Advancing Prediction of Risk of Intraoperative Massive Blood Transfusion in Liver Transplantation With Machine Learning Models. A Multicenter Retrospective Study

Chen, Sai; Liu, Leping; Wang, Yongjun; Zhou, Xionghui; Dong, Hang; Chen, Zi-wei; Wu, Jiang; Gui, Rong; Zhao, Qiancheng

doi:10.3389/fninf.2022.893452

Cited by 8 publications

(8 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…21 The Cat-Boost algorithm has also been specifically used in the field of organ transplantation to predict bleeding after liver transplantation. 22 In this study, we used SHAP for explainability analysis, a technique that has been used to explain ML prediction models in the pediatric transplant population. 23 We additionally used SHAP for feature selection.…”

Section: Discussionmentioning

confidence: 99%

“…In this study, we used the CatBoost for 1‐year and 3‐year mortality prediction, as this ML algorithm minimizes errors introduced by categorical variables and has been previously used multiple fields, including cardiology 21 . The CatBoost algorithm has also been specifically used in the field of organ transplantation to predict bleeding after liver transplantation 22 . In this study, we used SHAP for explainability analysis, a technique that has been used to explain ML prediction models in the pediatric transplant population 23 .…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Machine learning‐based prediction of mortality after heart transplantation in adults with congenital heart disease: A UNOS database analysis

Kampaktsis

Siouras²,

Doulamis

et al. 2022

Clinical Transplantation

View full text Add to dashboard Cite

Background Machine learning (ML) is increasingly being applied in Cardiology to predict outcomes and assist in clinical decision‐making. We sought to develop and validate an ML model for the prediction of mortality after heart transplantation (HT) in adults with congenital heart disease (ACHD). Methods The United Network for Organ Sharing (UNOS) database was queried from 2000 to 2020 for ACHD patients who underwent isolated HT. The study cohort was randomly split into derivation (70%) and validation (30%) datasets that were used to train and test a CatBoost ML model. Feature selection was performed using SHapley Additive exPlanations (SHAP). Recipient, donor, procedural, and post‐transplant characteristics were tested for their ability to predict mortality. We additionally used SHAP for explainability analysis, as well as individualized mortality risk assessment. Results The study cohort included 1033 recipients (median age 34 years, 61% male). At 1 year after HT, there were 205 deaths (19.9%). Out of a total of 49 variables, 10 were selected as highly predictive of 1‐year mortality and were used to train the ML model. Area under the curve (AUC) and predictive accuracy for the 1‐year ML model were .80 and 75.2%, respectively, and .69 and 74.2% for the 3‐year model, respectively. Based on SHAP analysis, hemodialysis of the recipient post‐HT had overall the strongest relative impact on 1‐year mortality after HΤ, followed by recipient‐estimated glomerular filtration rate, age and ischemic time. Conclusions ML models showed satisfactory predictive accuracy of mortality after HT in ACHD and allowed for individualized mortality risk assessment.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Machine learning‐based prediction of mortality after heart transplantation in adults with congenital heart disease: A UNOS database analysis

Kampaktsis

Siouras²,

Doulamis

et al. 2022

Clinical Transplantation

View full text Add to dashboard Cite

show abstract

“…This study aimed to train and assess five different machine-learning models utilizing machine-learning-related data from the NHANES (National Health and Nutrition Examination Survey). Based on the papers of other researchers [ 22 , 23 , 24 ], we hypothesized that CATBoost outperformed the other four machine-learning models in predicting diabetes.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Models for Data-Driven Prediction of Diabetes by Lifestyle Type

Qin

et al. 2022

IJERPH

View full text Add to dashboard Cite

The prevalence of diabetes has been increasing in recent years, and previous research has found that machine-learning models are good diabetes prediction tools. The purpose of this study was to compare the efficacy of five different machine-learning models for diabetes prediction using lifestyle data from the National Health and Nutrition Examination Survey (NHANES) database. The 1999–2020 NHANES database yielded data on 17,833 individuals data based on demographic characteristics and lifestyle-related variables. To screen training data for machine models, the Akaike Information Criterion (AIC) forward propagation algorithm was utilized. For predicting diabetes, five machine-learning models (CATBoost, XGBoost, Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM)) were developed. Model performance was evaluated using accuracy, sensitivity, specificity, precision, F1 score, and receiver operating characteristic (ROC) curve. Among the five machine-learning models, the dietary intake levels of energy, carbohydrate, and fat, contributed the most to the prediction of diabetes patients. In terms of model performance, CATBoost ranks higher than RF, LG, XGBoost, and SVM. The best-performing machine-learning model among the five is CATBoost, which achieves an accuracy of 82.1% and an AUC of 0.83. Machine-learning models based on NHANES data can assist medical institutions in identifying diabetes patients.

show abstract

“…At a tertiary referral center, resources like massive transfusion protocols, blood availability, and surgical expertise can be quickly mobilized in the setting of an acute surgical emergency. 6,7 However, these resources are not universally available. In settings where immediate support is not present, such as rural or resource-limited settings, having lead time to predict an event like the need for massive transfusion may well mean the difference between life and death for a patient.…”

mentioning

confidence: 99%

“…The importance of having the ability to predict resource needs rapidly, before an event occurs, is as important now as ever. At a tertiary referral center, resources like massive transfusion protocols, blood availability, and surgical expertise can be quickly mobilized in the setting of an acute surgical emergency . However, these resources are not universally available.…”

mentioning

confidence: 99%

Novel Use of a Real-Time Prediction Model to Enhance Early Detection of Need for Massive Transfusion—Artificial Intelligence Behind the Drapes

Villarreal

Forrester

2022

JAMA Netw Open

View full text Add to dashboard Cite

Intraoperative massive hemorrhage can be associated with high mortality if not treated promptly. 1 Lee et al 2 conducted a prognostic study incorporating intraoperative hemodynamic monitoring data into a validated artificial intelligence (AI)-assisted massive transfusion prediction model. The authors compared the ability of an AI prediction model using intraoperative hemodynamic monitoring with a current reference-standard model (using only preoperative data) to detect need for massive transfusion intraoperatively-10 minutes before the need for massive transfusion occurred. 2 The massive transfusion index prediction model outperformed the standard model in the internal and external validation single-center cohorts (area under the receiver operating characteristic curve, 0.972 vs area under the curve, 0.824). 2 The authors' novel use of real-time intraoperative data to predict upcoming need for a massive transfusion is intriguing and pushes the limits of most current AI clinical applications.The emerging field of surgical data science is transforming surgery and care of the surgical patient. 3 The advent of preoperative risk calculators and postoperative prediction tools for various potential surgical complications has been studied, yet correlation with actual patient outcomes in real-time remain futuristic, limiting informative prognostic discussions with patients and their families at bedside. 4 Encouragingly, more sophisticated AI predictive surgical models and safety systems have become available to help guide difficult goals-of-care discussions with patients and are able to codify intraoperative team communication, real-time patient monitoring, and intraoperative video analysis. 3,5 However, real-time prognostication for the surgical patient for most outcomes of interest remain far off.The importance of having the ability to predict resource needs rapidly, before an event occurs, is as important now as ever. At a tertiary referral center, resources like massive transfusion protocols, blood availability, and surgical expertise can be quickly mobilized in the setting of an acute surgical emergency. 6,7 However, these resources are not universally available. In settings where immediate support is not present, such as rural or resource-limited settings, having lead time to predict an event like the need for massive transfusion may well mean the difference between life and death for a patient. Once the authors' algorithm has been clinically validated with a broader patient population, we think the use of this AI predictive model could be helpful in resource-limited hospital settings, prompting resource allocation prior to an event occurring, rather than playing catch-up. Even more importantly, knowing a patient is at risk of an event, before the event occurs, may encourage and enhance communication between surgical and anesthesia teams helping mitigate the event entirely, or minimizing any untoward downstream effects.There are important limitations to this study to keep in mind. There was lack of granular detail o...

show abstract

Advancing Prediction of Risk of Intraoperative Massive Blood Transfusion in Liver Transplantation With Machine Learning Models. A Multicenter Retrospective Study

Cited by 8 publications

References 38 publications

Machine learning‐based prediction of mortality after heart transplantation in adults with congenital heart disease: A UNOS database analysis

Machine learning‐based prediction of mortality after heart transplantation in adults with congenital heart disease: A UNOS database analysis

Machine Learning Models for Data-Driven Prediction of Diabetes by Lifestyle Type

Novel Use of a Real-Time Prediction Model to Enhance Early Detection of Need for Massive Transfusion—Artificial Intelligence Behind the Drapes

Contact Info

Product

Resources

About