Toward Replacing Late Gadolinium Enhancement With Artificial Intelligence Virtual Native Enhancement for Gadolinium-Free Cardiovascular Magnetic Resonance Tissue Characterization in Hypertrophic Cardiomyopathy

• Published in: Circulation (New York, N.Y.) N.Y.), 2021-08, Vol.144 (8), p.589-599
• Main Authors: Zhang, Qiang, Burrage, Matthew K., Lukaschuk, Elena, Shanmuganathan, Mayooran, Popescu, Iulia A., Nikolaidou, Chrysovalantou, Mills, Rebecca, Werys, Konrad, Hann, Evan, Barutcu, Ahmet, Polat, Suleyman D., Salerno, Michael, Jerosch-Herold, Michael, Kwong, Raymond Y., Watkins, Hugh C., Kramer, Christopher M., Neubauer, Stefan, Ferreira, Vanessa M., Piechnik, Stefan K.
• Format: Article
• Language: English
• Subjects: Artificial Intelligence; Cardiomyopathy, Hypertrophic - diagnostic imaging; Cardiomyopathy, Hypertrophic - etiology; Cardiomyopathy, Hypertrophic - pathology; Contrast Media; Deep Learning; Gadolinium; Humans; Image Enhancement; Image Processing, Computer-Assisted; Magnetic Resonance Imaging - methods; Original s
• ISSN: 0009-7322; 1524-4539
• DOI: 10.1161/CIRCULATIONAHA.121.054432

Late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) imaging is the gold standard for noninvasive 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. 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 multicenter Hypertrophic Cardiomyopathy Registry, using hypertrophic cardiomyopathy as an exemplar. The datasets were randomized into 2 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 a nonparametric Wilcoxon test. Intra- and interobserver agreement was analyzed using intraclass correlation coefficients (ICC). Lesion quantification by VNE and LGE were compared using linear regression and ICC. A total of 1348 hypertrophic cardiomyopathy 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 were used for independent testing. VNE had significantly better image quality than LGE, as assessed by 4 operators (n=345 datasets;