Machine Learning for the Prediction of Red Blood Cell Transfusion in Patients During or After Liver Transplantation Surgery.
Liu, L. P., et al. (2021).Frontiers in Medicine 8: 632210.
Aims
This study aimed to establish critical preoperative risk factors linked to red blood cell (RBC) transfusion, and to develop and validate machine learning algorithms for predicting the RBC transfusion during or after liver transplantation.
Interventions
The study cohort was randomly divided into two sets: the training set and the validation set.
Participants
1193 liver transplant patients.
Outcomes
The study identified key risk factors linked to RBC transfusion during or after liver transplantation, and developed an RBC transfusion prediction model using machine learning algorithms.
Follow-up
N/A
CET Conclusions
This interesting study developed and validated machine learning models for the prediction of transfusion requirements during liver transplantation. The authors used a dataset from 3 Chinese transplant centres to develop and validate their models, and subsequently prospectively validated the resulting prediction tool in a small prospective cohort. The best model derived demonstrated good predictive performance and may have utility in predicting transfusion requirements which may help with pre-operative planning and risk stratification. The authors should be commended for making their tool publicly available on the web for other centres to validate, and for their attempts to create explainable models showing the weighting of the various factors in the decision-making process. It will be interesting to see how well the tool validates in other populations with different disease mixes and surgical techniques, and whether benefits can be measured in prospective use.
Trial registration
N/A