FIGURE 1 - uploaded by Shamin Mohd Saffian
Content may be subject to copyright.
Scatter plots of observed maintenance dose versus predicted maintenance dose for each dosing method. The solid blue line is the line of identity. The red line is a loess fitted function. Top rows are algorithms based only on patient characteristics, middle rows are algorithms based on patient characteristics and INR response, and bottom is a Bayesian method.
Source publication
The aim of this study was to compare the predictive performance of different warfarin dosing methods.
Data from 46 patients who were initiating warfarin therapy were available for analysis. Nine recently published dosing tools including 8 dose prediction algorithms and a Bayesian forecasting method were compared to each other in terms of their abil...
Citations
... Warfarin dosing prediction algorithms are commonly used by clinicians to optimize treatment and reduce the unpredictability of warfarin responses [9,10]. Recently, machine learning methods have been used for developing and validating algorithms that leverage patient information to guide warfarin dosing and facilitate individualized treatment [11][12][13][14][15][16]. ...
Background
Warfarin dosing in cardiac surgery patients is complicated by a heightened sensitivity to the drug, predisposing patients to adverse events. Predictive algorithms are therefore needed to guide warfarin dosing in cardiac surgery patients.
Objective
This study aimed to develop and validate an algorithm for predicting the warfarin dose needed to attain a therapeutic international normalized ratio (INR) at the time of discharge in cardiac surgery patients.
Methods
We abstracted variables influencing warfarin dosage from the records of 1031 encounters initiating warfarin between April 1, 2011, and November 29, 2019, at St Michael’s Hospital in Toronto, Ontario, Canada. We compared the performance of penalized linear regression, k-nearest neighbors, random forest regression, gradient boosting, multivariate adaptive regression splines, and an ensemble model combining the predictions of the 5 regression models. We developed and validated separate models for predicting the warfarin dose required for achieving a discharge INR of 2.0-3.0 in patients undergoing all forms of cardiac surgery except mechanical mitral valve replacement and a discharge INR of 2.5-3.5 in patients receiving a mechanical mitral valve replacement. For the former, we selected 80% of encounters (n=780) who had initiated warfarin during their hospital admission and had achieved a target INR of 2.0-3.0 at the time of discharge as the training cohort. Following 10-fold cross-validation, model accuracy was evaluated in a test cohort comprised solely of cardiac surgery patients. For patients requiring a target INR of 2.5-3.5 (n=165), we used leave-p-out cross-validation (p=3 observations) to estimate model performance. For each approach, we determined the mean absolute error (MAE) and the proportion of predictions within 20% of the true warfarin dose. We retrospectively evaluated the best-performing algorithm in clinical practice by comparing the proportion of cardiovascular surgery patients discharged with a therapeutic INR before (April 2011 and July 2019) and following (September 2021 and May 2, 2022) its implementation in routine care.
Results
Random forest regression was the best-performing model for patients with a target INR of 2.0-3.0, an MAE of 1.13 mg, and 39.5% of predictions of falling within 20% of the actual therapeutic discharge dose. For patients with a target INR of 2.5-3.5, the ensemble model performed best, with an MAE of 1.11 mg and 43.6% of predictions being within 20% of the actual therapeutic discharge dose. The proportion of cardiovascular surgery patients discharged with a therapeutic INR before and following implementation of these algorithms in clinical practice was 47.5% (305/641) and 61.1% (11/18), respectively.
Conclusions
Machine learning algorithms based on routinely available clinical data can help guide initial warfarin dosing in cardiac surgery patients and optimize the postsurgical anticoagulation of these patients.
... To improve treatment quality and shorten the time spent on warfarin dose adjustment, many prediction algorithms and models have been proposed based on clinical variables and pharmacogenetics [4][5][6][7][8]. The International Warfarin Pharmacogenetics Consortium (IWPC) reported that various clinical factors account for 26% of the dose variability and the influence of CYP2C9 and VKORC1 genotypes accounts for 43% of that variability [9]. ...
To improve the accuracy of warfarin daily dose prediction, we develop an evolutionary synthetic oversampling technique (ESMOTE) with a cocktail ensemble model (CEM) called ESMOTE-CEM. Different from conventional oversampling methods, ESMOTE finds the near-optimal oversampling parameters by evolving the parameter representation based on the pre-predicted warfarin dose and then synthesizes new samples to balance the data. The CEM, which improves the performance of random forest (RF) and boosted regression tree (BRT) models using a hybrid mechanism in the regression calculation, estimates the daily dose of warfarin. We test the ESMOTE-CEM on a dataset of 733 samples collected from the First Affiliated Hospital of Soochow University and the International Warfarin Pharmacogenetics Consortium (IWPC). The results show that ESMOTE outperformed the other oversampling methods by at least 6.98% for R² and 5.03% for the mean squared error (MSE). In terms of the percentage of patients whose predicted warfarin dose is within 20% of the actual stable therapeutic dose (20%-p value), the ESMOTE-CEM achieves a 20%-p value of 50%. Moreover, compared to RF, BRT and AdaBoost models, the CEM is the most suitable base predictive model for ESMOTE.
... And this process is mainly depended on doctor's experience. To improve the predictive accuracy, many works have developed their warfarin dose predictive models [4][5][6][7][8] based on linear regression, such as the well-known predictive model of International Warfarin Pharmacogenetics Consortium (IWPC), Warfarindosing predictive tool, and Yu model, etc. [9][10][11]. Machine learning methods [12] such as boosted regression tree (BRT) [13], artificial neural networks (ANNs) [14,15] and support vector regression (SVR) [16] can provide highly-accurate prediction in warfarin daily dose. ...
Background
Vitamin K antagonist (warfarin) is the most classical and widely used oral anticoagulant with assuring anticoagulant effect, wide clinical indications and low price. Warfarin dosage requirements of different patients vary largely. For warfarin daily dosage prediction, the data imbalance in dataset leads to inaccurate prediction on the patients of rare genotype, who usually have large stable dosage requirement. To balance the dataset of patients treated with warfarin and improve the predictive accuracy, an appropriate partition of majority and minority groups, together with an oversampling method, is required.
Method
To solve the data-imbalance problem mentioned above, we developed a clustering-based oversampling technique denoted as DBCSMOTE, which combines density-based spatial clustering of application with noise (DBCSCAN) and synthetic minority oversampling technique (SMOTE). DBCSMOTE automatically finds the minority groups by acquiring the association between samples in terms of the clinical features/genotypes and the warfarin dosage, and creates an extended dataset by adding the new synthetic samples of majority and minority groups. Meanwhile, two ensemble models, boosted regression tree (BRT) and random forest (RF), which are built on the extended dataset generateed by DBCSMOTE, accomplish the task of warfarin daily dosage prediction.
Results
DBCSMOTE and the comparison methods were tested on the datasets derived from our Hospital and International Warfarin Pharmacogenetics Consortium (IWPC). As the results, DBCSMOTE-BRT obtained the highest R -squared ( R ² ) of 0.424 and the smallest mean squared error (mse) of 1.08. In terms of the percentage of patients whose predicted dose of warfarin is within 20% of the actual stable therapeutic dose (20%- p ), DBCSMOTE-BRT can achieve the largest value of 47.8% among predictive models. The more important thing is that DBCSMOTE saved about 68% computational time to achieve the same or better performance than the Evolutionary SMOTE, which was the best oversampling method in warfarin dose prediction by far. Meanwhile, in warfarin dose prediction, it is discovered that DBCSMOTE is more effective in integrating BRT than RF for warfarin dose prediction.
Conclusion
Our finding is that the genotypes, CYP2C9 and VKORC1, no doubt contribute to the predictive accuracy. It was also discovered left atrium diameter, glutamic pyruvic transaminase and serum creatinine included in the model actually improved the predictive accuracy; When congestive heart failure, diabetes mellitus and valve replacement were absent in DBCSMOTE-BRT/RF, the predictive accuracy of DBCSMOTE-BRT/RF decreased. The oversampling ratio and number of minority clusters have a large impact on the effect of oversampling. According to our test, the predictive accuracy was high when the number of minority clusters was 6 ~ 8. The oversampling ratio for small minority clusters should be large (> 1.2) and for large minority clusters should be small (< 0.2). If the dataset becomes larger, the DBCSMOTE would be re-optimized and its BRT/RF model should be re-trained. DBCSMOTE-BRT/RF outperformed the current commonly-used tool called Warfarindosing. As compared to Evolutionary SMOTE-BRT and RF models, DBCSMOTE-BRT and RF models take only a small computational time to achieve the same or higher performance in many cases. In terms of predictive accuracy, RF is not as good as BRT. However, RF still has a powerful ability in generating a highly accurate model as the dataset increases; the software “WarfarinSeer v2.0” is a test version, which packed DBCSMOTE-BRT/RF. It could be a convenient tool for clinical application in warfarin treatment.
... The potential association between vitamin K intake and coagulation instability has been explored previously in case reports and small retrospective studies 6 . Clinical studies evaluating the effects of vitamin K intake on coagulation parameters in patients using anticoagulants demonstrate that vitamin K intake is directly related to INR 30,31 . ...
Vitamin K is found in higher concentrations in dark green plant and in vegetable oils. The adequate intake of vitamin K is 90 and 120μg/day for adult elderly men and women, respectively. The main function of vitamin K is to act as an enzymatic cofactor for hepatic prothrombin synthesis, blood coagulation factors, and anticoagulant proteins. Prominent among the many available anticoagulants is warfarin, an antagonist of vitamin K, which exerts its anticoagulant effects by inhibiting the synthesis of vitamin K1 and vitamin KH2. From the beginning of the therapy it is necessary that the patients carry out the monitoring through the prothrombin time and the international normalized ratio. However, it is known that very low intake and/or fluctuations in vitamin K intake are as harmful as high consumption. In addition, other foods can interact with warfarin, despite their content of vitamin K. The aim of this study was to gather information on the drug interaction of warfarin with vitamin K and with dietary supplements and other foods.
... As measures of accuracy we also calculated the mean prediction error (MPE), defined as the average of the differences between the predicted and the actual dose, and the mean absolute error (MAE), defined as the average of the absolute value (in the mathematical sense) of the difference between the predicted and the actual doses. MAE is usually reported as a measure of predictive accuracy, but rather is a measure of variability of the difference distribution, with a similar interpretation of root mean square error [28]. ...
We assessed the predictive accuracy of the Warfarin Pharmacogenetics Consortium (IWPC) algorithm in a prospective cohort of 376 high-risk elderly patients (≥65 years) who required new treatment with warfarin for either medical (non valvular atrial fibrillation) or surgical conditions (heart valve replacement), had ≥1 comorbid conditions, and regularly used ≥2 other drugs. Follow-up visits were performed according to clinical practice and lasted for a maximum of 1 year. Two hundred and eighty-three (75%) patients achieved a stable maintenance dose. Warfarin maintenance doses were low on average (median 20.3 mg/week, interquartile range, 14.1–27.7 mg/week) and were substantially overestimated by the IWPC algorithm. Overall the percentage of patients whose predicted dose of warfarin was within 20% of the actual stable dose was equal to 37.5%, (95% CI 32.0–43.3%). IWPC algorithm explained only 31% of the actual warfarin dose variability. Modifications of the IWPC algorithm are needed in high-risk elderly people.
... Рис. 16. Параметры кривой генерации тромбина [86] Lag time (LT) -время задержки (лаг-период), мин; Peak time (PT) -время достижения пика, мин; Peak height (PH) -высота пика, нМ; Endogenous thrombin poteintial (ETP) -потенциал эндогенного тромбина (площадь под кривой), нМ • мин; V -максимальная скорость генерации тромбина -тангенс угла наклона левой прямолинейной части кривой -V = PH / (PT --LT), нМ/мин. ...
ноМенКлатура вариантов поСледовательноСтеЙ днК, рнК и белКов
МЕЖДУНАРОДНЫЕ РЕКОМЕНДАЦИИ ПО ИХ ОПИСАНИЮ
... Рис. 16. Параметры кривой генерации тромбина [86] Lag time (LT) -время задержки (лаг-период), мин; Peak time (PT) -время достижения пика, мин; Peak height (PH) -высота пика, нМ; Endogenous thrombin poteintial (ETP) -потенциал эндогенного тромбина (площадь под кривой), нМ • мин; V -максимальная скорость генерации тромбина -тангенс угла наклона левой прямолинейной части кривой -V = PH / (PT --LT), нМ/мин. ...
... The most common of these methods is the Bayesian algorithm. There are many studies carried out with the Bayesian algorithm [4][5][6][7][8][9][10][11]. When these studies were examined, numerical values of warfarin dosing were tried to be found. ...
... When these studies were examined, numerical values of warfarin dosing were tried to be found. However, in another study performed, a Bayesian algorithm was applied on a dataset obtained from 46 patient data and an attempt was made to estimate the dose over 8 mg [11]. In another study, a decision support system was implemented using Bayesian networks from data obtained from over 3,000 patients [12]. ...
... And this process is mainly depended on doctor's experience. To reducing the number of dosage adjustments, some warfarin dose predictive models, such as International Warfarin Pharmacogenetics Consortium (IWPC) model and Warfarindosing, have been developed with a set of clinical features and pharmacogenetics [4][5][6]. IWPC reported that the influence of CYP2C9 and VKORC1 genotypes can give a large contribution. These models are essentially built by multivariable regression techniques. ...
... To reduce the time and side effects of dosage adjustment, various warfarin dosage prediction tools based on pharmacogenetics and clinical features have been developed [5]- [9], among which regression analysis is prevalent [10]- [12]. However, the estimated regression line can mislead the direction of conclusion because a linear representation of highly complex systems may sometimes be insufficient. ...
An evolutionary ensemble modeling (EEM) method is developed to improve the accuracy of warfarin dose prediction. In EEM, genetic programming (GP) evolves diverse base models, and the genetic algorithm optimizes the parameters of the GP. The EEM model is assembled by using the prepared base models through a technique called “bagging.” In the experiment, a dataset of 289 Chinese patients, which was provided by the First Affiliated Hospital of Soochow University, is used for training, validation, and testing. The EEM model with selected feature groups is benchmarked with four machine-learning methods and three conventional regression models. Results show that the EEM model with the M2+G group, namely age, height, weight, gender, CYP2C9, VKORC1, and amiodarone, presents the largest coefficients of determination (R
<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>
), the highest percentage of the predicted dose within 20% of the actual dose (20%-p), the smallest mean absolute error, mean squared error, and root-mean-squared error on the test set, and the least decrease in R
<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>
from the training set to the test set. In conclusion, the EEM method with M2+G delivers superior performance and can, therefore, be a suitable prediction model of warfarin dose for clinical applications.
