes made on the basis of the LGE data. Streamlining the acquisition process of such information in terms of acquisition time required, without the need for IV access, or worry about allergy or kidney function would be welcome.