⚡ Quick Summary
This study evaluated the effectiveness of machine learning models in predicting biliary complications and mortality rates in liver transplant patients, utilizing a dataset of 1,799 observations. The Random Survival Forest (RSF) model demonstrated the highest performance for both biliary complications and mortality predictions, achieving C-index values of 0.699 and 0.784, respectively.
🔍 Key Details
- 📊 Dataset: 1,799 observations with 40 outcome predictors
- 🧩 Features used: Various clinical parameters including graft types, recipient BMI, and post-transplant lab values
- ⚙️ Technology: Seven machine learning algorithms including LASSO, Ridge, and RSF
- 🏆 Performance: RSF with Ridge (C-index: 0.699 for BC), RSF with RSF (C-index: 0.784 for mortality)
🔑 Key Takeaways
- 📊 Biliary complications are a significant cause of mortality post-liver transplantation.
- 💡 Machine learning models can effectively predict these complications and mortality rates.
- 👩🔬 Key predictors for biliary complications include LT graft types and recipient’s BMI.
- 🏆 Mortality predictors include post-transplant AST, creatinine, and recipient’s age.
- 🤖 Data preprocessing involved feature scaling and one-hot encoding to enhance model performance.
- 🌍 Study conducted with a median follow-up of 19 months, providing robust longitudinal data.
- 🆔 Hyperparameter tuning and random oversampling were crucial for addressing data imbalance.
📚 Background
Liver transplantation is a life-saving procedure for patients with end-stage liver disease. However, the survival rates post-transplantation have stagnated, primarily due to complications such as biliary issues. Understanding the risk factors associated with these complications is essential for improving patient outcomes and enhancing decision-making in clinical settings.
🗒️ Study
The study analyzed longitudinal data from liver transplant patients, focusing on the prediction of biliary complications and mortality rates. Utilizing a comprehensive dataset, researchers employed various machine learning algorithms to assess the effectiveness of different models in predicting these critical outcomes.
📈 Results
The findings revealed that the Random Survival Forest (RSF) model, when combined with Ridge regression, achieved the highest performance for predicting biliary complications (C-index: 0.699). For mortality prediction, the RSF model alone reached a C-index of 0.784. These results underscore the potential of machine learning in enhancing predictive accuracy in clinical settings.
🌍 Impact and Implications
The implications of this study are significant for the field of liver transplantation. By identifying key predictors of biliary complications and mortality, healthcare professionals can make more informed decisions, ultimately improving patient outcomes. The integration of machine learning technologies into clinical practice could lead to more personalized care and better management of liver transplant patients.
🔮 Conclusion
This research highlights the transformative potential of machine learning in predicting biliary complications and mortality in liver transplant patients. By leveraging advanced algorithms, clinicians can enhance their decision-making processes, leading to improved patient care. Continued exploration in this area is essential for further advancements in liver transplantation outcomes.
💬 Your comments
What are your thoughts on the use of machine learning in predicting complications in liver transplant patients? We invite you to share your insights! 💬 Leave your comments below or connect with us on social media:
Comparison of time-to-event machine learning models in predicting biliary complication and mortality rate in liver transplant patients.
Abstract
Post-Liver transplantation (LT) survival rates stagnate, with biliary complications (BC) as a major cause of death. We analyzed longitudinal data with a median 19-month follow-up. BC was diagnosed with ultrasounds and MRCP. Missing data was imputed using mean and median. Data preprocessing involved feature scaling and one-hot encoding. Survival analysis used filter (Cox-P, Cox-c) and embedded (RSF, LASSO) feature selection methods. Seven survival machine learning algorithms were used: LASSO, Ridge, RSF, E-NET, GBS, C-GBS, and FS-SVM. Model development employed 5-fold cross-validation, random oversampling, and hyperparameter tuning. Random oversampling addressed data imbalance. Optimal hyperparameters were determined based on average C-index. Features importance was assessed using standardized regression coefficients and permutation importance for top models. Stability was evaluated using 5-fold cross-validation standard deviation. Finally, 1799 observations with 40 outcome predictors were included. RSF with Ridge achieved the highest performance (C-index: 0.699) for BC prediction, while RSF with RSF had the highest performance (C-index: 0.784) for mortality prediction. Top BC predictors were LT graft types, IBD in recipients, recipient’s BMI, recipient’s history of PVT, and previous LT history. For mortality, they were post-transplant AST, creatinine, recipient’s age, post-transplant ALT, and tacrolimus consumption. We identified BC and mortality risk factors, improving decision-making and outcomes.
Author: [‘Andishgar A’, ‘Bazmi S’, ‘Lankarani KB’, ‘Taghavi SA’, ‘Imanieh MH’, ‘Sivandzadeh G’, ‘Saeian S’, ‘Dadashpour N’, ‘Shamsaeefar A’, ‘Ravankhah M’, ‘Deylami HN’, ‘Tabrizi R’, ‘Imanieh MH’]
Journal: Sci Rep
Citation: Andishgar A, et al. Comparison of time-to-event machine learning models in predicting biliary complication and mortality rate in liver transplant patients. Comparison of time-to-event machine learning models in predicting biliary complication and mortality rate in liver transplant patients. 2025; 15:4768. doi: 10.1038/s41598-025-89570-4