{"doi":"10.1016/j.mri.2024.04.027","title":"Rapid 2D 23Na MRI of the calf using a denoising convolutional neural network","abstract":null,"journal":"Magnetic Resonance Imaging","year":2024,"id":617319,"datarank":0.4159486598934356,"base_score":2.0794415416798357,"endowment":2.0794415416798357,"self_citation_contribution":0.31191623125197543,"citation_network_contribution":0.10403242864146016,"self_endowment_contribution":0.31191623125197543,"citer_contribution":0.10403242864146016,"corpus_percentile":null,"corpus_rank":null,"citation_count":7,"citer_count":6,"citers_with_citation_signal":2,"citers_with_endowment":2,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":null,"is_data_producer":false,"deposit_databanks":null,"is_oa":false,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":null,"fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":850330,"name":"Vivek Muthurangu","orcid":null,"position":1,"is_corresponding":false},{"id":1591968,"name":"Marilena Rega","orcid":null,"position":2,"is_corresponding":false},{"id":109008,"name":"Stephen B. Walsh","orcid":"0000-0002-8693-1353","position":3,"is_corresponding":false},{"id":280608,"name":"Jennifer A. Steeden","orcid":"0000-0002-9792-2022","position":4,"is_corresponding":false},{"id":1591967,"name":"Rebecca R. Baker","orcid":null,"position":0,"is_corresponding":false}],"reference_count":0,"raw_metadata":{"has_enrichment":true,"resolved":true,"title":"Rapid 2D 23Na MRI of the calf using a denoising convolutional neural network","abstract":"23Na MRI can be used to quantify in-vivo tissue sodium concentration (TSC), but the inherently low 23Na signal leads to long scan times and/or noisy or low-resolution images. Reconstruction algorithms such as compressed sensing (CS) have been proposed to mitigate low signal-to-noise ratio (SNR); although, these can result in unnatural images, suboptimal denoising and long processing times. Recently, machine learning has been increasingly used to denoise 1H MRI acquisitions; however, this approach typically requires large volumes of high-quality training data, which is not readily available for 23Na MRI. Here, we propose using 1H data to train a denoising convolutional neural network (CNN), which we subsequently demonstrate on prospective 23Na images of the calf. 1893 1H fat-saturated transverse slices of the knee from the open-source fastMRI dataset were used to train denoising CNNs for different levels of noise. Synthetic low SNR images were generated by adding gaussian noise to the high-quality 1H k-space data before reconstruction to create paired training data. For prospective testing, 23Na images of the calf were acquired in 10 healthy volunteers with a total of 150 averages over ten minutes, which were used as a reference throughout the study. From this data, images with fewer averages were retrospectively reconstructed using a non-uniform fast Fourier transform (NUFFT) as well as CS, with the NUFFT images subsequently denoised using the trained CNN. CNNs were successfully applied to 23Na images reconstructed with 50, 40 and 30 averages. Muscle and skin apparent TSC quantification from CNN-denoised images were equivalent to those from CS images, with <0.9 mM bias compared to reference values. Estimated SNR was significantly higher in CNN-denoised images compared to NUFFT, CS and reference images. Quantitative edge sharpness was equivalent for all images. For subjective image quality ranking, CNN-denoised images ranked equally best with reference images and significantly better than NUFFT and CS images. Denoising CNNs trained on 1H data can be successfully applied to 23Na images of the calf; thus, allowing scan time to be reduced from ten minutes to two minutes with little impact on image quality or apparent TSC quantification accuracy.","is_dataset_classified":null,"base_score":1.9459101490553132,"endowment":1.9459101490553132,"datacite_reuse_total":0,"file_count":0,"downloads":0,"views":0,"has_version_chain":false,"is_dataset":false,"is_oa":false,"pmid":"38642779","pmcid":null,"openalex_id":"https://openalex.org/W4394910220","authors":[],"funders":[{"funder_name":"Michael J Fox Foundation for Parkinson's Research","grant_id":"MJFF-021438","title":null},{"funder_name":"UKRI Medical Research Council","grant_id":"MR/S032290/1","title":"Towards 10-minute Magnetic Resonance Scanning in Children - Developing Accelerated Imaging Using Machine Learning"},{"funder_name":"UK Research and Innovation","grant_id":"","title":null}],"total_grants":3,"fwci":1.5847,"citation_percentile":0.82173271,"influential_citations":0,"citation_trend":[{"year":2024,"count":1},{"year":2025,"count":1},{"year":2026,"count":4}],"oa_status":"hybrid","license":"cc-by","oa_locations":[{"url":"https://doi.org/10.1016/j.mri.2024.04.027","host_type":"journal"},{"url":"https://doi.org/10.1016/j.mri.2024.04.027","host_type":"publisher"},{"url":"https://api.elsevier.com/content/article/PII:S0730725X24001401?httpAccept=text/xml","host_type":"publisher"},{"url":"https://api.elsevier.com/content/article/PII:S0730725X24001401?httpAccept=text/plain","host_type":"publisher"},{"url":"https://pubmed.ncbi.nlm.nih.gov/38642779","host_type":"repository"},{"url":"https://discovery.ucl.ac.uk/id/eprint/10191819/","host_type":"repository"},{"url":"https://discovery.ucl.ac.uk/10191819/1/1-s2.0-S0730725X24001401-main.pdf","host_type":"repository"},{"url":"https://dx.doi.org/10.48550/arxiv.2406.02553","host_type":""},{"url":"http://arxiv.org/abs/2406.02553","host_type":""},{"url":"https://discovery-pp.ucl.ac.uk/id/eprint/10191819/","host_type":""}],"fields_of_study":["MRI in cancer diagnosis","Advanced MRI Techniques and Applications","Animal health and immunology","02 engineering and technology","0202 electrical engineering, electronic engineering, information engineering","Magnetic Resonance Imaging","Neural Networks, Computer","Humans","Signal-To-Noise Ratio","Image Processing, Computer-Assisted","Algorithms","Leg","Male","Adult","Female","Sodium Isotopes","Prospective Studies","Sodium","Healthy Volunteers","Muscle, Skeletal"],"mesh_terms":["Adult","Algorithms","Female","Humans","Image Processing, Computer-Assisted","Leg","Magnetic Resonance Imaging","Male","Prospective Studies","Sodium","Sodium Isotopes","Neural Networks, Computer","Muscle, Skeletal","Signal-To-Noise Ratio","Healthy Volunteers"],"keywords":["Noise reduction","Computer science","Convolutional neural network","Artificial intelligence","Pattern recognition (psychology)","Compressed sensing","Noise (video)","Image quality","Reduction (mathematics)","Computer vision","Image (mathematics)","Mathematics","Sodium","Denoising","Machine Learning","(23)Na Mri","X-nuclei","Male","Adult","Leg","FOS: Physical sciences","Signal-To-Noise Ratio","Physics - Medical Physics","Magnetic Resonance Imaging","Healthy Volunteers","Image Processing, Computer-Assisted","Humans","Female","Sodium Isotopes","Neural Networks, Computer","Prospective Studies","Medical Physics (physics.med-ph)","Muscle, Skeletal","Algorithms"],"sdg_mappings":[],"linked_datasets":[],"clinical_trials":[],"software_tools":[],"database_accessions":[],"source":"live","citation_network_status":"fetched"},"created_at":"2026-08-03T01:23:17.725887Z","pmid":null,"pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}