{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,24]],"date-time":"2026-01-24T01:46:33Z","timestamp":1769219193659,"version":"3.49.0"},"reference-count":29,"publisher":"Wiley","issue":"8","license":[{"start":{"date-parts":[[2020,6,3]],"date-time":"2020-06-03T00:00:00Z","timestamp":1591142400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2020,8]]},"abstract":"<jats:sec><jats:title>Purpose<\/jats:title><jats:p>T<jats:sub>1<\/jats:sub>\u2010weighted dynamic contrast\u2010enhanced Magnetic Resonance Imaging (DCE\u2010MRI) is typically quantified by least squares (LS) fitting to a pharmacokinetic (PK) model to yield parameters of microvasculature and perfusion in normal and disease tissues. Such fitting is both time\u2010consuming as well as subject to inaccuracy and instability in parameter estimates. Here, we propose a novel neural network approach to estimate the PK parameters by extracting long and short time\u2010dependent features in DCE\u2010MRI.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>A Long Short\u2010Term Memory (LSTM) network, widely used for processing sequence data, was employed to map DCE\u2010MRI time\u2010series accompanied with an arterial input function to parameters of the extended Tofts model. Head and neck DCE\u2010MRI from 103 patients were used for training and testing the LSTM model. Arterial input functions (AIFs) from 78 patients were used to generate synthetic DCE\u2010MRI time\u2010series for training, during which data augmentation was used to overcome the limited size of <jats:italic>in vivo<\/jats:italic> data. The model was tested on independent synthesized DCE data using AIFs from 25 patients. The LSTM performance was optimized for the numbers of layers and hidden state features. The performance of the LSTM was tested for different temporal resolution, total acquisition time, and contrast\u2010to\u2010noise ratio (CNR), and compared to the conventional LS fitting and a CNN\u2010based method.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Compared to LS fitting, the LSTM model had comparable accuracy in PK parameter estimations from fully temporal\u2010sampled DCE\u2010MRI data (~3\u00a0s per frame), but much better accuracy for the data with temporally subsampling (4s or greater per frame), total acquisition time truncation by 48%\u201016%, or low CNR (5 and 10). The LSTM reduced normalized root mean squared error by 40.4%, 46.9%, and 53.0% for sampling intervals of 4s, 5s, and 6s, respectively, compared to LS fitting. Compared to the CNN model, the LSTM model reduced the error in the parameter estimates up to 55.2%. Also, the LSTM improved the inference time by\u00a0~\u00a014 times on CPU compared to LS fitting.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>Our study suggests that the LSTM model could achieve improved robustness and computation speed for PK parameter estimation compared to LS fitting and the CNN based network, particularly for suboptimal data.<\/jats:p><\/jats:sec>","DOI":"10.1002\/mp.14222","type":"journal-article","created":{"date-parts":[[2020,5,7]],"date-time":"2020-05-07T20:24:13Z","timestamp":1588883053000},"page":"3447-3457","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":44,"title":["Estimation of pharmacokinetic parameters from DCE\u2010MRI by extracting long and short time\u2010dependent features using an LSTM network"],"prefix":"10.1002","volume":"47","author":[{"given":"Jiaren","family":"Zou","sequence":"first","affiliation":[{"name":"Department of Radiation Oncology University of Michigan Ann Arbor MI 48109 USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"James M.","family":"Balter","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of Michigan Ann Arbor MI 48109 USA"},{"name":"Department of Biomedical Engineering University of Michigan Ann Arbor MI 48109 USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yue","family":"Cao","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology University of Michigan Ann Arbor MI 48109 USA"},{"name":"Department of Biomedical Engineering University of Michigan Ann Arbor MI 48109 USA"},{"name":"Department of Radiology University of Michigan Ann Arbor MI 48109 USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2020,6,3]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2008.08.024"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.3174\/ajnr.A4458"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2019.01118"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.18383\/j.tom.2018.00035"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4898202"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.21391"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1002\/(SICI)1522-2594(199911)42:5<952::AID-MRM16>3.0.CO;2-S"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.24751"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.acra.2009.10.021"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.20022"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.25659"},{"key":"e_1_2_8_13_1","doi-asserted-by":"crossref","unstructured":"UlasC TettehG ThrippletonMJ et al.Direct Estimation of Pharmacokinetic Parameters from DCE\u2010MRI using Deep CNN with Forward Physical Model Loss. 2018;11070 LNCS:39\u201047.","DOI":"10.1007\/978-3-030-00928-1_5"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.3389\/fneur.2018.01147"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2599174"},{"key":"e_1_2_8_17_1","unstructured":"ChenJ YangL ZhangY AlberM ChenDZ.Combining fully convolutional and recurrent neural networks for 3D biomedical image segmentation. In: Advances in Neural Information Processing Systems.; 2016:3044\u20133052.http:\/\/arxiv.org\/abs\/1609.01006."},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1117\/1.JMI.5.1.014503"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.1880070113"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.mric.2009.01.007"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.1999737"},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1002\/jmri.21624"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/aafce7"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.1910360209"},{"key":"e_1_2_8_26_1","unstructured":"D Robben P Suetens Perfusion parameter estimation using neural networks and data augmentation"},{"key":"e_1_2_8_27_1","unstructured":"2015 DP Kingma BJL Adam Adam: A method for stochastic optimization"},{"key":"e_1_2_8_28_1","volume-title":"Proceedings \u2010 International Symposium on Biomedical Imaging","author":"Arbelle A"},{"key":"e_1_2_8_29_1","volume-title":"2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)","author":"Zhang D"},{"key":"e_1_2_8_30_1","first-page":"1","article-title":"GAN and dual\u2010input two\u2010compartment model\u2010based training of a neural network for robust quantification of contrast uptake rate in gadoxetic acid\u2010enhanced MRI","author":"Simeth J","year":"2020","journal-title":"Med Phys"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1002%2Fmp.14222","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.14222","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.14222","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.14222","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,23]],"date-time":"2023-08-23T07:56:40Z","timestamp":1692777400000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.14222"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,6,3]]},"references-count":29,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2020,8]]}},"alternative-id":["10.1002\/mp.14222"],"URL":"https:\/\/doi.org\/10.1002\/mp.14222","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,6,3]]},"assertion":[{"value":"2019-11-25","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-04-29","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-06-03","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}