Dataset Information

Machine learning-based prediction of transfusion.

ABSTRACT: BACKGROUND:The ability to predict transfusions arising during hospital admission might enable economized blood supply management and might furthermore increase patient safety by ensuring a sufficient stock of red blood cells (RBCs) for a specific patient. We therefore investigated the precision of four different machine learning-based prediction algorithms to predict transfusion, massive transfusion, and the number of transfusions in patients admitted to a hospital. STUDY DESIGN AND METHODS:This was a retrospective, observational study in three adult tertiary care hospitals in Western Australia between January 2008 and June 2017. Primary outcome measures for the classification tasks were the area under the curve for the receiver operating characteristics curve, the F1 score, and the average precision of the four machine learning algorithms used: neural networks (NNs), logistic regression (LR), random forests (RFs), and gradient boosting (GB) trees. RESULTS:Using our four predictive models, transfusion of at least 1 unit of RBCs could be predicted rather accurately (sensitivity for NN, LR, RF, and GB: 0.898, 0.894, 0.584, and 0.872, respectively; specificity: 0.958, 0.966, 0.964, 0.965). Using the four methods for prediction of massive transfusion was less successful (sensitivity for NN, LR, RF, and GB: 0.780, 0.721, 0.002, and 0.797, respectively; specificity: 0.994, 0.995, 0.993, 0.995). As a consequence, prediction of the total number of packed RBCs transfused was also rather inaccurate. CONCLUSION:This study demonstrates that the necessity for intrahospital transfusion can be forecasted reliably, however the amount of RBC units transfused during a hospital stay is more difficult to predict.

SUBMITTER: Mitterecker A

PROVIDER: S-EPMC7540018 | biostudies-literature | 2020 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning-based prediction of transfusion.

Mitterecker Andreas A Hofmann Axel A Trentino Kevin M KM Lloyd Adam A Leahy Michael F MF Schwarzbauer Karin K Tschoellitsch Thomas T Böck Carl C Hochreiter Sepp S Meier Jens J

Transfusion 20200628 9

<h4>Background</h4>The ability to predict transfusions arising during hospital admission might enable economized blood supply management and might furthermore increase patient safety by ensuring a sufficient stock of red blood cells (RBCs) for a specific patient. We therefore investigated the precision of four different machine learning-based prediction algorithms to predict transfusion, massive transfusion, and the number of transfusions in patients admitted to a hospital.<h4>Study design and m ...[more]

PMID: 32596877

Similar Datasets

Project description:ObjectivePostoperative red blood cell (RBC) transfusion is widely used during the perioperative period but is often associated with a high risk of infection and complications. However, prediction models for RBC transfusion in patients with orthopedic surgery have not yet been developed. We aimed to identify predictors and constructed prediction models for RBC transfusion after orthopedic surgery using interpretable machine learning algorithms.MethodsThis retrospective cohort study reviewed a total of 59,605 patients undergoing orthopedic surgery from June 2013 to January 2019 across 7 tertiary hospitals in China. Patients were randomly split into training (80%) and test subsets (20%). The feature selection method of recursive feature elimination (RFE) was used to identify an optimal feature subset from thirty preoperative variables, and six machine learning algorithms were applied to develop prediction models. The Shapley Additive exPlanations (SHAP) value was employed to evaluate the contribution of each predictor towards the prediction of postoperative RBC transfusion. For simplicity of the clinical utility, a risk score system was further established using the top risk factors identified by machine learning models.ResultsOf the 59,605 patients with orthopedic surgery, 19,921 (33.40%) underwent postoperative RBC transfusion. The CatBoost model exhibited an AUC of 0.831 (95% CI: 0.824-0.836) on the test subset, which significantly outperformed five other prediction models. The risk of RBC transfusion was associated with old age (>60 years) and low RBC count (<4.0 × 1012/L) with clear threshold effects. Extremes of BMI, low albumin, prolonged activated partial thromboplastin time, repair and plastic operations on joint structures were additional top predictors for RBC transfusion. The risk score system derived from six risk factors performed well with an AUC of 0.801 (95% CI: 0.794-0.807) on the test subset.ConclusionBy applying an interpretable machine learning framework in a large-scale multicenter retrospective cohort, we identified novel modifiable risk factors and developed prediction models with good performance for postoperative RBC transfusion in patients undergoing orthopedic surgery. Our findings may allow more precise identification of high-risk patients for optimal control of risk factors and achieve personalized RBC transfusion for orthopedic patients.

Project description:BackgroundAccurate estimation of surgical transfusion risk is essential for efficient allocation of blood bank resources and for other aspects of anesthetic planning. This study hypothesized that a machine learning model incorporating both surgery- and patient-specific variables would outperform the traditional approach that uses only procedure-specific information, allowing for more efficient allocation of preoperative type and screen orders.MethodsThe American College of Surgeons National Surgical Quality Improvement Program Participant Use File was used to train four machine learning models to predict the likelihood of red cell transfusion using surgery-specific and patient-specific variables. A baseline model using only procedure-specific information was created for comparison. The models were trained on surgical encounters that occurred at 722 hospitals in 2016 through 2018. The models were internally validated on surgical cases that occurred at 719 hospitals in 2019. Generalizability of the best-performing model was assessed by external validation on surgical cases occurring at a single institution in 2020.ResultsTransfusion prevalence was 2.4% (73,313 of 3,049,617), 2.2% (23,205 of 1,076,441), and 6.7% (1,104 of 16,053) across the training, internal validation, and external validation cohorts, respectively. The gradient boosting machine outperformed the baseline model and was the best- performing model. At a fixed 96% sensitivity, this model had a positive predictive value of 0.06 and 0.21 and recommended type and screens for 36% and 30% of the patients in internal and external validation, respectively. By comparison, the baseline model at the same sensitivity had a positive predictive value of 0.04 and 0.144 and recommended type and screens for 57% and 45% of the patients in internal and external validation, respectively. The most important predictor variables were overall procedure-specific transfusion rate and preoperative hematocrit.ConclusionsA personalized transfusion risk prediction model was created using both surgery- and patient-specific variables to guide preoperative type and screen orders and showed better performance compared to the traditional procedure-centric approach.Editor’s perspective

Project description:Objective The study aims to establish and validate an effective CT-based radiation pneumonitis (RP) prediction model using the multiomics method of radiomics and EQD2-based dosiomics. Materials and Methods The study performed a retrospective analysis on 91 nonsmall cell lung cancer patients who received radiotherapy from 2019 to 2021 in our hospital. The patients with RP grade ≥1 were labeled as 1, and those with RP grade < 1 were labeled as 0. The whole lung excluding clinical target volume (lung-CTV) was used as the region of interest (ROI). The radiomic and dosiomic features were extracted from the lung-CTV area's image and dose distribution. Besides, the equivalent dose of the 2 Gy fractionated radiation (EQD2) model was used to convert the physical dose to the isoeffect dose, and then, the EQD2-based dosiomic (eqd-dosiomic) features were extracted from the isoeffect dose distribution. Four machine learning (ML) models, including DVH, radiomics combined with DVH (radio + DVH), radiomics combined with dosiomics (radio + dose), and radiomics combined with eqd-dosiomics (radio + eqdose), were established to construct the prediction model via eleven different classifiers. The fivefold cross-validation was used to complete the classification experiment. The area under the curve (AUC) of the receiver operating characteristics (ROC), accuracy, precision, recall, and F1-score were calculated to assess the performance level of the prediction models. Results Compared with the DVH, radio + DVH, and radio + dose model, the value of the training AUC, accuracy, and F1-score of radio + eqdose was higher, and the difference was statistically significant (p < 0.05). Besides, the average value of the precision and recall of radio + eqdose was higher, but the difference was not statistically significant (p > 0.05). Conclusion The performance of using the ML-based multiomics method of radiomics and eqd-dosiomics to predict RP is more efficient and effective.

Dataset Information

Machine learning-based prediction of transfusion.

Publications

Machine learning-based prediction of transfusion.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets