43 93

Cited 0 times in

Machine-Learning Model for the Prediction of Hypoxaemia during Endoscopic Retrograde Cholangiopancreatography under Monitored Anaesthesia Care

 Huapyong Kang  ;  Bora Lee  ;  Jung Hyun Jo  ;  Hee Seung Lee  ;  Jeong Youp Park  ;  Seungmin Bang  ;  Seung Woo Park  ;  Si Young Song  ;  Joonhyung Park  ;  Hajin Shim  ;  Jung Hyun Lee  ;  Eunho Yang  ;  Eun Hwa Kim  ;  Kwang Joon Kim  ;  Min-Soo Kim  ;  Moon Jae Chung  
 YONSEI MEDICAL JOURNAL, Vol.64(1) : 25-34, 2023-01 
Journal Title
Issue Date
Anesthesia* / adverse effects ; Cholangiopancreatography, Endoscopic Retrograde* / adverse effects ; Humans ; Hypoxia / diagnosis ; Hypoxia / etiology ; Machine Learning ; Retrospective Studies
Cholangiopancreatography, endoscopic retrograde ; hypoxaemia ; machine learning ; monitored anaesthesia care ; prediction model
Purpose: Hypoxaemia is a significant adverse event during endoscopic retrograde cholangiopancreatography (ERCP) under monitored anaesthesia care (MAC); however, no model has been developed to predict hypoxaemia. We aimed to develop and compare logistic regression (LR) and machine learning (ML) models to predict hypoxaemia during ERCP under MAC.

Materials and methods: We collected patient data from our institutional ERCP database. The study population was randomly divided into training and test sets (7:3). Models were fit to training data and evaluated on unseen test data. The training set was further split into k-fold (k=5) for tuning hyperparameters, such as feature selection and early stopping. Models were trained over k loops; the i-th fold was set aside as a validation set in the i-th loop. Model performance was measured using area under the curve (AUC).

Results: We identified 6114 cases of ERCP under MAC, with a total hypoxaemia rate of 5.9%. The LR model was established by combining eight variables and had a test AUC of 0.693. The ML and LR models were evaluated on 30 independent data splits. The average test AUC for LR was 0.7230, which improved to 0.7336 by adding eight more variables with an l1 regularisation-based selection technique and ensembling the LRs and gradient boosting algorithm (GBM). The high-risk group was discriminated using the GBM ensemble model, with a sensitivity and specificity of 63.6% and 72.2%, respectively.

Conclusion: We established GBM ensemble model and LR model for risk prediction, which demonstrated good potential for preventing hypoxaemia during ERCP under MAC.
Files in This Item:
T202300518.pdf Download
Appears in Collections:
1. College of Medicine (의과대학) > Dept. of Anesthesiology and Pain Medicine (마취통증의학교실) > 1. Journal Papers
1. College of Medicine (의과대학) > Dept. of Internal Medicine (내과학교실) > 1. Journal Papers
Yonsei Authors
Kang, Huapyong(강화평) ORCID logo https://orcid.org/0000-0003-1790-0809
Kim, Kwang Joon(김광준) ORCID logo https://orcid.org/0000-0002-5554-8255
Kim, Min Soo(김민수) ORCID logo https://orcid.org/0000-0001-8760-4568
Park, Seung Woo(박승우) ORCID logo https://orcid.org/0000-0001-8230-964X
Park, Jeong Youp(박정엽) ORCID logo https://orcid.org/0000-0003-0110-8606
Bang, Seungmin(방승민) ORCID logo https://orcid.org/0000-0001-5209-8351
Song, Si Young(송시영) ORCID logo https://orcid.org/0000-0002-1417-4314
Lee, Bo Ra(이보라) ORCID logo https://orcid.org/0000-0002-7699-967X
Lee, Hee Seung(이희승) ORCID logo https://orcid.org/0000-0002-2825-3160
Chung, Moon Jae(정문재) ORCID logo https://orcid.org/0000-0002-5920-8549
Jo, Jung Hyun(조중현) ORCID logo https://orcid.org/0000-0002-2641-8873
사서에게 알리기


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.