Key result
Deep learning reconstruction of cine bSSFP up to 5-fold acceleration maintains diagnostic quality and comparable volumes.
Why the study?
Breath-holding for cine bSSFP imaging is challenging for patients with impaired capacity, and deep learning-based reconstruction of undersampled k-space could shorten breath-holds while preserving image quality.
Does deep learning-based reconstruction of undersampled cine bSSFP images preserve image quality and biventricular volumetric accuracy compared to fully sampled acquisitions in patients with pectus excavatum?
Observational (n=15)
Does deep learning-based reconstruction of undersampled cine bSSFP images preserve image quality and biventricular volumetric accuracy compared to fully sampled acquisitions in patients with pectus excavatum?
p-value: p=0.447
Deep learning-based reconstruction of cine cardiac MRI allows up to 5-fold acceleration without significantly compromising diagnostic image quality or biventricular volumetric measurements.
No takes yet. Share an insight, caveat, or question.
May support faster CMR in pectus excavatum; hypothesis-generating pending prospective validation.
Pednekar et al. (2023) conducted an observational in Pectus excavatum (n=15). Deep learning-based reconstruction (DLR) of undersampled k-space vs. Fully sampled k-space cine bSSFP acquisition was evaluated on Biventricular volumetric indices (p=0.447). Deep learning reconstruction of undersampled cine bSSFP images up to acceleration factor 5 yielded diagnostically adequate image quality and comparable biventricular volumetric indices (P=0.447).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: