{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T07:03:28Z","timestamp":1784099008312,"version":"3.55.0"},"reference-count":44,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T00:00:00Z","timestamp":1784073600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Pediatr."],"abstract":"\n Background<\/jats:title>\n Hypophosphatemia in very preterm and small for gestational age (SGA) infants is common and associated with serious complications. Diagnosis relies on phosphate concentrations, measured in serum or plasma. Machine learning (ML) may enable indirect estimation of serum phosphate based on routinely collected clinical data and results from blood gas analysis.<\/jats:p>\n <\/jats:sec>\n \n Objectives<\/jats:title>\n To (1) describe first week electrolyte trajectories in the ImNuT-trial (Clinicaltrials.gov ID: NCT03555019), and (2) develop and externally validate ML models to estimate serum phosphate levels during the first postnatal week in very preterm infants, using routinely collected clinical and nutritional data.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n We performed a retrospective analysis of 120 infants born &lt;29 weeks' gestation, all managed under a standardized nutritional protocol. Electrolyte trajectories (calcium, potassium, sodium, phosphate) were described using daily means stratified by hypophosphatemia and SGA status. For ML, we defined three regression tasks: concurrent phosphate prediction at blood gas analysis, and forecasts 12\u2005h and 24\u2005h ahead. Twenty-two candidate predictors from blood gases, anthropometry, SGA status and nutrient intakes were subjected to RENT-based variable selection within leave-one-group-out cross-validation. Multiple algorithms (elastic net, gradient boosting, random forests, k-nearest neighbours, kernel ridge, XGBoost) were trained under three preprocessing strategies (complete cases, complete cases without outliers, imputed data). Performance was evaluated with RMSE and\n R<\/jats:italic>\n 2<\/jats:sup>\n using nested cross-validation and an independent external cohort of 40 very preterm infants.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n Among 119 infants with phosphate measurements (607 samples), 33.6% developed hypophosphatemia in week one. SGA infants had a markedly higher and earlier risk of hypophosphatemia compared to appropriate-for-gestational-age infants (log-rank\n p<\/jats:italic>\n \u2009&lt;\u20090.0001). Six predictors (calcium, potassium, gestational age, SGA status, birth weight\n z<\/jats:italic>\n -score, daily weight\n z<\/jats:italic>\n -score) were consistently retained across all prediction tasks. Elastic net models performed best externally on all three prediction tasks. Performance was robust across preprocessing strategies.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n SGA status is a major determinant of early hypophosphatemia risk in very preterm infants. ML models using routinely available clinical and nutritional data can moderately but consistently estimate serum phosphate levels, including 12\u201324\u2005h ahead, and may serve as screening tools to guide nutritional management and reduce unnecessary blood sampling. Prospective validation is warranted before clinical implementation.<\/jats:p>\n <\/jats:sec>","DOI":"10.3389\/fped.2026.1881474","type":"journal-article","created":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T06:03:23Z","timestamp":1784095403000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Predicting serum phosphate levels in very preterm infants using machine learning"],"prefix":"10.3389","volume":"14","author":[{"given":"\u00c5sbj\u00f8rn S.","family":"Westvik","sequence":"first","affiliation":[{"name":"Institute of Clinical Medicine, University of Oslo","place":["Oslo, Norway"]},{"name":"Department of Neonatal Intensive Care, Oslo University Hospital","place":["Oslo, Norway"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Oliver","family":"Tomic","sequence":"additional","affiliation":[{"name":"Faculty of Science and Technology, Norwegian University of Life Sciences","place":["\u00c5s, Norway"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Charlotte","family":"Tscherning","sequence":"additional","affiliation":[{"name":"Institute of Clinical Medicine, University of Oslo","place":["Oslo, Norway"]},{"name":"Department of Neonatal Intensive Care, Oslo University Hospital","place":["Oslo, Norway"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sissel J.","family":"Moltu","sequence":"additional","affiliation":[{"name":"Department of Neonatal Intensive Care, Oslo University Hospital","place":["Oslo, Norway"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"207","DOI":"10.1016\/j.clnu.2012.09.004","article-title":"Enhanced feeding in very-low-birth-weight infants may cause electrolyte disturbances and septicemia\u2013a randomized, controlled trial","volume":"32","author":"Moltu","year":"2013","journal-title":"Clin Nutr"},{"key":"B2","doi-asserted-by":"publisher","first-page":"717","DOI":"10.1038\/jp.2013.28","article-title":"Refeeding syndrome in very-low-birth-weight intrauterine growth-restricted neonates","volume":"33","author":"Ross","year":"2013","journal-title":"J Perinatol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"65","DOI":"10.1002\/jpen.1934","article-title":"Neonatal refeeding syndrome and clinical outcome in extremely low-birth-weight babies: secondary cohort analysis from the ProVIDe trial","volume":"45","author":"Cormack","year":"2021","journal-title":"J Parenter Enteral Nutr"},{"key":"B4","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1136\/archdischild-2024-327029","article-title":"Early hypophosphataemia and refeeding syndrome in extremely low birthweight babies and outcomes to 2 years of age: secondary cohort analysis from the ProVIDe trial","volume":"110","author":"Ford","year":"2025","journal-title":"Arch Dis Child Fetal Neonatal Ed"},{"key":"B5","doi-asserted-by":"publisher","first-page":"771","DOI":"10.5863\/1551-6776-26.8.771","article-title":"Refeeding syndrome in the neonatal intensive care unit","volume":"26","author":"Bradford","year":"2021","journal-title":"J Pediatr Pharmacol Ther"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1529","DOI":"10.1007\/s00431-020-03622-7","article-title":"Factors associated with early phosphate levels in preterm infants","volume":"179","author":"Yakubovich","year":"2020","journal-title":"Eur J Pediatr"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1077","DOI":"10.1111\/apa.13093","article-title":"Extremely preterm infants who are small for gestational age have a high risk of early hypophosphatemia and hypokalemia","volume":"104","author":"Boubred","year":"2015","journal-title":"Acta Paediatr"},{"key":"B8","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1016\/j.jpeds.2017.12.059","article-title":"Consensus based definition of growth restriction in the newborn","volume":"196","author":"Beune","year":"2018","journal-title":"J Pediatr"},{"key":"B9","doi-asserted-by":"publisher","first-page":"248","DOI":"10.1097\/MPG.0000000000003642","article-title":"Enteral nutrition in preterm infants (2022): a position paper from the ESPGHAN committee on nutrition and invited experts","volume":"76","author":"Embleton","year":"2023","journal-title":"J Pediatr Gastroenterol Nutr"},{"key":"B10","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1159\/000260135","article-title":"Fatal effects of a neonatal high-protein diet in low-birth-weight piglets used as a model of intrauterine growth restriction","volume":"97","author":"Jamin","year":"2010","journal-title":"Neonatology"},{"key":"B11","doi-asserted-by":"publisher","first-page":"101635","DOI":"10.1016\/j.siny.2025.101635","article-title":"Parenteral nutrition for preterm infants: benefits and risks in 2025","volume":"30","author":"Embleton","year":"2025","journal-title":"Semin Fetal Neonatal Med"},{"key":"B12","doi-asserted-by":"publisher","first-page":"460","DOI":"10.1002\/jpen.2739","article-title":"Association between early phosphate intake and refeeding syndrome in extremely low-birth-weight infants: a retrospective cohort study","volume":"49","author":"Wright","year":"2025","journal-title":"J Parenter Enteral Nutr"},{"key":"B13","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1186\/s12887-020-02425-x","article-title":"Effects of nutrition therapy on growth, inflammation and metabolism in immature infants: a study protocol of a double-blind randomized controlled trial (ImNuT)","volume":"21","author":"Wendel","year":"2021","journal-title":"BMC Pediatr"},{"key":"B14","doi-asserted-by":"publisher","first-page":"251","DOI":"10.1016\/j.clnesp.2022.12.024","article-title":"A standardized feeding protocol ensured recommended nutrient intakes and prevented growth faltering in preterm infants <29 weeks gestation","volume":"53","author":"Rossholt","year":"2023","journal-title":"Clin Nutr ESPEN"},{"key":"B15","doi-asserted-by":"publisher","first-page":"176","DOI":"10.1016\/j.clnu.2023.11.037","article-title":"Arachidonic and docosahexaenoic acid supplementation and brain maturation in preterm infants; a double blind RCT","volume":"43","author":"Moltu","year":"2024","journal-title":"Clin Nutr"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1379","DOI":"10.1002\/ncp.10986","article-title":"Quality improvement project in a neonatal intensive care unit reduced the prevalence and duration of hypophosphatemia with significant and sustainable results","volume":"38","author":"Lair","year":"2023","journal-title":"Nutr Clin Pract"},{"key":"B17","doi-asserted-by":"publisher","first-page":"2660","DOI":"10.1080\/14767058.2019.1670800","article-title":"Iatrogenic blood loss in extreme preterm infants due to frequent laboratory tests and procedures","volume":"34","author":"Counsilman","year":"2021","journal-title":"J Matern Fetal Neonatal Med"},{"key":"B18","doi-asserted-by":"publisher","first-page":"191","DOI":"10.17219\/acem\/32065","article-title":"Prevention of iatrogenic Anemia in critical and neonatal care","volume":"25","author":"Jakacka","year":"2016","journal-title":"Adv Clin Exp Med"},{"key":"B19","doi-asserted-by":"publisher","first-page":"549","DOI":"10.1016\/j.cnc.2018.07.013","article-title":"Neonatal pain: perceptions and current practice","volume":"30","author":"Perry","year":"2018","journal-title":"Crit Care Nurs Clin North Am"},{"key":"B20","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1016\/j.earlhumdev.2017.03.003","article-title":"The impact of cumulative pain\/stress on neurobehavioral development of preterm infants in the NICU","volume":"108","author":"Cong","year":"2017","journal-title":"Early Hum Dev"},{"key":"B21","doi-asserted-by":"publisher","first-page":"57","DOI":"10.2217\/pmt.13.61","article-title":"Early repetitive pain in preterm infants in relation to the developing brain","volume":"4","author":"Ranger","year":"2014","journal-title":"Pain Manag"},{"key":"B22","year":""},{"key":"B23","doi-asserted-by":"publisher","first-page":"2582","DOI":"10.3390\/diagnostics13152582","article-title":"What is machine learning, artificial neural networks and deep learning?\u2014examples of practical applications in medicine","volume":"13","author":"Kufel","year":"2023","journal-title":"Diagnostics"},{"key":"B24","doi-asserted-by":"publisher","first-page":"416","DOI":"10.1038\/s41372-025-02445-4","article-title":"AI to predict extrauterine growth restriction during transitional nutrition of preterm infants: a retrospective study","volume":"46","author":"Bozzetti","year":"2026","journal-title":"J Perinatol"},{"key":"B25","doi-asserted-by":"publisher","first-page":"2580067","DOI":"10.1080\/0886022X.2025.2580067","article-title":"Interpretable machine learning for the early prediction of acute kidney injury in preterm infants in the neonatal intensive care unit","volume":"47","author":"Liu","year":"2025","journal-title":"Renal Fail"},{"key":"B26","doi-asserted-by":"publisher","first-page":"1282","DOI":"10.3390\/diagnostics16091282","article-title":"Machine-Learning models outperform clinicians in predicting postnatal growth failure among very low birth weight infants","volume":"16","author":"Lim","year":"2026","journal-title":"Diagnostics"},{"key":"B27","doi-asserted-by":"publisher","first-page":"464","DOI":"10.1186\/s12884-026-08955-z","article-title":"Prediction of preterm and low birth weight risk using a physiology based artificial neural network integrating hematological, dental, and periodontal index markers: a cross sectional study based on machine learning","volume":"26","author":"Temur","year":"2026","journal-title":"BMC Pregnancy Childbirth"},{"key":"B28","doi-asserted-by":"publisher","first-page":"e72880","DOI":"10.1371\/journal.pone.0072880","article-title":"Initial amino acid intake influences phosphorus and calcium homeostasis in preterm infants\u2013it is time to change the composition of the early parenteral nutrition","volume":"8","author":"Bonsante","year":"2013","journal-title":"PLoS One"},{"key":"B29","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1016\/j.clnu.2022.11.012","article-title":"Effect of arachidonic and docosahexaenoic acid supplementation on respiratory outcomes and neonatal morbidities in preterm infants","volume":"42","author":"Wendel","year":"2023","journal-title":"Clin Nutr"},{"key":"B30","doi-asserted-by":"publisher","first-page":"8","DOI":"10.1186\/1471-2431-8-8","article-title":"Continuous growth reference from 24th week of gestation to 24 months by gender","volume":"8","author":"Niklasson","year":"2008","journal-title":"BMC Pediatr"},{"key":"B31","doi-asserted-by":"publisher","first-page":"152333","DOI":"10.1109\/ACCESS.2021.3126429","article-title":"RENT\u2014repeated elastic net technique for feature selection","volume":"9","author":"Jenul","year":"2021","journal-title":"IEEE Access"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1026","DOI":"10.21105\/joss.01026","article-title":"Pingouin: statistics in python","volume":"3","author":"Vallat","year":"2018","journal-title":"J Open Source Softw"},{"key":"B33","author":"Wilson","year":""},{"key":"B34","first-page":"2623","article-title":"Optuna: a next-generation hyperparameter optimization framework","author":"Akiba","year":"2019"},{"key":"B35","doi-asserted-by":"publisher","first-page":"440","DOI":"10.1034\/j.1600-0412.2000.079006440.x","article-title":"Birthweight by gestational age in Norway","volume":"79","author":"Skjaerven","year":"2000","journal-title":"Acta Obstet Gynecol Scand"},{"key":"B36","volume-title":"R: A Language and Environment for Statistical Computing","year":"2021"},{"key":"B37","author":"Therneau","year":""},{"key":"B38","first-page":"2825","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"B39","author":"Kassambara","year":"2026"},{"key":"B40","doi-asserted-by":"publisher","first-page":"90","DOI":"10.1109\/MCSE.2007.55","article-title":"Matplotlib: a 2D graphics environment","volume":"9","author":"Hunter","year":"2007","journal-title":"Comput Sci Eng"},{"key":"B41","doi-asserted-by":"publisher","first-page":"3021","DOI":"10.21105\/joss.03021","article-title":"Seaborn: statistical data visualization","volume":"6","author":"Waskom","year":"2021","journal-title":"J Open Source Softw"},{"key":"B42","doi-asserted-by":"publisher","first-page":"e53","DOI":"10.1097\/MPG.0000000000003532","article-title":"Higher parenteral electrolyte intakes in preterm infants during first week of life: effects on electrolyte imbalances","volume":"75","author":"Sp\u00e4th","year":"2022","journal-title":"J Pediatr Gastroenterol Nutr"},{"key":"B43","doi-asserted-by":"publisher","first-page":"58","DOI":"10.1002\/jpr3.12029","article-title":"High protein intake can lead to serious hypophosphatemia and hypokalemia in growth restricted preterm newborns","volume":"5","author":"Kor\u010dek","year":"2024","journal-title":"JPGN Rep"},{"key":"B44","doi-asserted-by":"publisher","first-page":"841","DOI":"10.17219\/acem\/70081","article-title":"Early hypophosphatemia in very low birth weight preterm infants","volume":"27","author":"Paj\u0105k","year":"2018","journal-title":"Adv Clin Exp Med"}],"container-title":["Frontiers in Pediatrics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fped.2026.1881474\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T06:03:24Z","timestamp":1784095404000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fped.2026.1881474\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,15]]},"references-count":44,"alternative-id":["10.3389\/fped.2026.1881474"],"URL":"https:\/\/doi.org\/10.3389\/fped.2026.1881474","relation":{},"ISSN":["2296-2360"],"issn-type":[{"value":"2296-2360","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,15]]},"article-number":"1881474"}}