Machine Learning-Based Mortality Prediction for Acute Gastrointestinal Bleeding Patients Admitted to Intensive Care Unit

Abstract

Objective

The study aimed to develop machine learning (ML) models to predict the mortality of patients with acute gastrointestinal bleeding (AGIB) in the intensive care unit (ICU) and compared their prognostic performance with that of Acute Physiology and Chronic Health Evaluation II (APACHE-II) score.

Methods

A total of 961 AGIB patients admitted to the ICU of Renmin Hospital of Wuhan University from January 2020 to December 2023 were enrolled. Patients were randomly divided into the training cohort (n = 768) and the validation cohort (n= 193). Clinical data were collected within the first 24 h of ICU admission. ML models were constructed using Python V.3.7 package, employing 3 different algorithms: XGBoost, Random Forest (RF) and Gradient Boosting Decision Tree (GBDT). The area under the receiver operating characteristic (ROC) curve (AUC) was used to evaluate the performance of different models.

Results

A total of 94 patients died with an overall mortality of 9.78% (11.32% in the training cohort and 8.96% in the validation cohort). Among the 3 ML models, the GBDT algorithm demonstrated the highest predictive performance, achieving an AUC of 0.95 (95% CI 0.90-0.99), while the AUCs of XGBoost and RF models were 0.89 (95% CI 0.82-0.96) and 0.90 (95% CI 0.84-0.96), respectively. In comparison, the APACHE-II model achieved an AUC of 0.74 (95% CI 0.69-0.87), with a specificity of 70.97% (95% CI 64.07-77.01). When APACHE-II score was incorporated into the GBDT algorithm, the ensemble model achieved an AUC of 0.98 (95% CI 0.96-0.99) with a sensitivity of 85.71% and a specificity up to 95.15%.

Conclusions

The GBDT model serves as a reliable tool for accurately predicting the in-hospital mortality for AGIB patients. When integrated with the APACHE-II score, the ensemble GBDT algorithm further enhances predictive accuracy and provides valuable insights for prognostic evaluation.

Keywords

Acute gastrointestinal bleeding / Intensive care unit / APACHE-II / Machine learning / Artificial intelligence / Mortality

© The Author(s), under exclusive licence to Huazhong University of Science and Technology 2025

Funding The study was supported by the Cross Innovation Talent Project, Renmin Hospital of Wuhan University (No. JCR- CYG-2022-005, No. JCRCZN-2022-017); the National Natural Science Foundation of China (No. 82302418); the Knowledge Innovation Program of Wuhan, Science and Technology Innovation of Wuhan (No. 2023020201010165); The Key Project of Education Foundation, Guang Zhou 21st Century Education Foundation (No. 2023HX0054).

Availability of Supporting Data All raw data could be acquired by requested for the corresponding author by e-mail. The following information was supplied regarding data availability. The raw data tables are available in Supplement File.

Declarations

Conflict of Interest All authors declare no conflict interest in this research.

Ethical Approval and Consent to Participate The study has been approved by the Ethics Committee of Renmin Hospital of Wuhan University (WDRY2024-K103). This was a retrospective and noninterventional study that fully complied with the Helsinki Declaration. The ethics committee exempted the requirement for patient informed consent.

Human Ethics The study has been approved by the Ethics Committee of Renmin Hospital of Wuhan University (WDRY2024-K103).

Consent for Publication The authors all approved for publication and signed copyright transfer statement.

References

[1] Oakland K. Changing epidemiology and etiology of upper and lower gastrointestinal bleeding. Best Pract Res Clin Gastroenterol. 2019;42-43:101610.

[2] Yao Y, Ba T, Bao B, et al. Sepsis as a Potential Risk Factor for Upper Gastrointestinal Bleeding in Critically Ill Patients: A Systematic Review and Meta-analysis. J Intensive Care Med. 2024:8850666241252048.

[3] Laine L, Barkun AN, Saltzman JR, et al. ACG Clinical Guideline: Upper Gastrointestinal and Ulcer Bleeding. Am J Gastroenterol. 2021; 116(5):899-917.

[4] Costable NJ, Greenwald DA. Upper Gastrointestinal Bleeding. Clin Geriatr Med. 2021; 37(1):155172.

[5] Sengupta N, Feuerstein JD, Jairath V, et al. Management of Patients With Acute Lower Gastrointestinal Bleeding: An Updated ACG Guideline. Am J Gastroenterol. 2023; 118(2):208-231.

[6] Aoki T, Hirata Y, Yamada A, et al. Initial management for acute lower gastrointestinal bleeding. World J Gastroenterol. 2019; 25(1):69-84.

[7] Banigan M, Kranenburg L, Vise J. Upper Gastrointestinal Bleeding: Evaluation and Diagnosis. Gastroenterol Nurs. 2023; 46(5):348-358.

[8] Saltzman JR, Tabak YP, Hyett BH, et al. A simple risk score accurately predicts in-hospital mortality, length of stay, and cost in acute upper GI bleeding. Gastrointest Endosc. 2011; 74(6):1215-1224.

[9] Sengupta N, Tapper EB, Patwardhan VR, et al. High Glasgow Blatchford Score at admission is associated with recurrent bleeding after discharge for patients hospitalized with upper gastrointestinal bleeding. Endoscopy. 2016; 48(1):9-15.

[10] Uysal Y, Babus SB, Kose A, et al. The prognostic significance of the risk scores at upper gastrointestinal bleeding. Niger J Clin Pract. 2019; 22(8):1099-1108.

[11] Dogru U, Yuksel M, Ay MO, et al. The effect of the shock index and scoring systems for predicting mortality among geriatric patients with upper gastrointestinal bleeding: a prospective cohort study. Sao Paulo Med J. 2022; 140(4):531-539.

[12] Nayarisseri A, Khandelwal R, Tanwar P, et al. Artificial Intelligence, Big Data and Machine Learning Approaches in Precision Medicine & Drug Discovery. Curr Drug Targets. 2021; 22(6):631-655.

[13] The Lancet Digital H. Walking the tightrope of artificial intelligence guidelines in clinical practice. Lancet Digit Health. 2019; 1(3):e100.

[14] Gutierrez G. Artificial Intelligence in the Intensive Care Unit. Crit Care. 2020; 24(1):101.

[15] Deshmukh F, Merchant SS. Explainable Machine Learning Model for Predicting GI Bleed Mortality in the Intensive Care Unit. Am J Gastroenterol. 2020; 115(10):16571668.

[16] Ungureanu BS, Gheonea DI, Florescu DN, et al. Predicting mortality in patients with nonvariceal upper gastrointestinal bleeding using machine-learning. Front Med (Lausanne). 2023; 10:1134835.

[17] Sungono V, Hariyanto H, Soesilo TEB, et al. Cohort study of the APACHE II score and mortality for different types of intensive care unit patients. Postgrad Med J. 2022; 98(1166):914-918.

[18] Kisuule I, Seremba E, Kagimu M. Prevalence of gastrointestinal bleeding and frequency of selected predictors of mortality on the medical emergency ward at Mulago hospital. Afr Health Sci. 2023; 23(1):622-630.

[19] Chang DW, Shapiro MF. Association Between Intensive Care Unit Utilization During Hospitalization and Costs, Use of Invasive Procedures, and Mortality. JAMA Intern Med. 2016; 176(10):1492-1499.

[20] Liu Z, Zhang L, Li G, et al. A Nomogram Model for Prediction of Mortality Risk of Patients with Dangerous Upper Gastrointestinal Bleeding: A Two-center Retrospective Study. Curr Med Sci. 2023; 43(4):723-732.

[21] Zhang X, Ni J, Zhang H, et al. A nomogram to predict in-hospital mortality of gastrointestinal bleeding patients in the intensive care unit. Front Med (Lausanne). 2023; 10:1204099.

[22] Zhang Z, Jung C. GBDT-MO: Gradient-Boosted Decision Trees for Multiple Outputs. IEEE Trans Neural Netw Learn Syst. 2021; 32(7):3156-3167.

[23] Hou N, Li M, He L, et al. Predicting 30-days mortality for MIMIC-III patients with sepsis-3: a machine learning approach using XGboost. J Transl Med. 2020; 18(1):462.

[24] Wang W, Li L, Gu H, et al. Random Forest-Based Prediction of Acute Respiratory Distress Syndrome in Patients Undergoing Cardiac Surgery. Heart Surg Forum. 2022; 25(6):E854-e859.

[25] Deng K, Jing W, Yang J. Adding the analysis or discussion to the APACHE II score may clearly explain the predictive risk associated with upper gastrointestinal bleeding via the Re. Co. De score. Gastrointest Endosc. 2022; 96(2):385.

[26] Lu X, Zhang X, Chen H. Comparison of the AIMS65 score with the Glasgow-Blatchford and Rockall scoring systems for the prediction of the risk of in-hospital death among patients with upper gastrointestinal bleeding. Rev Esp Enferm Dig. 2020; 112(6):467-473.

[27] Rivieri S, Carron PN, Schoepfer A, et al. External validation and comparison of the Glasgow-Blatchford score, modified Glasgow-Blatchford score, Rockall score and AIMS65 score in patients with upper gastrointestinal bleeding: a cross-sectional observational study in Western Switzerland. Eur J Emerg Med. 2023; 30(1):32-39.

[28] Wilkins T, Wheeler B, Carpenter M. Upper Gastrointestinal Bleeding in Adults: Evaluation and Management. Am Fam Physician. 2020; 101(5):294-300.

[29] Gralnek IM, Camus Duboc M, Garcia-Pagan JC, et al. Endoscopic diagnosis and management of esophagogastric variceal hemorrhage: European Society of Gastrointestinal Endoscopy (ESGE) Guideline. Endoscopy. 2022; 54(11):1094-1120.

[30] García-Iglesias P, Botargues JM, Feu Caballé F, et al. Management of non variceal upper gastrointestinal bleeding: position statement of the Catalan Society of Gastroenterology. Gastroenterol Hepatol. 2017; 40(5):363-374.

[31] Guo CG, Cheung KS, Zhang F, et al. Delay in Retreatment of Helicobacter pylori Infection Increases Risk of Upper Gastrointestinal Bleeding. Clin Gastroenterol Hepatol. 2021; 19(2):314-322. e312.

[32] Yan YC, Chen XQ, Yu J, et al. An unusual cause of acute gastrointestinal bleeding: Hookworm infection. Am J Med Sci. 2023; 366(2):e21-e22.

[33] Chien YW, Chuang HN, Wang YP, et al. Short-term, mediumterm, and long-term risks of nonvariceal upper gastrointestinal bleeding after dengue virus infection. PLoS Negl Trop Dis. 2022; 16(1):e0010039.

[34] Xiao J, Zhang R, Chen W, et al. An unusual case of acute lower gastrointestinal bleeding due to fungal infection. J Int Med Res. 2020; 48(11):300060520967820.

[35] Agrawal D, Ravula PK, Verma G. An unusual opportunistic infection causing lower gastrointestinal tract bleeding in a patient with severe alcoholic hepatitis, post-liver transplantation. Transpl Infect Dis. 2023; 25(4):e14062.

[36] Li X, Huang S, Lu J, et al. Upper Gastrointestinal Bleeding Caused by SARS-CoV-2 Infection. Am J Gastroenterol. 2020; 115(9):1541-1542.

[37] Cappell MS, Friedel DM. Gastrointestinal Bleeding in COVID-19-Infected Patients. Gastroenterol Clin North Am. 2023; 52(1):77-102.

[38] Huang AH, Liu Y, Hsien YC, et al. Survival Impact and Clinical Predictors of Acute Gastrointestinal Bleeding in Patients With Bloodstream Infection. J Intensive Care Med. 2021; 36(1):63-69.

[39] Macedo Silva V, Freitas M, Arieira C, et al. Direct oral anticoagulants are associated with potentially bleeding lesions in suspected mid-gastrointestinal bleeding. Scand J Gastroenterol. 2022; 57(4):486-492.

[40] Akhila Arya PV, Thulaseedharan NK, Raj R, et al. AIMS65, Glasgow-Blatchford bleeding score and modified Glasgow-Blatchford bleeding score in predicting outcomes of upper gastrointestinal bleeding: An accuracy and calibration study. Indian J Gastroenterol. 2023; 42(4):496-504.

[41] Robertson M, Ng J, Abu Shawish W, et al. Risk stratification in acute variceal bleeding: Comparison of the AIMS65 score to established upper gastrointestinal bleeding and liver disease severity risk stratification scoring systems in predicting mortality and rebleeding. Dig Endosc. 2020; 32(5):761-768.

[42] Wang J, Ren H, Gu B, et al. Differential Analysis of Blood Routine Examination Parameters in Patients with Upper Gastrointestinal Bleeding and Lower Gastrointestinal Bleeding. Altern Ther Health Med. 2024; 30(11):98-102.

[43] Nicolai L, Massberg S. Platelets as key players in inflammation and infection. Curr Opin Hematol. 2020; 27(1):34-40.

[44] Chopra D, Rosenberg M, Moayyedi P, et al. Is Blood Urea Concentration an Independent Predictor of Positive Endoscopic Findings in Presumed Upper Gastrointestinal Bleeding? Dig Dis. 2020; 38(1):77-84.

[45] Huaringa-Marcelo J, Huaman MR, Brañez-Condorena A, et al. Vasoactive Agents for the Management of Acute Variceal Bleeding: A Systematic Review and Meta-analysis. J Gastrointestin Liver Dis. 2021; 30(1):110-121.