Che Ngufor

Ensemble learning approaches to predicting complications of blood transfusion

Of the 21 million blood components transfused in the United States during 2011, approximately 1 in 414 resulted in complication [1]. Two complications in particular, transfusion-related acute lung injury (TRALI) and transfusion-associated circulatory overload (TACO), are especially concerning. These two alone accounted for 62% of reported transfusion-related fatalities in 2013 [2]. We have previously developed a set of machine learning base models for predicting the likelihood of these adverse reactions, with a goal towards better informing the clinician prior to a transfusion decision. Here we describe recent work incorporating ensemble learning approaches to predicting TACO/TRALI. In particular we describe combining base models via majority voting, stacking of model sets with varying diversity, as well as a resampling/boosting combination algorithm called RUSBoost. We find that while the performance of many models is very good, the ensemble models do not yield significantly better performance in terms of AUC.

Predicting Adverse Reactions to Blood Transfusion

In 2011 approximately 21 million blood components were transfused in the United States, with roughly 1 in 414 causing an adverse reaction [1]. Two adverse reactions in particular, transfusion-related acute lung injury (TRALI) and transfusion-associated circulatory overload (TACO), accounted for 62% of reported transfusion-related fatalities in 2013 [2]. We describe newly developed models for predicting the likelihood of these adverse reactions, with a goal towards better informing the clinician prior to a transfusion decision. Our models include both traditional logistic regression as well as modern machine learning techniques, and incorporate over sampling methods to deal with severe class imbalance. We focus on a minimal set of predictors in order to maximize potential application. Results from 8 models demonstrate AUCs ranging from 0.72 to 0.84, with sensitivities tunable by threshold choice across ranges up to 0.93. Many of the models rank the same predictors amongst the most important, perhaps yielding insight into the mechanisms underlying TRALI and TACO. These models are currently being implemented in a Clinical Decision Support System [3] in perioperative environments at Mayo Clinic.

Multi-task learning with selective cross-task transfer for predicting bleeding and other important patient outcomes

In blood transfusion studies, its is often desirable before a surgical procedure to estimate the likelihood of a patient bleeding, need for blood products, re-operation due to bleeding and other important patient outcomes. Such prediction rules are crucial in allowing for optimal planning, more efficient use of blood bank resources, and identification of high-risk patient cohort for specific perioperative interventions. The goal of this study is to present a simple and efficient algorithm that could estimate the risk of multiple outcomes simultaneously. Specifically, a heterogeneous multi-task learning method is presented for learning important surgical outcomes such as bleeding, intraoperative RBC transfusion, need for ICU care, length of stay and mortality. To improve the performance of the method, a post-learning strategy is implemented to further learn the relationship between the trained tasks by a simple “goodness of fit” measure. Specifically, two tasks are considered similar if the model parameters of one tasks improves predictive performance of the other. This strategy allows tasks to be grouped in clusters where selective cross-task transfer of knowledge is explicitly encouraged. To further improve prediction accuracy, a number of operative measurements or surgical outcomes whose predictions are not of direct interest are incorporated in the multi-task model as supplementary tasks to donate information and help the performance of relevant tasks. Results for predicting bleeding and need for blood transfusion for patients undergoing non-cardiac operations from an institutional transfusion datamart show that the proposed methods can improve prediction accuracy over standard single-tasks learning methods. Additional experiments on a real public available data set show that the method is accurate and competitive with some existing methods in the literature.

A Heterogeneous Multi-Task Learning for Predicting RBC Transfusion and Perioperative Outcomes

It would be desirable before a surgical procedure to have a prediction rule that could accurately estimate the probability of a patient bleeding, need for blood transfusion, and other important outcomes. Such a prediction rule would allow optimal planning, more efficient use of blood bank resources, and identification of high-risk patient cohort for specific perioperative interventions. The goal of this study is to develop an efficient and accurate algorithm that could estimate the risk of multiple outcomes simultaneously. Specifically, a heterogeneous multi-task learning method is proposed for learning outcomes such as perioperative bleeding, intraoperative RBC transfusion, ICU care, and ICU length of stay. Additional outcomes not normally predicted are incorporated in the model for transfer learning and help improve the performance of relevant outcomes. Results for predicting perioperative bleeding and need for blood transfusion for patients undergoing non-cardiac operations from an institutional transfusion datamart show that the proposed method significantly increases AUC and G-Mean by more than 6% and 5% respectively over standard single-task learning methods.

A Clinical Decision Support System for Preventing Adverse Reactions to Blood Transfusion

During 2011 approximately 21 million blood components were transfused in the United States, with roughly 1 in 414 resulting in complication. For Americans, the two leading causes of transfusion-related death are the respiratory complications Transfusion-related acute lung injury (TRALI) and Transfusion-associated circulatory overload (TACO). Each of these complications results in significantly longer ICU and hospital stays as well as significantly greater rates of mortality. We have developed a set of machine learning models for predicting the likelihood of these adverse reactions in surgical populations. Here we describe deploying these models into a perioperative critical care environment via a continuous monitoring and alerting clinical decision support system. The goal of this system, which directly integrates our suite of machine learning models running in the R statistical environment into a traditional health information system, is to improve transfusion-related outcomes in the perioperative environment. By identifying high-risk patients prior to transfusion, the clinical team may be able to choose a more appropriate therapy or therapeutic course. Identifying high-risk patients for increased observation after transfusion may also allow for a more timely intervention, thereby potentially improving care delivery and resulting patient outcome. An early prototype of this system is currently running in two Mayo Clinic perioperative environments.

Trajectories of Glycemic Change in a National Cohort of Adults with Previously Controlled Type 2 Diabetes

Background: Individualized diabetes management would benefit from prospectively identifying well-controlled patients at risk of losing glycemic control. Objectives: To identify patterns of hemoglobin A1c (HbA1c) change among patients with stable controlled diabetes. Research Design: Cohort study using OptumLabs Data Warehouse, 2001–2013. We develop and apply a machine learning framework that uses a Bayesian estimation of the mixture of generalized linear mixed effect models to discover glycemic trajectories, and a random forest feature contribution method to identify patient characteristics predictive of their future glycemic trajectories. Subjects: The study cohort consisted of 27,005 US adults with type 2 diabetes, age 18 years and older, and stable index HbA1c <7.0%. Measures: HbA1c values during 24 months of observation. Results: We compared models with k=1, 2, 3, 4, 5 trajectories and baseline variables including patient age, sex, race/ethnicity, comorbidities, medications, and HbA1c. The k=3 model had the best fit, reflecting 3 distinct trajectories of glycemic change: (T1) rapidly deteriorating HbA1c among 302 (1.1%) youngest (mean, 55.2 y) patients with lowest mean baseline HbA1c, 6.05%; (T2) gradually deteriorating HbA1c among 902 (3.3%) patients (mean, 56.5 y) with highest mean baseline HbA1c, 6.53%; and (T3) stable glycemic control among 25,800 (95.5%) oldest (mean, 58.5 y) patients with mean baseline HbA1c 6.21%. After 24 months, HbA1c rose to 8.75% in T1 and 8.40% in T2, but remained stable at 6.56% in T3. Conclusions: Patients with controlled type 2 diabetes follow 3 distinct trajectories of glycemic control. This novel application of advanced analytic methods can facilitate individualized and population diabetes care by proactively identifying high risk patients.

Effects of Plasma Transfusion on Perioperative Bleeding Complications: A Machine Learning Approach.

Perioperative bleeding (PB) is associated with increased patient morbidity and mortality, and results in substantial health care resource utilization. To assess bleeding risk, a routine practice in most centers is to use indicators such as elevated values of the International Normalized Ratio (INR). For patients with elevated INR, the routine therapy option is plasma transfusion. However, the predictive accuracy of INR and the value of plasma transfusion still remains unclear. Accurate methods are therefore needed to identify early the patients with increased risk of bleeding. The goal of this work is to apply advanced machine learning methods to study the relationship between preoperative plasma transfusion (PPT) and PB in patients with elevated INR undergoing noncardiac surgery. The problem is cast under the framework of causal inference where robust meaningful measures to quantify the effect of PPT on PB are estimated. Results show that both machine learning and standard statistical methods generally agree that PPT negatively impacts PB and other important patient outcomes. However, machine learning methods show significant results, and machine learning boosting methods are found to make less errors in predicting PB.

Mixed Effect Machine Learning: a framework for predicting longitudinal change in hemoglobin A1c

Accurate and reliable prediction of clinical progression over time has the potential to improve the outcomes of chronic disease. The classical approach to analyzing longitudinal data is to use (generalized) linear mixed-effect models (GLMM). However, linear parametric models are predicated on assumptions, which are often difficult to verify. In contrast, data-driven machine learning methods can be applied to derive insight from the raw data without a priori assumptions. However, the underlying theory of most machine learning algorithms assume that the data is independent and identically distributed, making them inefficient for longitudinal supervised learning. In this study, we formulate an analytic framework, which integrates the random-effects structure of GLMM into non-linear machine learning models capable of exploiting temporal heterogeneous effects, sparse and varying-length patient characteristics inherent in longitudinal data. We applied the derived mixed-effect machine learning (MEml) framework to predict longitudinal change in glycemic control measured by hemoglobin A1c (HbA1c) among well controlled adults with type 2 diabetes. Results show that MEml is competitive with traditional GLMM, but substantially outperformed standard machine learning models that do not account for random-effects. Specifically, the accuracy of MEml in predicting glycemic change at the 1st, 2nd, 3rd, and 4th clinical visits in advanced was 1.04, 1.08, 1.11, and 1.14 times that of the gradient boosted model respectively, with similar results for the other methods. To further demonstrate the general applicability of MEml, a series of experiments were performed using real publicly available and synthetic data sets for accuracy and robustness. These experiments reinforced the superiority of MEml over the other methods. Overall, results from this study highlight the importance of modeling random-effects in machine learning approaches based on longitudinal data. Our MEml method is highly resistant to correlated data, readily accounts for random-effects, and predicts change of a longitudinal clinical outcome in real-world clinical settings with high accuracy.

Stacked classifiers for individualized prediction of glycemic control following initiation of metformin therapy in type 2 diabetes

OBJECTIVE Metformin is the preferred first-line medication for management of type 2 diabetes and prediabetes. However, over a third of patients experience primary or secondary therapeutic failure. We developed machine learning models to predict which patients initially prescribed metformin will achieve and maintain control of their blood glucose after one year of therapy. MATERIALS AND METHODS We performed a retrospective analysis of administrative claims data for 12,147 commercially-insured adults and Medicare Advantage beneficiaries with prediabetes or diabetes. Several machine learning models were trained using variables available at the time of metformin initiation to predict achievement and maintenance of hemoglobin A1c (HbA1c) < 7.0% after one year of therapy. RESULTS AUC performances based on five-fold cross-validation ranged from 0.58 to 0.75. The most influential variables driving the predictions were baseline HbA1c, starting metformin dosage, and presence of diabetes with complications. CONCLUSIONS Machine learning models can effectively predict primary or secondary metformin treatment failure within one year. This information can help identify effective individualized treatment strategies. Most of the implemented models outperformed traditional logistic regression, highlighting the potential for applying machine learning to problems in medicine.

Multitask LS-Svm for Predicting Bleeding and Re-operation Due to Bleeding

Individualized blood transfusion management would benefit from the ability to prospectively identify patients at risk of complications of blood transfusion, and target them for closer monitoring or intervention. This study presents a simple and efficient multi-task learning method for predicting multiple surgical outcomes based on the weighted least squares support vector machine. To accelerate the training process, the input data is mapped onto a low dimensional randomized feature space leading to a simple linear system that can be solved with any existing fast linear or gradient based methods. Results for predicting early re-operation due to bleeding for patients undergoing non-cardiac operations from an institutional transfusion datamart illustrates that the method can reduce misclassification errors by as much as 13 compared to learning independent models. To further demonstrate the general applicability of the proposed method, a series of experiments are performed on synthetic data sets for scalability and on a real public data set for accuracy and robustness.