{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T20:44:58Z","timestamp":1785876298650,"version":"3.56.0"},"reference-count":42,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Transfusion thresholds in upper gastrointestinal bleeding are debated; hemoglobin cutoffs of 70\u201080 g\/L are widely cited yet inconsistently applied. Common risk scores offer limited individualized guidance and rarely provide calibrated, interpretable predictions for transfusion decisions.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to develop and validate a two-stage, clinically constrained gradient-boosting framework (Medically Constrained Gradient Boosting [MCGB]) that predicts transfusion need and estimates transfusion dose with quantified uncertainty and to implement a prototype recommendation system for clinical use.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n We analyzed a retrospective multicenter cohort of 849 adults with endoscopically confirmed upper gastrointestinal bleeding admitted to 3 hospitals in Chongqing, China (January 2019 to August 2025). Predictors available before the transfusion decision included demographics, first recorded vital signs, initial laboratory indices, and clinician-adjudicated etiology. Stage 1 used a calibrated classifier with prespecified monotonic constraints and stability-screened, clinically justified interactions. Stage 2 modeled transfusion dose via quantile predictions with conformal adjustment to generate 95% prediction intervals. Performance was assessed using a cross-site hold-out design. Overall, 2 hospitals were used as the development cohort, within which stratified 5-fold cross-validation was performed for model development, hyperparameter tuning, interaction screening, and calibration. The remaining hospital was held out as an independent test cohort for final evaluation. Hospital-wise alternating external testing was further conducted as a supplementary robustness analysis to assess performance stability across institutions. Classification performance was evaluated using discrimination metrics (area under the receiver operating characteristic curve and area under the precision-recall curve), calibration metrics, and decision-curve analysis; regression performance was evaluated using\n R<\/jats:italic>\n \u00b2, mean absolute error, and prediction-interval coverage. A graphical user interface was implemented to enable clinicians to input patient data and obtain calibrated predictions of transfusion probability and corresponding dose recommendations.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n MCGB achieved strong discrimination and good calibration across subgroups (area under the receiver operating characteristic curve=0.97 and area under the precision-recall curve=0.91). At a reference probability threshold of .50, sensitivity, specificity, and\n F<\/jats:italic>\n 1<\/jats:sub>\n -scores were 0.99, 0.87, and 0.85, respectively, providing a representative operating point for comparison. For dose prediction among transfused patients, MCGB achieved\n R<\/jats:italic>\n \u00b2 of 0.95 and mean absolute error 0.04; 95% prediction-interval coverage was 0.94, indicating accurate point estimates with reliable uncertainty quantification. The software prototype further demonstrated feasibility of real-time decision support at the bedside.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n MCGB provides calibrated, interpretable predictions of transfusion need and individualized dose in upper gastrointestinal bleeding and may support bedside decision-making and blood-bank planning, with a prototype interface demonstrating potential for clinical deployment. External validation in additional settings is warranted to confirm generalizability.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/83889","type":"journal-article","created":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T17:50:08Z","timestamp":1781286608000},"page":"e83889-e83889","source":"Crossref","is-referenced-by-count":0,"title":["Prediction of Blood Transfusion Need and Dose in Patients With Upper Gastrointestinal Bleeding: Retrospective Multicenter Prediction Model Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1616-4985","authenticated-orcid":false,"given":"Xiaoyu","family":"Li","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7208-247X","authenticated-orcid":false,"given":"Yuqin","family":"He","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2056-6154","authenticated-orcid":false,"given":"Mingyang","family":"Hou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2305-0492","authenticated-orcid":false,"given":"Zhi","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-4772-493X","authenticated-orcid":false,"given":"Shuai","family":"Miao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6368-1008","authenticated-orcid":false,"given":"Zhiyong","family":"Huang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6802-7862","authenticated-orcid":false,"given":"Min","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,8,4]]},"reference":[{"key":"R1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajem.2024.04.052","article-title":"Emergency medicine updates: upper gastrointestinal bleeding","volume":"81","author":"Long","journal-title":"Am J Emerg Med"},{"key":"R2","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.l536","article-title":"Management of acute upper gastrointestinal bleeding","volume":"364","author":"Stanley","journal-title":"BMJ"},{"issue":"1","key":"R3","doi-asserted-by":"publisher","DOI":"10.1136\/bmjmed-2022-000202","article-title":"Update on the management of upper gastrointestinal bleeding","volume":"1","author":"Orpen-Palmer","journal-title":"BMJ Med"},{"issue":"11","key":"R4","doi-asserted-by":"publisher","first-page":"805","DOI":"10.7326\/M19-1795","article-title":"Management of nonvariceal upper gastrointestinal bleeding: guideline recommendations from the International Consensus Group","volume":"171","author":"Barkun","journal-title":"Ann Intern Med"},{"issue":"3","key":"R5","doi-asserted-by":"publisher","first-page":"300","DOI":"10.1055\/a-1369-5274","article-title":"Endoscopic diagnosis and management of nonvariceal upper gastrointestinal hemorrhage (NVUGIH): European Society of Gastrointestinal Endoscopy (ESGE) Guideline - update 2021","volume":"53","author":"Gralnek","journal-title":"Endoscopy"},{"issue":"5","key":"R6","doi-asserted-by":"publisher","first-page":"899","DOI":"10.14309\/ajg.0000000000001245","article-title":"ACG clinical guideline: upper gastrointestinal and ulcer bleeding","volume":"116","author":"Laine","journal-title":"Am J Gastroenterol"},{"issue":"4","key":"R7","doi-asserted-by":"publisher","first-page":"959","DOI":"10.1016\/j.jhep.2021.12.022","article-title":"Baveno VII - renewing consensus in portal hypertension","volume":"76","author":"de Franchis","journal-title":"J Hepatol"},{"issue":"12","key":"R8","doi-asserted-by":"publisher","DOI":"10.1002\/14651858.CD002042.pub5","article-title":"Transfusion thresholds for guiding red blood cell transfusion","volume":"12","author":"Carson","journal-title":"Cochrane Database Syst Rev"},{"issue":"12","key":"R9","doi-asserted-by":"publisher","first-page":"1368","DOI":"10.1007\/s00134-021-06531-x","article-title":"Transfusion strategies in bleeding critically ill adults: a clinical practice guideline from the European Society of Intensive Care Medicine","volume":"47","author":"Vlaar","journal-title":"Intensive Care Med"},{"issue":"9238","key":"R10","doi-asserted-by":"publisher","first-page":"1318","DOI":"10.1016\/S0140-6736(00)02816-6","article-title":"A risk score to predict need for treatment for upper-gastrointestinal haemorrhage","volume":"356","author":"Blatchford","journal-title":"Lancet"},{"issue":"3","key":"R11","doi-asserted-by":"publisher","first-page":"316","DOI":"10.1136\/gut.38.3.316","article-title":"Risk assessment after acute upper gastrointestinal haemorrhage","volume":"38","author":"Rockall","journal-title":"Gut"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.i6432","article-title":"Comparison of risk scoring systems for patients presenting with upper gastrointestinal bleeding: international multicentre prospective study","volume":"356","author":"Stanley","journal-title":"BMJ"},{"issue":"4","key":"R13","doi-asserted-by":"publisher","first-page":"707","DOI":"10.1136\/gutjnl-2019-320002","article-title":"ABC score: a new risk score that accurately predicts mortality in acute upper and lower gastrointestinal bleeding: an international multicentre study","volume":"70","author":"Laursen","journal-title":"Gut"},{"issue":"6","key":"R14","doi-asserted-by":"publisher","first-page":"1215","DOI":"10.1016\/j.gie.2011.06.024","article-title":"A simple risk score accurately predicts in-hospital mortality, length of stay, and cost in acute upper GI bleeding","volume":"74","author":"Saltzman","journal-title":"Gastrointest Endosc"},{"issue":"1","key":"R15","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1111\/jgh.14811","article-title":"MAP(ASH): a new scoring system for the prediction of intervention and mortality in upper gastrointestinal bleeding","volume":"35","author":"Redondo-Cerezo","journal-title":"J Gastroenterol Hepatol"},{"issue":"1","key":"R16","doi-asserted-by":"publisher","first-page":"160","DOI":"10.1053\/j.gastro.2019.09.009","article-title":"Validation of a machine learning model that outperforms clinical risk scoring systems for upper gastrointestinal bleeding","volume":"158","author":"Shung","journal-title":"Gastroenterology"},{"issue":"8","key":"R17","doi-asserted-by":"publisher","first-page":"2078","DOI":"10.1007\/s10620-019-05645-z","article-title":"Machine learning to predict outcomes in patients with acute gastrointestinal bleeding: a systematic review","volume":"64","author":"Shung","journal-title":"Dig Dis Sci"},{"issue":"6","key":"R18","doi-asserted-by":"publisher","first-page":"1198","DOI":"10.1053\/j.gastro.2024.06.030","article-title":"Validation of an electronic health record-based machine learning model compared with clinical risk scores for gastrointestinal bleeding","volume":"167","author":"Shung","journal-title":"Gastroenterology"},{"issue":"2","key":"R19","doi-asserted-by":"publisher","first-page":"371","DOI":"10.14309\/ajg.0000000000002520","article-title":"Achieving value by risk stratification with machine learning model or clinical risk score in acute upper gastrointestinal bleeding: a cost minimization analysis","volume":"119","author":"Shung","journal-title":"Am J Gastroenterol"},{"issue":"1","key":"R20","doi-asserted-by":"publisher","first-page":"55","DOI":"10.7326\/M14-0697","article-title":"Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement","volume":"162","author":"Collins","journal-title":"Ann Intern Med"},{"issue":"1","key":"R21","doi-asserted-by":"publisher","first-page":"51","DOI":"10.7326\/M18-1376","article-title":"PROBAST: a tool to assess the risk of bias and applicability of prediction model studies","volume":"170","author":"Wolff","journal-title":"Ann Intern Med"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-022-01577-x","article-title":"Methodological conduct of prognostic prediction models developed using machine learning in oncology: a systematic review","volume":"22","author":"Dhiman","journal-title":"BMC Med Res Methodol"},{"key":"R23","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.n2281","article-title":"Risk of bias in studies on prediction models developed using supervised machine learning techniques: systematic review","volume":"375","author":"Andaur Navarro","journal-title":"BMJ"},{"issue":"29","key":"R24","doi-asserted-by":"publisher","first-page":"1925","DOI":"10.1093\/eurheartj\/ehu207","article-title":"Towards better clinical prediction models: seven steps for development and an ABCD for validation","volume":"35","author":"Steyerberg","journal-title":"Eur Heart J"},{"issue":"1","key":"R25","doi-asserted-by":"publisher","first-page":"128","DOI":"10.1097\/EDE.0b013e3181c30fb2","article-title":"Assessing the performance of prediction models: a framework for traditional and novel measures","volume":"21","author":"Steyerberg","journal-title":"Epidemiology"},{"key":"R26","doi-asserted-by":"publisher","DOI":"10.1016\/j.jclinepi.2015.12.005","article-title":"A calibration hierarchy for risk models was defined: from utopia to empirical data","volume":"74","author":"Van Calster","journal-title":"J Clin Epidemiol"},{"issue":"6","key":"R27","doi-asserted-by":"publisher","first-page":"565","DOI":"10.1177\/0272989X06295361","article-title":"Decision curve analysis: a novel method for evaluating prediction models","volume":"26","author":"Vickers","journal-title":"Med Decis Making"},{"key":"R28","doi-asserted-by":"publisher","DOI":"10.1186\/s41512-019-0064-7","article-title":"A simple, step-by-step guide to interpreting decision curve analysis","volume":"3","author":"Vickers","journal-title":"Diagn Progn Res"},{"issue":"1","key":"R29","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1175\/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2","article-title":"Verification of forecasts expressed in terms of probability","volume":"78","author":"Brier","journal-title":"Mon Wea Rev"},{"key":"R30","unstructured":"Romano Y Patterson E Cand\u00e8s EJ . Conformalized quantile regression. In: Advances in Neural Information Processing Systems. Vol 2019. :3543-3553."},{"issue":"11","key":"R31","doi-asserted-by":"publisher","DOI":"10.1111\/cts.70056","article-title":"Practical guide to SHAP analysis: explaining supervised machine learning model predictions in drug development","volume":"17","author":"Ponce-Bobadilla","journal-title":"Clin Transl Sci"},{"key":"R32","doi-asserted-by":"crossref","unstructured":"Chen T Guestrin C . XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery; 785-794. [doi: 10.1145\/2939672.2939785]","DOI":"10.1145\/2939672.2939785"},{"key":"R33","unstructured":"Prokhorenkova L Gusev G Vorobev A Dorogush AV Gulin A . CatBoost: unbiased boosting with categorical features. In: Advances in Neural Information Processing Systems. Vol 2018. :6638-6648."},{"key":"R34","unstructured":"Guo C Pleiss G Sun Y Weinberger KQ . On calibration of modern neural networks. Presented at: Proceedings of the 34th international conference on machine learning; Aug 6-11, 2017; Sydney, Australia."},{"issue":"3","key":"R35","doi-asserted-by":"publisher","first-page":"524","DOI":"10.1213\/ANE.0000000000006047","article-title":"Artificial intelligence and machine learning in patient blood management: a scoping review","volume":"135","author":"Meier","journal-title":"Anesth Analg"},{"issue":"1","key":"R36","doi-asserted-by":"publisher","first-page":"162","DOI":"10.1111\/trf.17582","article-title":"Machine learning in transfusion medicine: a scoping review","volume":"64","author":"Maynard","journal-title":"Transfusion"},{"issue":"1","key":"R37","doi-asserted-by":"publisher","first-page":"577","DOI":"10.1182\/hematology.2019000062","article-title":"Clinical decision support and improved blood use in patient blood management","volume":"2019","author":"Goodnough","journal-title":"Hematology Am Soc Hematol Educ Program"},{"issue":"5","key":"R38","doi-asserted-by":"publisher","first-page":"1358","DOI":"10.1111\/trf.12445","article-title":"Improved blood utilization using real-time clinical decision support","volume":"54","author":"Goodnough","journal-title":"Transfusion"},{"issue":"10","key":"R39","doi-asserted-by":"publisher","first-page":"2340","DOI":"10.1007\/s00268-023-07098-y","article-title":"Assessment of machine learning methods to predict massive blood transfusion in trauma","volume":"47","author":"Strickland","journal-title":"World J Surg"},{"issue":"10","key":"R40","doi-asserted-by":"publisher","first-page":"927","DOI":"10.1002\/ajh.24101","article-title":"Is there a \u201cmagic\u201d hemoglobin number? clinical decision support promoting restrictive blood transfusion practices","volume":"90","author":"Goodnough","journal-title":"Am J Hematol"},{"issue":"6","key":"R41","doi-asserted-by":"publisher","first-page":"761","DOI":"10.1111\/vox.13254","article-title":"A review of electronic medical records and safe transfusion practice for guideline development","volume":"117","author":"Crispin","journal-title":"Vox Sang"},{"key":"R42","doi-asserted-by":"publisher","DOI":"10.1016\/j.jclinane.2023.111272","article-title":"Development of a machine learning model to predict intraoperative transfusion and guide type and screen ordering","volume":"91","author":"Zapf","journal-title":"J Clin Anesth"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T19:52:56Z","timestamp":1785873176000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e83889"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,4]]},"references-count":42,"URL":"https:\/\/doi.org\/10.2196\/83889","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,4]]},"article-number":"v14i1e83889"}}