Generalized Super-Resolution 4D Flow MRI - Using Ensemble Learning to Extend Across the Cardiovascular System

• Published in: IEEE journal of biomedical and health informatics 2024, Vol.28 (12), p.7239-7250
• Main Authors: Ericsson, Leon, Hjalmarsson, Adam, Akbar, Muhammad Usman, Ferdian, Edward, Bonini, Mia, Hardy, Brandon, Schollenberger, Jonas, Aristova, Maria, Winter, Patrick, Burris, Nicholas, Fyrdahl, Alexander, Sigfridsson, Andreas, Schnell, Susanne, Figueroa, C. Alberto, Nordsletten, David, Young, Alistair A., Marlevi, David
• Format: Article
• Language: English
• Subjects: 4D flow MRI; Biomedical imaging; cardiovascular; Data models; Ensemble learning; Hemodynamics; Magnetic resonance imaging; super-resolution; Superresolution; Training
• ISSN: 2168-2194; 2168-2208; 2168-2208
• DOI: 10.1109/JBHI.2024.3429291

4D Flow Magnetic Resonance Imaging (4D Flow MRI) is a non-invasive measurement technique capable of quantifying blood flow across the cardiovascular system. While practical use is limited by spatial resolution and image noise, incorporation of trained super-resolution (SR) networks has potential to enhance image quality post-scan. However, these efforts have predominantly been restricted to narrowly defined cardiovascular domains, with limited exploration of how SR performance extends across the cardiovascular system; a task aggravated by contrasting hemodynamic conditions apparent across the cardiovasculature. The aim of our study was therefore to explore the generalizability of SR 4D Flow MRI using a combination of existing super-resolution base models, novel heterogeneous training sets, and dedicated ensemble learning techniques; the latter-most being effectively used for improved domain adaption in other domains or modalities, however, with no previous exploration in the setting of 4D Flow MRI. With synthetic training data generated across three disparate domains (cardiac, aortic, cerebrovascular), varying convolutional base and ensemble learners were evaluated as a function of domain and architecture, quantifying performance on both in-silico and acquired in-vivo data from the same three domains. Results show that both bagging and stacking ensembling enhance SR performance across domains, accurately predicting high-resolution velocities from low-resolution input data in-silico . Likewise, optimized networks successfully recover native resolution velocities from downsampled in-vivo data, as well as show qualitative potential in generating denoised SR-images from clinical-level input data. In conclusion, our work presents a viable approach for generalized SR 4D Flow MRI, with the novel use of ensemble learning in the setting of advanced full-field flow imaging extending utility across various clinical areas of interest.