{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T10:54:52Z","timestamp":1790765692182,"version":"4.1.0"},"posted":{"date-parts":[[2026,9,27]]},"group-title":"Cardiovascular Medicine","reference-count":45,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,9,27]],"date-time":"2026-09-27T00:00:00Z","timestamp":1790467200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"name":"NIH\/NHLBI","award":["R01 HL162671"],"award-info":[{"award-number":["R01 HL162671"]}]},{"name":"NIH\/NHLBI","award":["K23 HL163456"],"award-info":[{"award-number":["K23 HL163456"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,9,27]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:sec>\n                  <jats:title>Purpose<\/jats:title>\n                  <jats:p>To develop and evaluate a recurrent neural network that operates directly on undersampled golden-angle radial k-space, enabling low-latency cardiac cine reconstruction for interventional cardiac MRI (iCMR).<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>KOSTAL (K-space Optimized Spatial Temporal Architecture for Low latency) combines a per-coil k-space convolutional gated recurrent unit (convGRU), which learns a mask-conditioned view-sharing rule over the accumulated k-space, with an image-space recurrent module. It was trained and tested on the open-access multi-coil OCMR dataset using 10 simulated golden-angle spokes per frame against a 1000-spoke reference. Ablations varied recurrent module architecture and placement; temporal behavior was assessed using a imaging-plane transition test. Generalization was evaluated without retraining in 23 pediatric patients prospectively scanned during free-breathing using a radial golden-angle bSSFP sequence with binned XD-GRASP and clinical breath-held cine serving as references.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>At 10 spokes per frame, KOSTAL reached a median SSIM of 0.907 (gridding: 0.622) in under 34 ms of processing time per frame. Image quality (IQ) matched gridded results formed from 7-11 times more spokes. Recurrence in k-space rather than image space accounted for most of this gain (0.891 versus \u22640.650). IQ recovered from an abrupt plane transition in 75 ms, versus 303 ms for 200-spoke gridding. KOSTAL had the best agreement with the binned reference and the highest long-axis CNR. KOSTAL accurately tracked left-ventricular blood-pool volume (end-diastolic volume bias-0.7 mL) with a shorter lag (71 ms) than iterative reconstruction (239-287 ms).<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>Learned, mask-conditioned recurrence in k-space yields high image quality at a latency compatible with interventional guidance.<\/jats:p>\n                <\/jats:sec>","DOI":"10.64898\/2026.09.21.26363257","type":"posted-content","created":{"date-parts":[[2026,9,27]],"date-time":"2026-09-27T21:35:12Z","timestamp":1790544912000},"source":"Crossref","is-referenced-by-count":0,"title":["K-space Optimized Spatial Temporal Architecture for Low Latency in Interventional Cardiac MRI"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6938-3474","authenticated-orcid":false,"given":"Mark","family":"Buckup","sequence":"first","affiliation":[{"name":"Department of Bioengineering, Jacobs School of Engineering, UC San Diego"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Niraj","family":"Mahajan","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Jacobs School of Engineering, UC San Diego"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9544-547X","authenticated-orcid":false,"given":"Ana E","family":"Rodr\u00edguez-Soto","sequence":"additional","affiliation":[{"name":"Department of Bioengineering, Jacobs School of Engineering, UC San Diego"},{"name":"Department of Radiology, School of Medicine, UC San Diego"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Joni","family":"Blood","sequence":"additional","affiliation":[{"name":"Rady Children\u2019s Hospital San Diego, Division of Cardiology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7396-288X","authenticated-orcid":false,"given":"Sanjeet","family":"Hegde","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, School of Medicine, UC San Diego"},{"name":"Rady Children\u2019s Hospital San Diego, Division of Cardiology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9106-8486","authenticated-orcid":false,"given":"Eleanor L","family":"Schuchardt","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, School of Medicine, UC San Diego"},{"name":"Rady Children\u2019s Hospital San Diego, Division of Cardiology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3602-8411","authenticated-orcid":false,"given":"Beth","family":"Printz","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, School of Medicine, UC San Diego"},{"name":"Rady Children\u2019s Hospital San Diego, Division of Cardiology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hari K","family":"Narayan","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, School of Medicine, UC San Diego"},{"name":"Rady Children\u2019s Hospital San Diego, Division of Cardiology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7671-3766","authenticated-orcid":false,"given":"Brent L","family":"Gordon","sequence":"additional","affiliation":[{"name":"Department of Pediatrics, School of Medicine, UC San Diego"},{"name":"Rady Children\u2019s Hospital San Diego, Division of Cardiology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9616-3274","authenticated-orcid":false,"given":"Francisco","family":"Contijoch","sequence":"additional","affiliation":[{"name":"Department of Bioengineering, Jacobs School of Engineering, UC San Diego"},{"name":"Department of Radiology, School of Medicine, UC San Diego"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026093002550536000_2026.09.21.26363257v1.1","doi-asserted-by":"crossref","unstructured":"Ismail, T. F. et al. Cardiac MR: From Theory to Practice. Front. Cardiovasc. Med. 9, 826283 (2022).","DOI":"10.3389\/fcvm.2022.826283"},{"key":"2026093002550536000_2026.09.21.26363257v1.2","doi-asserted-by":"crossref","unstructured":"Ferreira, P. F. , Gatehouse, P. D. , Mohiaddin, R. H. & Firmin, D. N . Cardiovascular magnetic resonance artefacts. J. Cardiovasc. Magn. Reson. 15, 41 (2013).","DOI":"10.1186\/1532-429X-15-41"},{"key":"2026093002550536000_2026.09.21.26363257v1.3","doi-asserted-by":"publisher","DOI":"10.1002\/(SICI)1522-2594(199911)42:5<952::AID-MRM16>3.0.CO;2-S"},{"key":"2026093002550536000_2026.09.21.26363257v1.4","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.10171"},{"key":"2026093002550536000_2026.09.21.26363257v1.5","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.21391"},{"key":"2026093002550536000_2026.09.21.26363257v1.6","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.28187"},{"key":"2026093002550536000_2026.09.21.26363257v1.7","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.21236"},{"key":"2026093002550536000_2026.09.21.26363257v1.8","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.24389"},{"key":"2026093002550536000_2026.09.21.26363257v1.9","doi-asserted-by":"crossref","first-page":"1015","DOI":"10.1002\/mrm.25213","article-title":"Distributed MRI Reconstruction using Gadgetron based Cloud Computing","volume":"73","year":"2015","journal-title":"Magn. Reson. Med"},{"key":"2026093002550536000_2026.09.21.26363257v1.10","doi-asserted-by":"crossref","first-page":"935","DOI":"10.1002\/jmri.25749","article-title":"Real-time MRI guidance of cardiac interventions","volume":"46","year":"2017","journal-title":"J. Magn. Reson. Imaging JMRI"},{"key":"2026093002550536000_2026.09.21.26363257v1.11","doi-asserted-by":"crossref","first-page":"419","DOI":"10.1007\/s11886-022-01659-8","article-title":"MRI-Guided Cardiac Catheterization in Congenital Heart Disease: How to Get Started","volume":"24","year":"2022","journal-title":"Curr. Cardiol. Rep"},{"key":"2026093002550536000_2026.09.21.26363257v1.12","doi-asserted-by":"crossref","first-page":"767","DOI":"10.1002\/mrm.22463","article-title":"Combination of Compressed Sensing and Parallel Imaging for Highly Accelerated First-Pass Cardiac Perfusion MRI","volume":"64","year":"2010","journal-title":"Magn. Reson. Med. Off. J. Soc. Magn. Reson. Med. Soc. Magn. Reson. Med"},{"key":"2026093002550536000_2026.09.21.26363257v1.13","doi-asserted-by":"crossref","unstructured":"Sufian, M. A. & Niu, M . Hybrid deep learning for computational precision in cardiac MRI segmentation: Integrating Autoencoders, CNNs, and RNNs for enhanced structural analysis. Comput. Biol. Med. 186, 109597 (2025).","DOI":"10.1016\/j.compbiomed.2024.109597"},{"key":"2026093002550536000_2026.09.21.26363257v1.14","doi-asserted-by":"crossref","unstructured":"Wang, Z. , She, H. , Zhang, Y. & Du, Y. P . Parallel non-Cartesian spatial-temporal dictionary learning neural networks (stDLNN) for accelerating 4D-MRI. Med. Image Anal. 84, 102701 (2023).","DOI":"10.1016\/j.media.2022.102701"},{"key":"2026093002550536000_2026.09.21.26363257v1.15","doi-asserted-by":"crossref","unstructured":"Hossain, Md . B., Shinde , R. K., Oh , S., Kwon , K.-C. & Kim , N. A Systematic Review and Identification of the Challenges of Deep Learning Techniques for Undersampled Magnetic Resonance Image Reconstruction. Sensors 24, 753 (2024).","DOI":"10.3390\/s24030753"},{"key":"2026093002550536000_2026.09.21.26363257v1.16","doi-asserted-by":"crossref","first-page":"266","DOI":"10.1002\/mrm.29855","article-title":"HyperSLICE: HyperBand optimized spiral for low-latency interactive cardiac examination","volume":"91","year":"2024","journal-title":"Magn. Reson. Med"},{"key":"2026093002550536000_2026.09.21.26363257v1.17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00143"},{"key":"2026093002550536000_2026.09.21.26363257v1.18","doi-asserted-by":"crossref","first-page":"1195","DOI":"10.1002\/mrm.28485","article-title":"Multi-domain convolutional neural network (MDCNN) for radial reconstruction of dynamic cardiac MRI","volume":"85","year":"2021","journal-title":"Magn. Reson. Med"},{"key":"2026093002550536000_2026.09.21.26363257v1.19","doi-asserted-by":"crossref","unstructured":"Ratnayaka, K. et al. Radiation-free CMR diagnostic heart catheterization in children. J. Cardiovasc. Magn. Reson. 19, 65 (2017).","DOI":"10.1186\/s12968-017-0374-2"},{"key":"2026093002550536000_2026.09.21.26363257v1.20","unstructured":"Lipton, Z. C. , Berkowitz, J. & Elkan, C. A Critical Review of Recurrent Neural Networks for Sequence Learning. arXiv.org https:\/\/arxiv.org\/abs\/1506.00019v4 (2015)."},{"key":"2026093002550536000_2026.09.21.26363257v1.21","doi-asserted-by":"crossref","first-page":"193","DOI":"10.1002\/mp.14566","article-title":"A k-space-to-image reconstruction network for MRI using recurrent neural network","volume":"48","year":"2021","journal-title":"Med. Phys"},{"key":"2026093002550536000_2026.09.21.26363257v1.22","doi-asserted-by":"crossref","unstructured":"Alshirbaji, T. A. , Jalal, N. A. & M\u00f6ller, K . A convolutional neural network with a two-stage LSTM model for tool presence detection in laparoscopic videos. Curr. Dir. Biomed. Eng. 6, (2020).","DOI":"10.1515\/cdbme-2020-0002"},{"key":"2026093002550536000_2026.09.21.26363257v1.23","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"2026093002550536000_2026.09.21.26363257v1.24","doi-asserted-by":"publisher","DOI":"10.1109\/tmi.2017.2760978"},{"key":"2026093002550536000_2026.09.21.26363257v1.25","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1412.3555"},{"key":"2026093002550536000_2026.09.21.26363257v1.26","doi-asserted-by":"crossref","first-page":"68","DOI":"10.1016\/j.media.2019.04.011","article-title":"GRUU-Net: Integrated convolutional and gated recurrent neural network for cell segmentation","volume":"56","year":"2019","journal-title":"Med. Image Anal"},{"key":"2026093002550536000_2026.09.21.26363257v1.27","doi-asserted-by":"crossref","first-page":"10612","DOI":"10.1109\/TNNLS.2022.3169569","article-title":"A Deep Convolutional Gated Recurrent Unit for CT Image Reconstruction","volume":"34","year":"2023","journal-title":"IEEE Trans. Neural Netw. Learn. Syst"},{"key":"2026093002550536000_2026.09.21.26363257v1.28","doi-asserted-by":"crossref","unstructured":"Singh, D. et al. Emerging Trends in Fast MRI Using Deep-Learning Reconstruction on Undersampled k-Space Data: A Systematic Review. Bioengineering 10, 1012 (2023).","DOI":"10.3390\/bioengineering10091012"},{"key":"2026093002550536000_2026.09.21.26363257v1.29","doi-asserted-by":"crossref","first-page":"918","DOI":"10.1002\/jmri.1880060613","article-title":"Application of keyhole imaging to interventional MRI: A simulation study to predict sequence requirements","volume":"6","year":"1996","journal-title":"J. Magn. Reson. Imaging"},{"key":"2026093002550536000_2026.09.21.26363257v1.30","doi-asserted-by":"crossref","first-page":"231","DOI":"10.1002\/jmri.1880040223","article-title":"On the \u201ckeyhole\u201d technique","volume":"4","year":"1994","journal-title":"J. Magn. Reson. Imaging"},{"key":"2026093002550536000_2026.09.21.26363257v1.31","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.1880030419"},{"key":"2026093002550536000_2026.09.21.26363257v1.32","doi-asserted-by":"publisher","DOI":"10.1002\/1522-2594(200012)44:6<825::AID-MRM2>3.0.CO;2-D"},{"key":"2026093002550536000_2026.09.21.26363257v1.33","unstructured":"Chen, C. , et al. OCMR (v1.0)--Open-Access Multi-Coil k-Space Dataset for Cardiovascular Magnetic Resonance Imaging. arXiv.org https:\/\/arxiv.org\/abs\/2008.03410v2 (2020)."},{"key":"2026093002550536000_2026.09.21.26363257v1.34","doi-asserted-by":"crossref","first-page":"133","DOI":"10.1016\/j.mri.2007.04.010","article-title":"A software channel compression technique for faster reconstruction with many channels","volume":"26","year":"2008","journal-title":"Magn. Reson. Imaging"},{"key":"2026093002550536000_2026.09.21.26363257v1.35","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.21237"},{"key":"2026093002550536000_2026.09.21.26363257v1.36","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2212.08479"},{"key":"2026093002550536000_2026.09.21.26363257v1.37","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.28706"},{"key":"2026093002550536000_2026.09.21.26363257v1.38","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.25665"},{"key":"2026093002550536000_2026.09.21.26363257v1.39","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2504.03600"},{"key":"2026093002550536000_2026.09.21.26363257v1.40","doi-asserted-by":"crossref","first-page":"1904","DOI":"10.1002\/mrm.28834","article-title":"Real-time deep artifact suppression using recurrent U-Nets for low-latency cardiac MRI","volume":"86","year":"2021","journal-title":"Magn. Reson. Med"},{"key":"2026093002550536000_2026.09.21.26363257v1.41","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2018.2863670"},{"key":"2026093002550536000_2026.09.21.26363257v1.42","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1904.09390"},{"key":"2026093002550536000_2026.09.21.26363257v1.43","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2111.09639"},{"key":"2026093002550536000_2026.09.21.26363257v1.44","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2203.14769"},{"key":"2026093002550536000_2026.09.21.26363257v1.45","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.mri.2020.06.008","article-title":"Rapid 3D whole-heart cine imaging using golden ratio stack of spirals","volume":"72","year":"2020","journal-title":"Magn. Reson. Imaging"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.09.21.26363257","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T09:55:18Z","timestamp":1790762118000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.64898\/2026.09.21.26363257"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,27]]},"references-count":45,"URL":"https:\/\/doi.org\/10.64898\/2026.09.21.26363257","relation":{},"subject":[],"published":{"date-parts":[[2026,9,27]]},"subtype":"preprint"}}