{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T22:14:29Z","timestamp":1787004869649,"version":"build-2736575974"},"reference-count":37,"publisher":"Georg Thieme Verlag KG","issue":"03","license":[{"start":{"date-parts":[[2025,6,2]],"date-time":"2025-06-02T00:00:00Z","timestamp":1748822400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Pharmacopsychiatry"],"published-print":{"date-parts":[[2026,5]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>To develop an optimal model to predict valproic acid (VPA) concentrations by\n                        machine learning, ensuring that the VPA plasma concentration is in the\n                        effective treatment range, and thus effectively control the patient\u2019s\n                        epilepsy.<\/jats:p>\n                  <jats:p>This single-center, retrospective study included patients diagnosed with\n                        epilepsy from January 2014 to January 2022. Patients receiving VPA and\n                        having undergone therapeutic drug monitoring were enrolled. Top three\n                        algorithms exhibiting superior model performance were selected to establish\n                        the ensemble prediction model, with Shapley Additive exPlanations (SHAP)\n                        employed for model interpretation. An independent dataset was collected as a\n                        clinical validation group to verify the prediction model performance.<\/jats:p>\n                  <jats:p>The algorithms chosen for the ensemble model\u2014Light Gradient Boosting,\n                        Categorical Boosting, and Gradient Boosted Regression Trees\u2014demonstrated\n                        high R\n                  2 (0.549, 0.515, and 0.503, respectively).\n                        Post-feature selection, the final model incorporated 20 variables, proving\n                        superior in predictive performance compared to models considering all 24\n                        variables. The R\n                  \n                     2\n                  , mean absolute error, mean square\n                        error, absolute accuracy (\u00b120\u2009mg\/L), and relative accuracy (\u00b120%) of\n                        external validation were 0.621, 10.67, 221.50, 78.98%, and 66.48%,\n                        respectively. The importance and direction of each variable were visually\n                        represented using SHAP values, with VPA administration and liver function\n                        emerging as the most significant factors.<\/jats:p>\n                  <jats:p>The innovative application harnesses advanced multi-algorithm mining\n                        methodologies to forecast VPA concentrations in adult epileptic patients.\n                        Furthermore, it employs SHAP to elucidate the nuanced influence of each\n                        feature within the integrated prediction model, thereby providing a robust\n                        and plausible explanation for the determinants affecting VPA concentration\n                        predictions.<\/jats:p>","DOI":"10.1055\/a-2593-3125","type":"journal-article","created":{"date-parts":[[2025,6,2]],"date-time":"2025-06-02T19:24:32Z","timestamp":1748892272000},"page":"145-156","source":"Crossref","is-referenced-by-count":3,"title":["Ensemble Machine Learning Model for Real-Time Valproic Acid\n                    Prediction in Epilepsy Treatment"],"prefix":"10.1055","volume":"59","author":[{"given":"Jiangchuan","family":"Xie","sequence":"additional","affiliation":[{"name":"Department of pharmacy, The First Affiliated Hospital of Army Medical\n                    University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pan","family":"Ma","sequence":"additional","affiliation":[{"name":"Department of pharmacy, The First Affiliated Hospital of Army Medical\n                    University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinmei","family":"Pan","sequence":"additional","affiliation":[{"name":"Department of pharmacy, The First Affiliated Hospital of Army Medical\n                    University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liya","family":"Cao","sequence":"additional","affiliation":[{"name":"Department of pharmacy, The First Affiliated Hospital of Army Medical\n                    University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ruixiang","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of pharmacy, The First Affiliated Hospital of Army Medical\n                    University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lirong","family":"Xiong","sequence":"additional","affiliation":[{"name":"Department of pharmacy, The First Affiliated Hospital of Army Medical\n                    University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongqian","family":"Wang","sequence":"additional","affiliation":[{"name":"Medical Big Data and Artificial Intelligence Center, the First\n                    Affiliated Hospital of Army Medical University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xin","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of pharmacy, The First Affiliated Hospital of Army Medical\n                    University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Linli","family":"Xie","sequence":"additional","affiliation":[{"name":"Department of pharmacy, The First Affiliated Hospital of Army Medical\n                    University, Chongqing, China"},{"name":"Medical Big Data and Artificial Intelligence Center, the First\n                    Affiliated Hospital of Army Medical University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongchuan","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of pharmacy, The First Affiliated Hospital of Army Medical\n                    University, Chongqing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"194","published-online":{"date-parts":[[2025,6,2]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"7","DOI":"10.1111\/epi.14458","article-title":"Epilepsy in Asia: Disease burden, management barriers, and challenges","volume":"60","author":"E Trinka","year":"2019","journal-title":"Epilepsia"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"689","DOI":"10.1016\/S0140-6736(18)32596-0","article-title":"Epilepsy in adults","volume":"393","author":"R D Thijs","year":"2019","journal-title":"Lancet"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"1333","DOI":"10.1111\/epi.17237","article-title":"Methodology for classification and definition of epilepsy syndromes with list of\n                    syndromes: Report of the ILAE Task Force on Nosology and Definitions","volume":"63","author":"E C Wirrell","year":"2022","journal-title":"Epilepsia"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"1323","DOI":"10.1016\/j.clinbiochem.2013.06.012","article-title":"Adverse drug reactions induced by valproic acid","volume":"46","author":"R M Nanau","year":"2013","journal-title":"Clin Biochem"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"191","DOI":"10.1007\/s00406-017-0827-5","article-title":"Use and safety of antiepileptic drugs in psychiatric inpatients\u2014data from the\n                    AMSP study","volume":"268","author":"K Druschky","year":"2018","journal-title":"Eur Arch Psychiatry Clin Neurosci"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"e238201","DOI":"10.1371\/journal.pone.0238201","article-title":"Safety range of free valproic acid serum concentration in adult patients","volume":"15","author":"Y J Tseng","year":"2020","journal-title":"PLoS One"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"1114","DOI":"10.1016\/j.jfma.2020.09.004","article-title":"Factors to influence the accuracy of albumin adjusted free valproic acid\n                    concentration","volume":"120","author":"Y J Tseng","year":"2021","journal-title":"J Formos Med Assoc"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"e6","DOI":"10.1136\/ejhpharm-2020-002477","article-title":"Predictive performance of reported vancomycin population pharmacokinetic model\n                    in patients with different renal function status, especially those with\n                    augmented renal clearance","volume":"29","author":"Y X Yu","year":"2022","journal-title":"Eur J Hosp Pharm"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"50","DOI":"10.3760\/cma.j.cn121430-20190814-00009","article-title":"Predictive performance of population pharmacokinetic software on vancomycin\n                    steady-state trough concentration","volume":"32","author":"S Xue","year":"2020","journal-title":"Chinese Critical Care Medicine"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"1278","DOI":"10.1002\/cpt.2577","article-title":"Artificial neural network vs pharmacometric model for population prediction of\n                    plasma concentration in real-world data: A case study on valproic acid.","volume":"111","author":"H Soeorg","year":"2022","journal-title":"Clin Pharmacol Ther"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"975855","DOI":"10.3389\/fphar.2022.975855","article-title":"An interpretable stacking ensemble learning framework based on multi-dimensional\n                    data for real-time prediction of drug concentration: The example of\n                    olanzapine","volume":"13","author":"X Zhu","year":"2022","journal-title":"Front Pharmacol"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"985","DOI":"10.1097\/CCM.0000000000005894","article-title":"Bringing the promise of artificial intelligence to critical care: What the\n                    experience with sepsis analytics can teach us","volume":"51","author":"G Wardi","year":"2023","journal-title":"Crit Care Med"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"215","DOI":"10.1186\/s13054-022-04088-9","article-title":"External validation of a machine learning model to predict hemodynamic\n                    instability in intensive care unit","volume":"26","author":"C Dung-Hung","year":"2022","journal-title":"Crit Care"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"2714","DOI":"10.1111\/bcp.15734","article-title":"A machine learning model for predicting blood concentration of quetiapine in\n                    patients with schizophrenia and depression based on real-world data","volume":"89","author":"Y Hao","year":"2023","journal-title":"Br J Clin Pharmacol"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"2815","DOI":"10.1093\/jac\/dkae292","article-title":"Prediction of teicoplanin plasma concentration in critically ill patients: A\n                    combination of machine learning and population pharmacokinetics","volume":"79","author":"P Ma","year":"2024","journal-title":"J Antimicrob Chemother"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"808969","DOI":"10.3389\/fmed.2022.808969","article-title":"Construction and interpretation of prediction model of teicoplanin trough\n                    concentration via machine learning","volume":"9","author":"P Ma","year":"2022","journal-title":"Front Med (Lausanne)"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"816","DOI":"10.1111\/bcp.13510","article-title":"A systematic review of population pharmacokinetics of valproic acid","volume":"84","author":"J Methaneethorn","year":"2018","journal-title":"Br J Clin Pharmacol"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"479","DOI":"10.1007\/s10928-013-9323-3","article-title":"Sex related differences on valproic acid pharmacokinetics after oral single\n                    dose","volume":"40","author":"M Ibarra","year":"2013","journal-title":"J Pharmacokinet Pharmacodyn"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1016\/j.cell.2020.03.022","article-title":"How machine learning will transform biomedicine","volume":"181","author":"J Goecks","year":"2020","journal-title":"Cell"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"e934275","DOI":"10.12659\/MSM.934275","article-title":"Factors influencing sodium valproate serum concentrations in patients with\n                    epilepsy based on logistic regression analysis","volume":"27","author":"X Lan","year":"2021","journal-title":"Med Sci Monit"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"181","DOI":"10.1177\/0018578719831974","article-title":"Valproate interaction with carbapenems: Review and recommendations","volume":"55","author":"O Al-Quteimat","year":"2020","journal-title":"Hosp Pharm"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1055\/s-0043-116492","article-title":"Consensus guidelines for therapeutic drug monitoring in neuropsychopharmacology:\n                    Update 2017","volume":"51","author":"C Hiemke","year":"2018","journal-title":"Pharmacopsychiatry"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"5417","DOI":"10.21037\/apm-21-795","article-title":"Interaction between valproic acid and carbapenems: Decreased plasma\n                    concentration of valproic acid and liver injury","volume":"10","author":"Z Li","year":"2021","journal-title":"Ann Palliat Med"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"587","DOI":"10.1097\/FTD.0000000000000316","article-title":"The effect of different carbapenem antibiotics (ertapenem, imipenem\/cilastatin,\n                    and meropenem) on serum valproic acid concentrations","volume":"38","author":"C Wu","year":"2016","journal-title":"Ther Drug Monit"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"1685","DOI":"10.1002\/jps.20338","article-title":"Increased erythrocyte distribution of valproic acid in pharmacokinetic\n                    interaction with carbapenem antibiotics in rat and human","volume":"94","author":"K Omoda","year":"2005","journal-title":"J Pharm Sci"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"648","DOI":"10.1177\/0897190019882880","article-title":"Rhabdomyolysis and hepatotoxicity from valproic acid: Case reports","volume":"34","author":"C P Walker","year":"2021","journal-title":"J Pharm Pract"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"181","DOI":"10.1177\/0018578719831974","article-title":"Valproate interaction with carbapenems: Review and recommendations","volume":"55","author":"O Al-Quteimat","year":"2020","journal-title":"Hosp Pharm"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"1538","DOI":"10.1124\/dmd.110.032938","article-title":"Identification of valproic acid glucuronide hydrolase as a key enzyme for the\n                    interaction of valproic acid with carbapenem antibiotics","volume":"38","author":"E Suzuki","year":"2010","journal-title":"Drug Metab Dispos"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"S2","DOI":"10.1016\/j.amjcard.2020.12.018","article-title":"Overview of aspirin and platelet biology","volume":"144","author":"C G Santos-Gallego","year":"2021","journal-title":"Am J Cardiol"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1111\/bcpt.13208","article-title":"Identifying factors affecting the pharmacokinetics of voriconazole in patients\n                    with liver dysfunction: A population pharmacokinetic approach","volume":"125","author":"D Tang","year":"2019","journal-title":"Basic Clin Pharmacol Toxicol"},{"key":"ref32","doi-asserted-by":"publisher","first-page":"1890","DOI":"10.1111\/bcp.14578","article-title":"Population pharmacokinetics, safety and dosing optimization of voriconazole in\n                    patients with liver dysfunction: A prospective observational study","volume":"87","author":"D Tang","year":"2021","journal-title":"Br J Clin Pharmacol"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"900","DOI":"10.1002\/phar.1965","article-title":"Clinical importance of monitoring unbound valproic acid concentration in\n                    patients with hypoalbuminemia","volume":"37","author":"M Dore","year":"2017","journal-title":"Pharmacotherapy"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"222","DOI":"10.1016\/j.jss.2021.04.014","article-title":"Valproic acid protects against acute kidney injury in hemorrhage and trauma","volume":"266","author":"B E Biesterveld","year":"2021","journal-title":"J Surg Res"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"1220751833","DOI":"10.1177\/0300060520952281","article-title":"Impact of gender, albumin, and CYP2C19 polymorphisms on valproic acid in Chinese\n                    patients: A population pharmacokinetic model [J]","volume":"48","author":"J Guo","year":"2020","journal-title":"J Int Med Res"},{"key":"ref36","doi-asserted-by":"publisher","first-page":"641512","DOI":"10.3389\/fphar.2021.641512","article-title":"Evaluation of factors impacting the efficacy of single or combination therapies\n                    of valproic acid, carbamazepine, and oxcarbazepine: A longitudinal observation\n                    Study","volume":"12","author":"Q Peng","year":"2021","journal-title":"Front Pharmacol"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"256","DOI":"10.1111\/ijlh.13158","article-title":"Evaluation of the HemoCue WBC DIFF in leukopenic patient samples","volume":"42","author":"D K Kur","year":"2020","journal-title":"Int J Lab Hematol"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"886","DOI":"10.1016\/j.cmi.2015.05.002","article-title":"Optimization of dosing regimens and dosing in special populations","volume":"21","author":"F B Sime","year":"2015","journal-title":"Clin Microbiol Infect"}],"container-title":["Pharmacopsychiatry"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/www.thieme-connect.de\/products\/ejournals\/pdf\/10.1055\/a-2593-3125.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T22:30:59Z","timestamp":1778279459000},"score":1,"resource":{"primary":{"URL":"http:\/\/www.thieme-connect.de\/DOI\/DOI?10.1055\/a-2593-3125"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,2]]},"references-count":37,"journal-issue":{"issue":"03","published-online":{"date-parts":[[2026,5,8]]},"published-print":{"date-parts":[[2026,5]]}},"URL":"https:\/\/doi.org\/10.1055\/a-2593-3125","archive":["Portico","CLOCKSS"],"relation":{},"ISSN":["0176-3679","1439-0795"],"issn-type":[{"value":"0176-3679","type":"print"},{"value":"1439-0795","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,2]]}}}