Ensemble of machine learning algorithms using the stacked generalization approach to estimate the warfarin dose

Venous thromboembolism in the era of machine learning and artificial intelligence in medicine

In this review, we embark on a comprehensive exploration of venous thromboembolism (VTE) in the context of medical history and its current practice within medicine. We delve into the landscape of artificial intelligence (AI), exploring its present utility and envisioning its transformative roles within VTE management, from prevention to screening and beyond. Central to our discourse is a forward-looking perspective on the integration of AI within VTE in medicine, advocating for rigorous study design, robust validation processes, and meticulous statistical analysis to gauge the efficacy of AI applications. We further illuminate the potential of large language models and generative AI in revolutionizing VTE care, while acknowledging their inherent limitations and proposing innovative solutions to overcome challenges related to data availability and integrity, including the strategic use of synthetic data. The critical importance of navigating ethical, legal, and privacy concerns associated with AI is underscored, alongside the imperative for comprehensive governance and policy frameworks to regulate its deployment in VTE treatment. We conclude on a note of cautious optimism, where we highlight the significance of proactively addressing the myriad challenges that accompany the advent of AI in healthcare. Through diligent design, stringent validation, extensive education, and prudent regulation, we can harness AI's potential to significantly enhance our understanding and management of VTE. As we stand on the cusp of a new era, our commitment to these principles will be instrumental in ensuring that the promise of AI is fully realized within the realm of VTE care.

The Role of Vitamin D Metabolism Genes and Their Genomic Background in Shaping Cyclosporine A Dosage Parameters after Kidney Transplantation

Background: Kidney transplantation is followed by immunosuppressive therapy involving calcineurin inhibitors (CNIs) such as cyclosporin A. However, long-term high CNIs doses can lead to vitamin D deficiency, and genetic variations influencing vitamin D levels can indirectly impact the necessary CNIs dosage. This study investigates the impact of genetic variations of vitamin D binding protein (DBP) rs2282679 and CYP2R1 hydroxylase rs10741657 polymorphisms on the cyclosporin A dosage in kidney transplant recipients. Additional polymorphisims of genes that are predicted to influence the pharmacogenetic profile were included. Methods: Gene polymorphisms in 177 kidney transplant recipients were analyzed using data mining techniques, including the Random Forest algorithm and Classification and Regression Trees (C&RT). The relationship between the concentration/dose (C/D) ratio of cyclosporin A and genetic profiles was assessed to determine the predictive value of DBP rs2282679 and CYP2R1 rs10741657 polymorphisms. Results: Polymorphic variants of the DBP (rs2282679) demonstrated a strong predictive value for the cyclosporin A C/D ratio in post-kidney transplantation patients. By contrast, the CYP2R1 polymorphism (rs10741657) did not show predictive significance. Additionally, the immune response genes rs231775 CTLA4 and rs1800896 IL10 were identified as predictors of cyclosporin A response, though these did not result in statistically significant differences. Conclusions:DBP rs2282679 polymorphisms can significantly predict the cyclosporin A C/D ratio, potentially enhancing the accuracy of CNI dosing. This can help identify patient groups at risk of vitamin D deficiency, ultimately improving the management of kidney transplant recipients. Understanding these genetic influences allows for more personalized and effective treatment strategies, contributing to better long-term outcomes for patients.

Comprehensive hepatotoxicity prediction: ensemble model integrating machine learning and deep learning

Background

Chemicals may lead to acute liver injuries, posing a serious threat to human health. Achieving the precise safety profile of a compound is challenging due to the complex and expensive testing procedures. In silico approaches will aid in identifying the potential risk of drug candidates in the initial stage of drug development and thus mitigating the developmental cost.

Methods

In current studies, QSAR models were developed for hepatotoxicity predictions using the ensemble strategy to integrate machine learning (ML) and deep learning (DL) algorithms using various molecular features. A large dataset of 2588 chemicals and drugs was randomly divided into training (80%) and test (20%) sets, followed by the training of individual base models using diverse machine learning or deep learning based on three different kinds of descriptors and fingerprints. Feature selection approaches were employed to proceed with model optimizations based on the model performance. Hybrid ensemble approaches were further utilized to determine the method with the best performance.

Results

The voting ensemble classifier emerged as the optimal model, achieving an excellent prediction accuracy of 80.26%, AUC of 82.84%, and recall of over 93% followed by bagging and stacking ensemble classifiers method. The model was further verified by an external test set, internal 10-fold cross-validation, and rigorous benchmark training, exhibiting much better reliability than the published models.

Conclusion

The proposed ensemble model offers a dependable assessment with a good performance for the prediction regarding the risk of chemicals and drugs to induce liver damage.

Leveraging AI techniques for predicting spatial distribution and determinants of carbon emission in China's Yangtze River Delta

This study focuses on the prediction and management of carbon emissions (CE) under the backdrop of global warming, with a particular emphasis on developing spatial planning strategies for urban clusters. In this context, we integrate artificial intelligence technologies to devise an optimized spatial analysis method based on the attributes of multi-source, urban-level spatio-temporal big data on CE. This method enhances both the accuracy and interpretability of CE data processing. Our objectives are to accurately analyze the current status of CE, predict the future spatial distribution of urban CE in the Yangtze River Delta (YRD), and identify key driving factors. We aim to provide pragmatic recommendations for sustainable urban carbon management planning. The findings indicate that: (1) the algorithm designed by us demonstrates excellent fitting capabilities in the analysis of CE data in the YRD, achieving a fitting accuracy of 0.93; (2) it is predicted that from 2025 to 2030, areas with higher CE in the YRD will be primarily concentrated in the 'Provincial Capital Belt' and the 'Heavy Industry Belt'; (3) the economic foundation has been identified as the most significant factor influencing CE in the YRD; (4) projections suggest that CE in the YRD are likely to peak by 2030.

The Application of Artificial Intelligence in Atrial Fibrillation Patients: From Detection to Treatment

Atrial fibrillation (AF) is the most prevalent arrhythmia worldwide. Although the guidelines for AF have been updated in recent years, its gradual onset and associated risk of stroke pose challenges for both patients and cardiologists in real-world practice. Artificial intelligence (AI) is a powerful tool in image analysis, data processing, and for establishing models. It has been widely applied in various medical fields, including AF. In this review, we focus on the progress and knowledge gap regarding the use of AI in AF patients and highlight its potential throughout the entire cycle of AF management, from detection to drug treatment. More evidence is needed to demonstrate its ability to improve prognosis through high-quality randomized controlled trials.

Application of artificial intelligence in the diagnosis and treatment of cardiac arrhythmia

The rapid growth in computational power, sensor technology, and wearable devices has provided a solid foundation for all aspects of cardiac arrhythmia care. Artificial intelligence (AI) has been instrumental in bringing about significant changes in the prevention, risk assessment, diagnosis, and treatment of arrhythmia. This review examines the current state of AI in the diagnosis and treatment of atrial fibrillation, supraventricular arrhythmia, ventricular arrhythmia, hereditary channelopathies, and cardiac pacing. Furthermore, ChatGPT, which has gained attention recently, is addressed in this paper along with its potential applications in the field of arrhythmia. Additionally, the accuracy of arrhythmia diagnosis can be improved by identifying electrode misplacement or erroneous swapping of electrode position using AI. Remote monitoring has expanded greatly due to the emergence of contactless monitoring technology as wearable devices continue to develop and flourish. Parallel advances in AI computing power, ChatGPT, availability of large data sets, and more have greatly expanded applications in arrhythmia diagnosis, risk assessment, and treatment. More precise algorithms based on big data, personalized risk assessment, telemedicine and mobile health, smart hardware and wearables, and the exploration of rare or complex types of arrhythmia are the future direction.

Warfarin-A natural anticoagulant: A review of research trends for precision medication

BACKGROUND

Warfarin is a widely prescribed anticoagulant in the clinic. It has a more considerable individual variability, and many factors affect its variability. Mathematical models can quantify the quantitative impact of these factors on individual variability.

PURPOSE

The aim is to comprehensively analyze the advanced warfarin dosing algorithm based on pharmacometrics and machine learning models of personalized warfarin dosage.

METHODS

A bibliometric analysis of the literature retrieved from PubMed and Scopus was performed using VOSviewer. The relevant literature that reported the precise dosage of warfarin calculation was retrieved from the database. The multiple linear regression (MLR) algorithm was excluded because a recent systematic review that mainly reviewed this algorithm has been reported. The following terms of quantitative systems pharmacology, mechanistic model, physiologically based pharmacokinetic model, artificial intelligence, machine learning, pharmacokinetic, pharmacodynamic, pharmacokinetics, pharmacodynamics, and warfarin were added as MeSH Terms or appearing in Title/Abstract into query box of PubMed, then humans and English as filter were added to retrieve the literature.

RESULTS

Bibliometric analysis revealed important co-occuring MeShH and index keywords. Further, the United States, China, and the United Kingdom were among the top countries contributing in this domain. Some studies have established personalized warfarin dosage models using pharmacometrics and machine learning-based algorithms. There were 54 related studies, including 14 pharmacometric models, 31 artificial intelligence models, and 9 model evaluations. Each model has its advantages and disadvantages. The pharmacometric model contains biological or pharmacological mechanisms in structure. The process of pharmacometric model development is very time- and labor-intensive. Machine learning is a purely data-driven approach; its parameters are more mathematical and have less biological interpretation. However, it is faster, more efficient, and less time-consuming. Most published models of machine learning algorithms were established based on cross-sectional data sourced from the database.

CONCLUSION

Future research on personalized warfarin medication should focus on combining the advantages of machine learning and pharmacometrics algorithms to establish a more robust warfarin dosage algorithm. Randomized controlled trials should be performed to evaluate the established algorithm of warfarin dosage. Moreover, a more user-friendly and accessible warfarin precision medicine platform should be developed.

Prediction of naloxone dose in opioids toxicity based on machine learning techniques (artificial intelligence)

BACKGROUND

Treatment management for opioid poisoning is critical and, at the same time, requires specialized knowledge and skills. This study was designed to develop and evaluate machine learning algorithms for predicting the maintenance dose and duration of hospital stay in opioid poisoning, in order to facilitate appropriate clinical decision-making.

METHOD AND RESULTS

This study used artificial intelligence technology to predict the maintenance dose and duration of administration by selecting clinical and paraclinical features that were selected by Pearson correlation (filter method) (Stage 1) and then the (wrapper method) Recursive Feature Elimination Cross-Validated (RFECV) (Stage2). The duration of administration was divided into two categories: A (which includes a duration of less than or equal to 24 h of infusion) and B (more than 24 h of naloxone infusion). XGBoost algorithm model with an accuracy rate of 91.04%, a prediction rate of 91.34%, and a sensitivity rate of 91.04% and area under the Curve (AUC) 0.97 was best model for classification patients. Also, the best maintenance dose of naloxone was obtained with XGBoost algorithm with R2 = 0.678. Based on the selected algorithm, the most important features for classifying patients for the duration of treatment were bicarbonate, respiration rate, physical sign, The partial pressure of carbon dioxide (PCO2), diastolic blood pressure, pulse rate, naloxone bolus dose, Blood Creatinine(Cr), Body temperature (T). The most important characteristics for determining the maintenance dose of naloxone were physical signs, bolus dose of 4.5 mg/kg, Glasgow Coma Scale (GCS), Creatine Phosphokinase (CPK) and intensive care unit (ICU) add.

CONCLUSION

A predictive model can significantly enhance the decision-making and clinical care provided by emergency physicians in hospitals and medical settings. XGBoost was found to be the superior model.

Machine Learning-Based Predictive Models for Patients with Venous Thromboembolism: A Systematic Review

BACKGROUND

 Venous thromboembolism (VTE) is a chronic disorder with a significant health and economic burden. Several VTE-specific clinical prediction models (CPMs) have been used to assist physicians in decision-making but have several limitations. This systematic review explores if machine learning (ML) can enhance CPMs by analyzing extensive patient data derived from electronic health records. We aimed to explore ML-CPMs' applications in VTE for risk stratification, outcome prediction, diagnosis, and treatment.

METHODS

 Three databases were searched: PubMed, Google Scholar, and IEEE electronic library. Inclusion criteria focused on studies using structured data, excluding non-English publications, studies on non-humans, and certain data types such as natural language processing and image processing. Studies involving pregnant women, cancer patients, and children were also excluded. After excluding irrelevant studies, a total of 77 studies were included.

RESULTS

 Most studies report that ML-CPMs outperformed traditional CPMs in terms of receiver operating area under the curve in the four clinical domains that were explored. However, the majority of the studies were retrospective, monocentric, and lacked detailed model architecture description and external validation, which are essential for quality audit. This review identified research gaps and highlighted challenges related to standardized reporting, reproducibility, and model comparison.

CONCLUSION

 ML-CPMs show promise in improving risk assessment and individualized treatment recommendations in VTE. Apparently, there is an urgent need for standardized reporting and methodology for ML models, external validation, prospective and real-world data studies, as well as interventional studies to evaluate the impact of artificial intelligence in VTE.

Optimizing warfarin dosing for patients with atrial fibrillation using machine learning

While novel oral anticoagulants are increasingly used to reduce risk of stroke in patients with atrial fibrillation, vitamin K antagonists such as warfarin continue to be used extensively for stroke prevention across the world. While effective in reducing the risk of strokes, the complex pharmacodynamics of warfarin make it difficult to use clinically, with many patients experiencing under- and/or over- anticoagulation. In this study we employed a novel implementation of deep reinforcement learning to provide clinical decision support to optimize time in therapeutic International Normalized Ratio (INR) range. We used a novel semi-Markov decision process formulation of the Batch-Constrained deep Q-learning algorithm to develop a reinforcement learning model to dynamically recommend optimal warfarin dosing to achieve INR of 2.0-3.0 for patients with atrial fibrillation. The model was developed using data from 22,502 patients in the warfarin treated groups of the pivotal randomized clinical trials of edoxaban (ENGAGE AF-TIMI 48), apixaban (ARISTOTLE) and rivaroxaban (ROCKET AF). The model was externally validated on data from 5730 warfarin-treated patients in a fourth trial of dabigatran (RE-LY) using multilevel regression models to estimate the relationship between center-level algorithm consistent dosing, time in therapeutic INR range (TTR), and a composite clinical outcome of stroke, systemic embolism or major hemorrhage. External validation showed a positive association between center-level algorithm-consistent dosing and TTR (R2 = 0.56). Each 10% increase in algorithm-consistent dosing at the center level independently predicted a 6.78% improvement in TTR (95% CI 6.29, 7.28; p < 0.001) and a 11% decrease in the composite clinical outcome (HR 0.89; 95% CI 0.81, 1.00; p = 0.015). These results were comparable to those of a rules-based clinical algorithm used for benchmarking, for which each 10% increase in algorithm-consistent dosing independently predicted a 6.10% increase in TTR (95% CI 5.67, 6.54, p < 0.001) and a 10% decrease in the composite outcome (HR 0.90; 95% CI 0.83, 0.98, p = 0.018). Our findings suggest that a deep reinforcement learning algorithm can optimize time in therapeutic range for patients taking warfarin. A digital clinical decision support system to promote algorithm-consistent warfarin dosing could optimize time in therapeutic range and improve clinical outcomes in atrial fibrillation globally.

Improved stacking ensemble learning based on feature selection to accurately predict warfarin dose

Background

With the rapid development of artificial intelligence, prediction of warfarin dose via machine learning has received more and more attention. Since the dose prediction involve both linear and nonlinear problems, traditional machine learning algorithms are ineffective to solve such problems at one time.

Objective

Based on the characteristics of clinical data of Chinese warfarin patients, an improved stacking ensemble learning can achieve higher prediction accuracy.

Methods

Information of 641 patients from southern China who had reached a steady state on warfarin was collected, including demographic information, medical history, genotype, and co-medication status. The dataset was randomly divided into a training set (90%) and a test set (10%). The predictive capability is evaluated on a new test set generated by stacking ensemble learning. Additional factors associated with warfarin dose were discovered by feature selection methods.

Results

A newly proposed heuristic-stacking ensemble learning performs better than traditional-stacking ensemble learning in key metrics such as accuracy of ideal dose (73.44%, 71.88%), mean absolute errors (0.11 mg/day, 0.13 mg/day), root mean square errors (0.18 mg/day, 0.20 mg/day) and R2 (0.87, 0.82).

Conclusions

The developed heuristic-stacking ensemble learning can satisfactorily predict warfarin dose with high accuracy. A relationship between hypertension, a history of severe preoperative embolism, and warfarin dose is found, which provides a useful reference for the warfarin dose administration in the future.

Predicting quetiapine dose in patients with depression using machine learning techniques based on real-world evidence

BACKGROUND

Being one of the most widespread, pervasive, and troublesome illnesses in the world, depression causes dysfunction in various spheres of individual and social life. Regrettably, despite obtaining evidence-based antidepressant medication, up to 70% of people are going to continue to experience troublesome symptoms. Quetiapine, as one of the most commonly prescribed antipsychotic medication worldwide, has been reported as an effective augmentation strategy to antidepressants. The right quetiapine dose and personalized quetiapine treatment are frequently challenging for clinicians. This study aimed to identify important influencing variables for quetiapine dose by maximizing the use of data from real world, and develop a predictive model of quetiapine dose through machine learning techniques to support selections for treatment regimens.

METHODS

The study comprised 308 depressed patients who were medicated with quetiapine and hospitalized in the First Hospital of Hebei Medical University, from November 1, 2019, to August 31, 2022. To identify the important variables influencing the dose of quetiapine, a univariate analysis was applied. The prediction abilities of nine machine learning models (XGBoost, LightGBM, RF, GBDT, SVM, LR, ANN, DT) were compared. Algorithm with the optimal model performance was chosen to develop the prediction model.

RESULTS

Four predictors were selected from 38 variables by the univariate analysis (p < 0.05), including quetiapine TDM value, age, mean corpuscular hemoglobin concentration, and total bile acid. Ultimately, the XGBoost algorithm was used to create a prediction model for quetiapine dose that had the greatest predictive performance (accuracy = 0.69) out of nine models. In the testing cohort (62 cases), a total of 43 cases were correctly predicted of the quetiapine dose regimen. In dose subgroup analysis, AUROC for patients with daily dose of 100 mg, 200 mg, 300 mg and 400 mg were 0.99, 0.75, 0.93 and 0.86, respectively.

CONCLUSIONS

In this work, machine learning techniques are used for the first time to estimate the dose of quetiapine for patients with depression, which is valuable for the clinical drug recommendations.

Functional Outcome Prediction After Spinal Cord Injury Using Ensemble Machine Learning

OBJECTIVES

To establish a machine learning model to predict functional outcomes after SCI with Spinal Cord Independence Measure (SCIM) using features present at the time of rehabilitation admission.

STUDY DESIGN

A retrospective, single-center study. The following data were collected from the medical charts: age, sex, acute length of stay (LOS), level of injury, American Spinal Injury Association Impairment Scale (AIS), motor scores of each key muscle, Upper Extremity Motor Score (UEMS), Lower Extremity Motor Score (LEMS), SCIM total scores, and subtotal scores on admission and discharge. Based on the multivariate linear regression analysis, age, acute LOS, UEMS, LEMS, and SCIM subtotal scores were selected as features for machine learning algorithms. Random forest, support vector machine, neural network, and gradient boosting were used as the base models and combined using ridge regression as a metamodel.

SETTING

A spinal center in Tokyo, Japan.

PARTICIPANTS

Participants were individuals with SCI admitted to our hospital from March 2016 to October 2021 for the first rehabilitation after the injury. They were divided into 2 groups: training (n=140) and testing (n=70).

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

The root-mean-square error (RMSE), R2, and Mean Absolute Error (MAE) were used as accuracy measures.

RESULTS

RMSE, R2, and MAE of the meta-model using the testing group were 9.7453, 0.8835, and 7.4743, respectively, outperforming any other single base model.

CONCLUSIONS

Our study revealed that functional prognostication could be achieved using machine-learning methods with features present at the time of rehabilitation admission. Goals can be set at the beginning of rehabilitation. Moreover, our model can be used to evaluate advanced medical treatments, such as regenerative medicine.

Machine learning in medication prescription: A systematic review

BACKGROUND

Medication prescription is a complex process that could benefit from current research and development in machine learning through decision support systems. Particularly pediatricians are forced to prescribe medications "off-label" as children are still underrepresented in clinical studies, which leads to a high risk of an incorrect dose and adverse drug effects.

METHODS

PubMed, IEEE Xplore and PROSPERO were searched for relevant studies that developed and evaluated well-performing machine learning algorithms following the PRISMA statement. Quality assessment was conducted in accordance with the IJMEDI checklist. Identified studies were reviewed in detail, including the required variables for predicting the correct dose, especially of pediatric medication prescription.

RESULTS

The search identified 656 studies, of which 64 were reviewed in detail and 36 met the inclusion criteria. According to the IJMEDI checklist, five studies were considered to be of high quality. 19 of the 36 studies dealt with the active substance warfarin. Overall, machine learning algorithms based on decision trees or regression methods performed superior regarding their predictive power than algorithms based on neural networks, support vector machines or other methods. The use of ensemble methods like bagging or boosting generally enhanced the accuracy of the dose predictions. The required input and output variables of the algorithms were considerably heterogeneous and differ strongly among the respective substance.

CONCLUSIONS

By using machine learning algorithms, the prescription process could be simplified and dosing correctness could be enhanced. Despite the heterogenous results among the different substances and cases and the lack of pediatric use cases, the identified approaches and required variables can serve as an excellent starting point for further development of algorithms predicting drug doses, particularly for children. Especially the combination of physiologically-based pharmacokinetic models with machine learning algorithms represents a great opportunity to enhance the predictive power and accuracy of the developed algorithms.

Evaluation of stacked ensemble model performance to predict clinical outcomes: A COVID-19 study

BACKGROUND

The application of machine learning (ML) to analyze clinical data with the goal to predict patient outcomes has garnered increasing attention. Ensemble learning has been used in conjunction with ML to improve predictive performance. Although stacked generalization (stacking), a type of heterogeneous ensemble of ML models, has emerged in clinical data analysis, it remains unclear how to define the best model combinations for strong predictive performance. This study develops a methodology to evaluate the performance of "base" learner models and their optimized combination using "meta" learner models in stacked ensembles to accurately assess performance in the context of clinical outcomes.

METHODS

De-identified COVID-19 data was obtained from the University of Louisville Hospital, where a retrospective chart review was performed from March 2020 to November 2021. Three differently-sized subsets using features from the overall dataset were chosen to train and evaluate ensemble classification performance. The number of base learners chosen from several algorithm families coupled with a complementary meta learner was varied from a minimum of 2 to a maximum of 8. Predictive performance of these combinations was evaluated in terms of mortality and severe cardiac event outcomes using area-under-the-receiver-operating-characteristic (AUROC), F1, balanced accuracy, and kappa.

RESULTS

The results highlight the potential to accurately predict clinical outcomes, such as severe cardiac events with COVID-19, from routinely acquired in-hospital patient data. Meta learners Generalized Linear Model (GLM), Multi-Layer Perceptron (MLP), and Partial Least Squares (PLS) had the highest AUROC for both outcomes, while K-Nearest Neighbors (KNN) had the lowest. Performance trended lower in the training set as the number of features increased, and exhibited less variance in both training and validation across all feature subsets as the number of base learners increased.

CONCLUSION

This study offers a methodology to robustly evaluate ensemble ML performance when analyzing clinical data.

A Real-World Exploration into Clinical Outcomes of Direct Oral Anticoagulant Dosing Regimens in Morbidly Obese Patients Using Data-Driven Approaches

INTRODUCTION

The clinical outcomes of direct oral anticoagulant (DOAC) dosage regimens in morbid obesity are uncertain due to limited clinical evidence. This study seeks to bridge this evidence gap by identifying the factors associated with clinical outcomes following the dosing of DOACs in morbidly obese patients.

METHOD

A data-driven observational study was carried out using supervised machine learning (ML) models with a dataset extracted from electronic health records and preprocessed. Following 70%:30% partitioning of the overall dataset via stratified sampling, the selected ML classifiers (e.g., random forest, decision trees, bootstrap aggregation) were applied to the training dataset (70%). The outcomes of the models were evaluated against the test dataset (30%). Multivariate regression analysis explored the association between DOAC regimens and clinical outcomes.

RESULTS

A sample of 4,275 morbidly obese patients was extracted and analysed. The decision trees, random forest, and bootstrap aggregation classifiers achieved acceptable (excellent) values of precision, recall, and F1 scores in terms of their contribution to clinical outcomes. The length of stay, treatment days, and age were ranked highest for relevance to mortality and stroke. Among DOAC regimens, apixaban 2.5 mg twice daily ranked highest for its association with mortality, increasing the mortality risk by 43% (odds ratio [OR] 1.430, 95% confidence interval [CI] 1.181-1.732, p = 0.001). On the other hand, apixaban 5 mg twice daily reduced the odds of mortality by 25% (OR 0.751, 95% CI 0.632-0.905, p = 0.003) but increased the odds of stroke events. No clinically relevant non-major bleeding events occurred in this group.

CONCLUSION

Data-driven approaches can identify key factors associated with clinical outcomes following the dosing of DOACs in morbidly obese patients. This will help design further studies to explore well tolerated and effective DOAC doses for morbidly obese patients.

Azim N, Abd El-Aziz AA. Groundwater Management Based on Time Series and Ensembles of Machine Learning

Due to the necessity of effective water management, the issue of water scarcity has developed into a significant global issue. One way to collect water is through the water management method. The most common source of fresh water anywhere in the world is groundwater, which has developed into a significant global issue. Our previous research used machine learning (ML) for training models to classify groundwater quality. However, in this study, we used the time series and ensemble methods to propose a hybrid technique to enhance the multiclassification of groundwater quality. The proposed technique distinguishes between excellent drinking water, good drinking water, poor irrigation water, and very poor irrigation water. In this research, we used the GEOTHERM dataset, and we pre-processed it by replacing the missing and null values, solving the sparsity problem with our recommender system, which was previously proposed, and applying the synthetic minority oversampling technique (SMOTE). Moreover, we used the Pearson correlation coefficient (PCC) feature selection technique to select the relevant attributes. The dataset was divided into a training set (75%) and a testing set (25%). The time-series algorithm was used in the training phase to learn the four ensemble techniques (random forest (RF), gradient boosting, AdaBoost, and bagging. The four ensemble methods were used in the testing phase to validate the proposed hybrid technique. The experimental results showed that the RF algorithm outperformed the common ensemble methods in terms of multiclassification average precision, recall, disc similarity coefficient (DSC), and accuracy for the groundwater dataset by approximately 98%, 89.25%, 93%, and 95%, respectively. As a result, the evaluation of the proposed model revealed that, compared to other recent models, it produces unmatched tuning-based perception results.

Evaluation of supervised machine learning algorithms in predicting the poor anticoagulation control and stable weekly doses of warfarin

BACKGROUND

Machine learning algorithms (MLAs) carry a huge potential in identifying predicting factors and are being explored for their utility in the field of personalized medicine.

AIM

We aimed to investigate MLAs for identifying predictors (clinical and genetic) of poor anticoagulation status (ACS) and stable weekly warfarin dose (SWWD).

METHOD

Clinical factors, in addition to the CYP2C9, VKORC1, and CYP4F2 genotypes, were obtained for patients receiving warfarin for at least the previous six months. The C5.0 decision tree classification algorithm was used to predict poor ACS while classification and regression tree analysis (CART), in addition to the Chi-square automatic interaction detector (CHAID), was used to predict SWWD. The percentage of patients within 20% of the actual dose, root mean squared error (RMSE), and area under the receiver-operating characteristics curve (AUROC) were identified as performance indicators of the models.

RESULTS

In the C5.0 classification decision tree, the CYP4F2 genotype was the strongest predictor of ACS (AUROC = 0.53). In the CART analysis of SWWD, VKORC1 polymorphisms were the most significant predictor, followed by the CYP2C9 genotype (percentage of patients within 20% of the actual dose = 38.2%, RMSE = 13.6). For the CHAID algorithm, the percentage of patients within 20% of the actual dose was 49%, while the RMSE was found to be 13.4.

CONCLUSION

Genetic and non-genetic predictive factors were identified by the MLAs for ACS and SWWD. Further, the need to externally validate the MLAs in a prospective study was highlighted.

Building and analyzing machine learning-based warfarin dose prediction models using scikit-learn

For personalized drug dosing, prediction models may be utilized to overcome the inter-individual variability. Multiple linear regression has been used as a conventional method to model the relationship between patient features and optimal drug dose. However, linear regression cannot capture non-linear relationships and may be adversely affected by non-normal distribution and collinearity of data. To overcome this hurdle, machine learning models have been extensively adapted in drug dose prediction. In this tutorial, random forest and neural network models will be trained in tandem with a multiple linear regression model on the International Warfarin Pharmacogenetics Consortium dataset using the scikit-learn python library. Subsequent model analyses including performance comparison, permutation feature importance computation and partial dependence plotting will be demonstrated. The basic methods of model training and analysis discussed in this article may be implemented in drug dose-related studies.

Distributed information encoding and decoding using self-organized spatial patterns

Dynamical systems often generate distinct outputs according to different initial conditions, and one can infer the corresponding input configuration given an output. This property captures the essence of information encoding and decoding. Here, we demonstrate the use of self-organized patterns that generate high-dimensional outputs, combined with machine learning, to achieve distributed information encoding and decoding. Our approach exploits a critical property of many natural pattern-formation systems: in repeated realizations, each initial configuration generates similar but not identical output patterns due to randomness in the patterning process. However, for sufficiently small randomness, different groups of patterns that arise from different initial configurations can be distinguished from one another. Modulating the pattern-generation and machine learning model training can tune the tradeoff between encoding capacity and security. We further show that this strategy is scalable by implementing the encoding and decoding of all characters of the standard English keyboard.

Predicting Therapeutic Response to Unfractionated Heparin Therapy: Machine Learning Approach

Background

Unfractionated heparin (UFH) is an anticoagulant drug that is considered a high-risk medication because an excessive dose can cause bleeding, whereas an insufficient dose can lead to a recurrent embolic event. Therapeutic response to the initiation of intravenous UFH is monitored using activated partial thromboplastin time (aPTT) as a measure of blood clotting time. Clinicians iteratively adjust the dose of UFH toward a target, indication-defined therapeutic aPTT range using nomograms, but this process can be imprecise and can take ≥36 hours to achieve the target range. Thus, a more efficient approach is required.

Objective

In this study, we aimed to develop and validate a machine learning (ML) algorithm to predict aPTT within 12 hours after a specified bolus and maintenance dose of UFH.

Methods

This was a retrospective cohort study of 3019 patient episodes of care from January 2017 to August 2020 using data collected from electronic health records of 5 hospitals in Queensland, Australia. Data from 4 hospitals were used to build and test ensemble models using cross-validation, whereas data from the fifth hospital were used for external validation. We built 2 ML models: a regression model to predict the aPTT value after a UFH bolus dose and a multiclass model to predict the aPTT, classified as subtherapeutic (aPTT <70 seconds), therapeutic (aPTT 70-100 seconds), or supratherapeutic (aPTT >100 seconds). Modeling was performed using Driverless AI (H2O), an automated ML tool, and 17 different experiments were iteratively conducted to optimize model accuracy.

Results

In predicting aPTT, the best performing model was an ensemble with 4x LightGBM models with a root mean square error of 31.35 (SD 1.37). In predicting the aPTT class using a repurposed data set, the best performing ensemble model achieved an accuracy of 0.599 (SD 0.0289) and an area under the receiver operating characteristic curve of 0.735. External validation yielded similar results: root mean square error of 30.52 (SD 1.29) for the aPTT prediction model, and accuracy of 0.568 (SD 0.0315) and area under the receiver operating characteristic curve of 0.724 for the aPTT multiclassification model.

Conclusions

To the best of our knowledge, this is the first ML model applied to intravenous UFH dosing that has been developed and externally validated in a multisite adult general medical and surgical inpatient setting. We present the processes of data collection, preparation, and feature engineering for replication.

Warfarin sensitivity is associated with increased hospital mortality in critically Ill patients

Background

Warfarin is a widely used anticoagulant with a narrow therapeutic index and large interpatient variability in the therapeutic dose. Warfarin sensitivity has been reported to be associated with increased incidence of international normalized ratio (INR) > 5. However, whether warfarin sensitivity is a risk factor for adverse outcomes in critically ill patients remains unknown. In the present study, we aimed to evaluate the utility of different machine learning algorithms for the prediction of warfarin sensitivity and to determine the impact of warfarin sensitivity on outcomes in critically ill patients.

Methods

Nine different machine learning algorithms for the prediction of warfarin sensitivity were tested in the International Warfarin Pharmacogenetic Consortium cohort and Easton cohort. Furthermore, a total of 7,647 critically ill patients was analyzed for warfarin sensitivity on in-hospital mortality by multivariable regression. Covariates that potentially confound the association were further adjusted using propensity score matching or inverse probability of treatment weighting.

Results

We found that logistic regression (AUC = 0.879, 95% CI: 0.834–0.924) was indistinguishable from support vector machine with a linear kernel, neural network, AdaBoost and light gradient boosting trees, and significantly outperformed all the other machine learning algorithms. Furthermore, we found that warfarin sensitivity predicted by the logistic regression model was significantly associated with worse in-hospital mortality in critically ill patients with an odds ratio (OR) of 1.33 (95% CI, 1.01–1.77).

Conclusions

Our data suggest that the logistic regression model is the best model for the prediction of warfarin sensitivity clinically and that warfarin sensitivity is likely to be a risk factor for adverse outcomes in critically ill patients.

Nonlinear Machine Learning in Warfarin Dose Prediction: Insights from Contemporary Modelling Studies

Objective: This study aimed to systematically assess the characteristics and risk of bias of previous studies that have investigated nonlinear machine learning algorithms for warfarin dose prediction. Methods: We systematically searched PubMed, Embase, Cochrane Library, Chinese National Knowledge Infrastructure (CNKI), China Biology Medicine (CBM), China Science and Technology Journal Database (VIP), and Wanfang Database up to March 2022. We assessed the general characteristics of the included studies with respect to the participants, predictors, model development, and model evaluation. The methodological quality of the studies was determined, and the risk of bias was evaluated using the Prediction model Risk of Bias Assessment Tool (PROBAST). Results: From a total of 8996 studies, 23 were assessed in this study, of which 23 (100%) were retrospective, and 11 studies focused on the Asian population. The most common demographic and clinical predictors were age (21/23, 91%), weight (17/23, 74%), height (12/23, 52%), and amiodarone combination (11/23, 48%), while CYP2C9 (14/23, 61%), VKORC1 (14/23, 61%), and CYP4F2 (5/23, 22%) were the most common genetic predictors. Of the included studies, the MAE ranged from 1.47 to 10.86 mg/week in model development studies, from 2.42 to 5.18 mg/week in model development with external validation (same data) studies, from 12.07 to 17.59 mg/week in model development with external validation (another data) studies, and from 4.40 to 4.84 mg/week in model external validation studies. All studies were evaluated as having a high risk of bias. Factors contributing to the risk of bias include inappropriate exclusion of participants (10/23, 43%), small sample size (15/23, 65%), poor handling of missing data (20/23, 87%), and incorrect method of selecting predictors (8/23, 35%). Conclusions: Most studies on nonlinear-machine-learning-based warfarin prediction models show poor methodological quality and have a high risk of bias. The analysis domain is the major contributor to the overall high risk of bias. External validity and model reproducibility are lacking in most studies. Future studies should focus on external validity, diminish risk of bias, and enhance real-world clinical relevance.

Deciphering the Interactions of SARS-CoV-2 Proteins with Human Ion Channels Using Machine-Learning-Based Methods

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is accountable for the protracted COVID-19 pandemic. Its high transmission rate and pathogenicity led to health emergencies and economic crisis. Recent studies pertaining to the understanding of the molecular pathogenesis of SARS-CoV-2 infection exhibited the indispensable role of ion channels in viral infection inside the host. Moreover, machine learning (ML)-based algorithms are providing a higher accuracy for host-SARS-CoV-2 protein–protein interactions (PPIs). In this study, PPIs of SARS-CoV-2 proteins with human ion channels (HICs) were trained on the PPI-MetaGO algorithm. PPI networks (PPINs) and a signaling pathway map of HICs with SARS-CoV-2 proteins were generated. Additionally, various U.S. food and drug administration (FDA)-approved drugs interacting with the potential HICs were identified. The PPIs were predicted with 82.71% accuracy, 84.09% precision, 84.09% sensitivity, 0.89 AUC-ROC, 65.17% Matthews correlation coefficient score (MCC) and 84.09% F1 score. Several host pathways were found to be altered, including calcium signaling and taste transduction pathway. Potential HICs could serve as an initial set to the experimentalists for further validation. The study also reinforces the drug repurposing approach for the development of host directed antiviral drugs that may provide a better therapeutic management strategy for infection caused by SARS-CoV-2.

Explainable artificial intelligence (XAI) for exploring spatial variability of lung and bronchus cancer (LBC) mortality rates in the contiguous USA

Machine learning (ML) has demonstrated promise in predicting mortality; however, understanding spatial variation in risk factor contributions to mortality rate requires explainability. We applied explainable artificial intelligence (XAI) on a stack-ensemble machine learning model framework to explore and visualize the spatial distribution of the contributions of known risk factors to lung and bronchus cancer (LBC) mortality rates in the conterminous United States. We used five base-learners-generalized linear model (GLM), random forest (RF), Gradient boosting machine (GBM), extreme Gradient boosting machine (XGBoost), and Deep Neural Network (DNN) for developing stack-ensemble models. Then we applied several model-agnostic approaches to interpret and visualize the stack ensemble model's output in global and local scales (at the county level). The stack ensemble generally performs better than all the base learners and three spatial regression models. A permutation-based feature importance technique ranked smoking prevalence as the most important predictor, followed by poverty and elevation. However, the impact of these risk factors on LBC mortality rates varies spatially. This is the first study to use ensemble machine learning with explainable algorithms to explore and visualize the spatial heterogeneity of the relationships between LBC mortality and risk factors in the contiguous USA.

Key use cases for artificial intelligence to reduce the frequency of adverse drug events: a scoping review

Adverse drug events (ADEs) represent one of the most prevalent types of health-care-related harm, and there is substantial room for improvement in the way that they are currently predicted and detected. We conducted a scoping review to identify key use cases in which artificial intelligence (AI) could be leveraged to reduce the frequency of ADEs. We focused on modern machine learning techniques and natural language processing. 78 articles were included in the scoping review. Studies were heterogeneous and applied various AI techniques covering a wide range of medications and ADEs. We identified several key use cases in which AI could contribute to reducing the frequency and consequences of ADEs, through prediction to prevent ADEs and early detection to mitigate the effects. Most studies (73 [94%] of 78) assessed technical algorithm performance, and few studies evaluated the use of AI in clinical settings. Most articles (58 [74%] of 78) were published within the past 5 years, highlighting an emerging area of study. Availability of new types of data, such as genetic information, and access to unstructured clinical notes might further advance the field.

A novel fusion-based deep learning model for sentiment analysis of COVID-19 tweets

Undoubtedly, coronavirus (COVID-19) has caused one of the biggest challenges of all times. The ongoing COVID-19 pandemic has caused more than 150 million infected cases and one million deaths globally as of May 5, 2021. Understanding the sentiment of people expressed in their social media comments can help in monitoring, controlling, and ultimately eradicating the disease. This is a sensitive matter as the threat of infectious disease significantly affects the way people think and behave in various ways. In this study, we proposed a novel method based on the fusion of four deep learning and one classical supervised machine learning model for sentiment analysis of coronavirus-related tweets from eight countries. Also, we analyzed coronavirus-related searches using Google Trends to better understand the change in the sentiment pattern at different times and places. Our findings reveal that the coronavirus attracted the attention of people from different countries at different times in varying intensities. Also, the sentiment in their tweets is correlated to the news and events that occurred in their countries including the number of newly infected cases, number of recoveries and deaths. Moreover, common sentiment patterns can be observed in various countries during the spread of the virus. We believe that different social media platforms have great impact on raising people's awareness about the importance of this disease as well as promoting preventive measures among people in the community.

Machine Learning: An Overview and Applications in Pharmacogenetics

This narrative review aims to provide an overview of the main Machine Learning (ML) techniques and their applications in pharmacogenetics (such as antidepressant, anti-cancer and warfarin drugs) over the past 10 years. ML deals with the study, the design and the development of algorithms that give computers capability to learn without being explicitly programmed. ML is a sub-field of artificial intelligence, and to date, it has demonstrated satisfactory performance on a wide range of tasks in biomedicine. According to the final goal, ML can be defined as Supervised (SML) or as Unsupervised (UML). SML techniques are applied when prediction is the focus of the research. On the other hand, UML techniques are used when the outcome is not known, and the goal of the research is unveiling the underlying structure of the data. The increasing use of sophisticated ML algorithms will likely be instrumental in improving knowledge in pharmacogenetics.

Use of metamodels for rapid discovery of narrow bandgap oxide photocatalysts

New photocatalysts are traditionally identified through trial-and-error methods. Machine learning has shown considerable promise for improving the efficiency of photocatalyst discovery from a large potential pool. Here, we describe a multi-step, target-driven consensus method using a stacking meta-learning algorithm that robustly predicts bandgaps and H2 evolution activities of photocatalysts. Trained on small datasets, these models can rapidly screen a large space (>10 million materials) to identify promising, non-toxic compounds as candidate water splitting photocatalysts. Two effective compounds and two controls possessing optimal bandgap values (∼2 eV) but not photoactivity as predicted by the models were synthesized. Their experimentally measured bandgaps and H2 evolution activities were consistent with the predictions. Conspicuously, the two compounds with strong photoactivities under UV and visible light are promising visible-light-driven water splitting photocatalysts. This study demonstrates the power of machine learning and the potential of big data to accelerate discovery of next-generation photocatalysts.

Gene Polymorphisms of the Renin-Angiotensin System and Bleeding Complications of Warfarin: Genetic-Based Machine Learning Models

This study aimed to investigate the effects of genetic variants and haplotypes in the renin–angiotensin system (RAS) on the risk of warfarin-induced bleeding complications at therapeutic international normalized ratios (INRs). Four single nucleotide polymorphisms (SNPs) of AGT, two SNPs of REN, three SNPs of ACE, four SNPs of AGTR1, and one SNP of AGTR2, in addition to VKORC1 and CYP2C9 variants, were investigated. We utilized logistic regression and several machine learning methods for bleeding prediction. The study included 142 patients, among whom 21 experienced bleeding complications. We identified a haplotype, H2 (TCG), carrying three single nucleotide polymorphisms (SNPs) of ACE (rs1800764, rs4341, and rs4353), which showed a significant relation with bleeding complications. After adjusting covariates, patients with H2/H2 experienced a 0.12-fold (95% CI 0.02–0.99) higher risk of bleeding complications than the others. In addition, G allele carriers of AGT rs5050 and A allele carriers of AGTR1 rs2640543 had 5.0- (95% CI 1.8–14.1) and 3.2-fold (95% CI 1.1–8.9) increased risk of bleeding complications compared with the TT genotype and GG genotype carriers, respectively. The AUROC values (mean, 95% CI) across 10 random iterations using five-fold cross-validated multivariate logistic regression, elastic net, random forest, support vector machine (SVM)–linear kernel, and SVM–radial kernel models were 0.732 (0.694–0.771), 0.741 (0.612–0.870), 0.723 (0.589–0.857), 0.673 (0.517–0.828), and 0.680 (0.528–0.832), respectively. The highest quartile group (≥75th percentile) of weighted risk score had approximately 12.0 times (95% CI 3.1–46.7) increased risk of bleeding, compared to the 25–75th percentile group, respectively. This study demonstrated that RAS-related polymorphisms, including the H2 haplotype of the ACE gene, could affect bleeding complications during warfarin treatment for patients with mechanical heart valves. Our results could be used to develop individually tailored intervention strategies to prevent warfarin-induced bleeding.

A two level learning model for authorship authentication

Nowadays, forensic authorship authentication plays a vital role in identifying the number of unknown authors as a result of the world's rapidly rising internet use. This paper presents two-level learning techniques for authorship authentication. The learning technique is supplied with linguistic knowledge, statistical features, and vocabulary features to enhance its efficiency instead of learning only. The linguistic knowledge is represented through lexical analysis features such as part of speech. In this study, a two-level classifier has been presented to capture the best predictive performance for identifying authorship. The first classifier is based on vocabulary features that detect the frequency with which each author uses certain words. This classifier's results are fed to the second one which is based on a learning technique. It depends on lexical, statistical and linguistic features. All of the three sets of features describe the author's writing styles in numerical forms. Through this work, many new features are proposed for identifying the author's writing style. Although, the proposed new methodology is tested for Arabic writings, it is general and can be applied to any language. According to the used machine learning models, the experiment carried out shows that the trained two-level classifier achieves an accuracy ranging from 94% to 96.16%.

Comparison of multivariate linear regression and a machine learning algorithm developed for prediction of precision warfarin dosing in a Korean population

BACKGROUND

Personalized warfarin dosing is influenced by various factors including genetic and non-genetic factors. Multiple linear regression (LR) is known as a conventional method to develop predictive models. Recently, machine learning approaches have been extensively implemented for warfarin dosing due to the hypothesis of non-linear association between covariates and stable warfarin dose.

OBJECTIVE

To extend the multiple linear regression algorithm for personalized warfarin dosing in a Korean population and compare with a machine learning--based algorithm.

METHOD

From this cohort study, we collected information on 650 patients taking warfarin who achieved steady state including demographic information, indications, comorbidities, comedications, habits, and genetic factors. The dataset was randomly split into training set (90%) and test set (10%). The LR and machine learning (gradient boosting machine [GBM]) models were developed on the training set and were evaluated on the test set.

RESULT

LR and GBM models were comparable in terms of accuracy of ideal dose (75.38% and 73.85%), correlation (0.77 and 0.73), mean absolute error (0.58 mg/day and 0.64 mg/day), and root mean square error (0.82 mg/day and 0.9 mg/day), respectively. VKORC1 genotype, CYP2C9 genotype, age, and weight were the highest contributors and could obtain 80% of maximum performance in both models.

CONCLUSION

This study shows that our LR and GMB models are satisfactory to predict warfarin dose in our dataset. Both models showed similar performance and feature contribution characteristics. LR may be the appropriate model due to its simplicity and interpretability.

Health Condition Estimation of Bearings with Multiple Faults by a Composite Learning-Based Approach

Bearings are critical components found in most rotating machinery; their health condition is of immense importance to many industries. The varied conditions and environments in which bearings operate make them prone to single and multiple faults. Widespread interest in the improvements of single fault diagnosis meant limited attention was spent on multiple fault diagnosis. However, multiple fault diagnosis poses extra challenges due to the submergence of the weak fault by the strong fault, presence of non-Gaussian noise, coupling of the frequency components, etc. A number of existing convolutional neural network models operate on a distinct feature that is not enough to assure reliable results in the presence of these challenges. In this paper, extended feature sets in three homogenous deep learning models are used for multiple fault diagnosis. This ensures a measure of diversity is introduced to the health management dataset to obtain complementary solutions from the models. The outputs of the models are fused through blending ensemble learning. Experiments using vibration datasets based on bearing multiple faults show an accuracy of 98.54%, with an improvement of 2.74% in the overall effectiveness over the single models. Compared with other technologies, the results show that this approach provides an improved generalized diagnostic capability.

A post-hoc interpretable ensemble model to feature effect analysis in warfarin dose prediction for Chinese patients

To interprete the importance of clinical features and genotypes for warfarin daily dose prediction, we developed a post-hoc interpretable framework based on an ensemble predictive model. This framework includes permutation importance for global interpretation and local interpretable model-agnostic explanation (LIME) and shapley additive explanations (SHAP) for local explanation. The permutation importance globally ranks the importance of features on the whole data set. This can guide us to build a predictive model with less variables and the complexity of final predictive model can be reduced. LIME and SHAP together explain how the predictive model give the predicted dosage for specific samples. This help clinicians prescribe accurate doses to patients using more effective clinical variables. Results showed that both the permutation importance and SHAP demonstrated that VKORC1, age, serum creatinine (SCr), left atrium (LA) size, CYP2C9 and weight were the most important features on the whole data set. In specific samples, both SHAP and LIME discovered that in Chinese patients, wild-type VKORC1-AA, mutant-type CYP2C9*3, age over 60, abnormal LA size, SCr within the normal range, and using amiodarone definitely required dosage reduction, whereas mutant-type VKORC1-AG/GG, small age, SCr out of normal range, normal LA size, diabetes and heavy weight required dosage enhancement.

Integration of machine learning and genome-scale metabolic modeling identifies multi-omics biomarkers for radiation resistance

Resistance to ionizing radiation, a first-line therapy for many cancers, is a major clinical challenge. Personalized prediction of tumor radiosensitivity is not currently implemented clinically due to insufficient accuracy of existing machine learning classifiers. Despite the acknowledged role of tumor metabolism in radiation response, metabolomics data is rarely collected in large multi-omics initiatives such as The Cancer Genome Atlas (TCGA) and consequently omitted from algorithm development. In this study, we circumvent the paucity of personalized metabolomics information by characterizing 915 TCGA patient tumors with genome-scale metabolic Flux Balance Analysis models generated from transcriptomic and genomic datasets. Metabolic biomarkers differentiating radiation-sensitive and -resistant tumors are predicted and experimentally validated, enabling integration of metabolic features with other multi-omics datasets into ensemble-based machine learning classifiers for radiation response. These multi-omics classifiers show improved classification accuracy, identify clinical patient subgroups, and demonstrate the utility of personalized blood-based metabolic biomarkers for radiation sensitivity. The integration of machine learning with genome-scale metabolic modeling represents a significant methodological advancement for identifying prognostic metabolite biomarkers and predicting radiosensitivity for individual patients.

QSAR Models for Active Substances against Pseudomonas aeruginosa Using Disk-Diffusion Test Data

Pseudomonas aeruginosa is a Gram-negative bacillus included among the six “ESKAPE” microbial species with an outstanding ability to “escape” currently used antibiotics and developing new antibiotics against it is of the highest priority. Whereas minimum inhibitory concentration (MIC) values against Pseudomonas aeruginosa have been used previously for QSAR model development, disk diffusion results (inhibition zones) have not been apparently used for this purpose in the literature and we decided to explore their use in this sense. We developed multiple QSAR methods using several machine learning algorithms (support vector classifier, K nearest neighbors, random forest classifier, decision tree classifier, AdaBoost classifier, logistic regression and naïve Bayes classifier). We used four sets of molecular descriptors and fingerprints and three different methods of data balancing, together with the “native” data set. In total, 32 models were built for each set of descriptors or fingerprint and balancing method, of which 28 were selected and stacked to create meta-models. In terms of balanced accuracy, the best performance was provided by KNN, logistic regression and decision tree classifier, but the ensemble method had slightly superior results in nested cross-validation.

A machine learning approach to personalized dose adjustment of lamotrigine using noninvasive clinical parameters

The pharmacokinetic variability of lamotrigine (LTG) plays a significant role in its dosing requirements. Our goal here was to use noninvasive clinical parameters to predict the dose-adjusted concentrations (C/D ratio) of LTG based on machine learning (ML) algorithms. A total of 1141 therapeutic drug-monitoring measurements were used, 80% of which were randomly selected as the "derivation cohort" to develop the prediction algorithm, and the remaining 20% constituted the "validation cohort" to test the finally selected model. Fifteen ML models were optimized and evaluated by tenfold cross-validation on the "derivation cohort,” and were filtered by the mean absolute error (MAE). On the whole, the nonlinear models outperformed the linear models. The extra-trees’ regression algorithm delivered good performance, and was chosen to establish the predictive model. The important features were then analyzed and parameters of the model adjusted to develop the best prediction model, which accurately described the C/D ratio of LTG, especially in the intermediate-to-high range (≥ 22.1 μg mL⁻¹ g⁻¹ day), as illustrated by a minimal bias (mean relative error (%) =  + 3%), good precision (MAE = 8.7 μg mL⁻¹ g⁻¹ day), and a high percentage of predictions within ± 20% of the empirical values (60.47%). This is the first study, to the best of our knowledge, to use ML algorithms to predict the C/D ratio of LTG. The results here can help clinicians adjust doses of LTG administered to patients to minimize adverse reactions.

StackGenVis: Alignment of Data, Algorithms, and Models for Stacking Ensemble Learning Using Performance Metrics

In machine learning (ML), ensemble methods-such as bagging, boosting, and stacking-are widely-established approaches that regularly achieve top-notch predictive performance. Stacking (also called "stacked generalization") is an ensemble method that combines heterogeneous base models, arranged in at least one layer, and then employs another metamodel to summarize the predictions of those models. Although it may be a highly-effective approach for increasing the predictive performance of ML, generating a stack of models from scratch can be a cumbersome trial-and-error process. This challenge stems from the enormous space of available solutions, with different sets of data instances and features that could be used for training, several algorithms to choose from, and instantiations of these algorithms using diverse parameters (i.e., models) that perform differently according to various metrics. In this work, we present a knowledge generation model, which supports ensemble learning with the use of visualization, and a visual analytics system for stacked generalization. Our system, StackGenVis, assists users in dynamically adapting performance metrics, managing data instances, selecting the most important features for a given data set, choosing a set of top-performant and diverse algorithms, and measuring the predictive performance. In consequence, our proposed tool helps users to decide between distinct models and to reduce the complexity of the resulting stack by removing overpromising and underperforming models. The applicability and effectiveness of StackGenVis are demonstrated with two use cases: a real-world healthcare data set and a collection of data related to sentiment/stance detection in texts. Finally, the tool has been evaluated through interviews with three ML experts.

Identifying Parkinson's disease with mild cognitive impairment by using combined MR imaging and electroencephalogram

OBJECTIVES

To analyse the changes of quantitative electroencephalogram (qEEG) and cortex structural magnetic resonance (MR) imaging in Parkinson's disease with mild cognitive impairment (PD-MCI) and to explore the "composite marker"-based machine learning model in identifying PD-MCI.

METHODS

Retrospective analysis of patients with PD identified 36 PD-MCI and 35 PD with normal cognition (PD-NC). QEEG features of power spectrum and structural MR features of cortex based on surface-based morphometry (SBM) were extracted. Support vector machine (SVM) was established using combined features of structural MR and qEEG to identify PD-MCI. Feature importance evaluation algorithm of mean impact value (MIV) was established to sort the vital characteristics of qEEG and structural MR.

RESULTS

Compared with PD-NC, PD-MCI showed a statistically significant difference in 5 leads and waves of qEEG and 7 cortical region features of structural MR. The SVM model based on these qEEG and structural MR features yielded an accuracy of 0.80 in the training set and had a high prediction accuracy of 0.80 in the test set (sensitivity was 0.78, specificity was 0.83, area under the receiver operating characteristic curve was 0.77), which was higher than the model built by the feature separately. QEEG features of theta wave in C3 had a marked impact on the model for classification according to the MIV algorithm.

CONCLUSIONS

PD-MCI is characterized by widespread structural and EEG abnormality. "Composite markers" could be valuable for the individualized diagnosis of PD-MCI by machine learning.

KEY POINTS

• Explore the brain abnormalities in Parkinson's disease with mild cognitive impairment by using the quantitative electroencephalogram and cortex structural MR simultaneously. • Multimodal features based support vector machine for identifying Parkinson's disease with mild cognitive impairment has an acceptable performance. • Theta wave in C3 is the most influential feature of qEEG and cortex structure MR imaging in identifying Parkinson's disease with mild cognitive impairment using support vector machine.

Evaluating warfarin dosing models on multiple datasets with a novel software framework and evolutionary optimisation

Warfarin is an effective preventative treatment for arterial and venous thromboembolism, but requires individualised dosing due to its narrow therapeutic range and high individual variation. Many machine learning techniques have been demonstrated in this domain. This study evaluated the accuracy of the most promising algorithms on the International Warfarin Pharmacogenetics Consortium dataset and a novel clinical dataset of South African patients. Support vectors and linear regression were amongst the top performers in both datasets and performed comparably to recent stacked ensemble approaches, whilst neural networks were one of the worst performers in both datasets. We also introduced genetic programming to automatically optimise model architectures and hyperparameters without human guidance. Remarkably, the generated models were found to match the performance of the best models hand-crafted by human experts. Finally, we present a novel software framework (Warfit-learn) for warfarin dosing research. It leverages the most successful techniques in preprocessing, imputation, and parallel evaluation, with the goal of accelerating research and making results in this domain more reproducible.

Country-level pandemic risk and preparedness classification based on COVID-19 data: A machine learning approach

In this work we present a three-stage Machine Learning strategy to country-level risk classification based on countries that are reporting COVID-19 information. A K% binning discretisation (K = 25) is used to create four risk groups of countries based on the risk of transmission (coronavirus cases per million population), risk of mortality (coronavirus deaths per million population), and risk of inability to test (coronavirus tests per million population). The four risk groups produced by K% binning are labelled as 'low', 'medium-low', 'medium-high', and 'high'. Coronavirus-related data are then removed and the attributes for prediction of the three types of risk are given as the geopolitical and demographic data describing each country. Thus, the calculation of class label is based on coronavirus data but the input attributes are country-level information regardless of coronavirus data. The three four-class classification problems are then explored and benchmarked through leave-one-country-out cross validation to find the strongest model, producing a Stack of Gradient Boosting and Decision Tree algorithms for risk of transmission, a Stack of Support Vector Machine and Extra Trees for risk of mortality, and a Gradient Boosting algorithm for the risk of inability to test. It is noted that high risk for inability to test is often coupled with low risks for transmission and mortality, therefore the risk of inability to test should be interpreted first, before consideration is given to the predicted transmission and mortality risks. Finally, the approach is applied to more recent risk levels to data from September 2020 and weaker results are noted due to the growth of international collaboration detracting useful knowledge from country-level attributes which suggests that similar machine learning approaches are more useful prior to situations later unfolding.

Empowering Predictive Maintenance: A Hybrid Method to Diagnose Abnormal Situations

Aerospace systems are composed of hundreds or thousands of components and complex subsystems which need an appropriate health monitoring capability to enable safe operation in various conditions. In terms of monitoring systems, it is possible to find a considerable number of state-of-the-art works in the literature related to ad-hoc solutions. Still, it is challenging to reuse them even with subtle differences in analogous subsystems or components. This paper proposes the Generic Anomaly Detection Hybridization Algorithm (GADHA) aiming to build a more reusable algorithm to support anomaly detection. The solution consists of analyzing different supervised machine learning classification algorithms combined in ensemble techniques, with a physical model when available, and two levels of a decision to estimate the current state of the monitored system. Finally, the proposed algorithm assures at least equal, or, more typically, better, overall accuracy in fault detection and isolation than the application of such algorithms alone, through few adaptations.

How Will Machine Learning Inform the Clinical Care of Atrial Fibrillation?

Machine learning applications in cardiology have rapidly evolved in the past decade. With the availability of machine learning tools coupled with vast data sources, the management of atrial fibrillation (AF), a common chronic disease with significant associated morbidity and socioeconomic impact, is undergoing a knowledge and practice transformation in the increasingly complex healthcare environment. Among other advances, deep-learning machine learning methods, including convolutional neural networks, have enabled the development of AF screening pathways using the ubiquitous 12-lead ECG to detect asymptomatic paroxysmal AF in at-risk populations (such as those with cryptogenic stroke), the refinement of AF and stroke prediction schemes through comprehensive digital phenotyping using structured and unstructured data abstraction from the electronic health record or wearable monitoring technologies, and the optimization of treatment strategies, ranging from stroke prophylaxis to monitoring of antiarrhythmic drug (AAD) therapy. Although the clinical and population-wide impact of these tools continues to be elucidated, such transformative progress does not come without challenges, such as the concerns about adopting black box technologies, assessing input data quality for training such models, and the risk of perpetuating rather than alleviating health disparities. This review critically appraises the advances of machine learning related to the care of AF thus far, their potential future directions, and its potential limitations and challenges.

AIBH: Accurate Identification of Brain Hemorrhage Using Genetic Algorithm Based Feature Selection and Stacking

Brain hemorrhage is a type of stroke which is caused by a ruptured artery, resulting in localized bleeding in or around the brain tissues. Among a variety of imaging tests, a computerized tomography (CT) scan of the brain enables the accurate detection and diagnosis of a brain hemorrhage. In this work, we developed a practical approach to detect the existence and type of brain hemorrhage in a CT scan image of the brain, called Accurate Identification of Brain Hemorrhage, abbreviated as AIBH. The steps of the proposed method consist of image preprocessing, image segmentation, feature extraction, feature selection, and design of an advanced classification framework. The image preprocessing and segmentation steps involve removing the skull region from the image and finding out the region of interest (ROI) using Otsu’s method, respectively. Subsequently, feature extraction includes the collection of a comprehensive set of features from the ROI, such as the size of the ROI, centroid of the ROI, perimeter of the ROI, the distance between the ROI and the skull, and more. Furthermore, a genetic algorithm (GA)-based feature selection algorithm is utilized to select relevant features for improved performance. These features are then used to train the stacking-based machine learning framework to predict different types of a brain hemorrhage. Finally, the evaluation results indicate that the proposed predictor achieves a 10-fold cross-validation (CV) accuracy (ACC), precision (PR), Recall, F1-score, and Matthews correlation coefficient (MCC) of 99.5%, 99%, 98.9%, 0.989, and 0.986, respectively, on the benchmark CT scan dataset. While comparing AIBH with the existing state-of-the-art classification method of the brain hemorrhage type, AIBH provides an improvement of 7.03%, 7.27%, and 7.38% based on PR, Recall, and F1-score, respectively. Therefore, the proposed approach considerably outperforms the existing brain hemorrhage classification approach and can be useful for the effective prediction of brain hemorrhage types from CT scan images (The code and data can be found here: http://cs.uno.edu/~tamjid/Software/AIBH/code_data.zip).

Machine Learning Algorithm for Predicting Warfarin Dose in Caribbean Hispanics Using Pharmacogenetic Data

Despite some previous examples of successful application to the field of pharmacogenomics, the utility of machine learning (ML) techniques for warfarin dose predictions in Caribbean Hispanic patients has yet to be fully evaluated. This study compares seven ML methods to predict warfarin dosing in Caribbean Hispanics. This is a secondary analysis of genetic and non-genetic clinical data from 190 cardiovascular Hispanic patients. Seven ML algorithms were applied to the data. Data was divided into 80 and 20% to be used as training and test sets. ML algorithms were trained with the training set to obtain the models. Model performance was determined by computing the corresponding mean absolute error (MAE) and % patients whose predicted optimal dose were within ±20% of the actual stabilization dose, and then compared between groups of patients with “normal” (i.e., > 21 but <49 mg/week), low (i.e., ≤21 mg/week, “sensitive”), and high (i.e., ≥49 mg/week, “resistant”) dose requirements. Random forest regression (RFR) significantly outperform all other methods, with a MAE of 4.73 mg/week and 80.56% of cases within ±20% of the actual stabilization dose. Among those with “normal” dose requirements, RFR performance is also better than the rest of models (MAE = 2.91 mg/week). In the “sensitive” group, support vector regression (SVR) shows superiority over the others with lower MAE of 4.79 mg/week. Finally, multivariate adaptive splines (MARS) shows the best performance in the resistant group (MAE = 7.22 mg/week) and 66.7% of predictions within ±20%. Models generated by using RFR, MARS, and SVR algorithms showed significantly better predictions of weekly warfarin dosing in the studied cohorts than other algorithms. Better performance of the ML models for patients with “normal,” “sensitive,” and “resistant” to warfarin were obtained when compared to other populations and previous statistical models.

Clinical Model for Predicting Warfarin Sensitivity

Warfarin is a widely used anticoagulant with a narrow therapeutic index and large interpatient variability in the therapeutic dose. Complications from inappropriate warfarin dosing are one of the most common reasons for emergency room visits. Approximately one third of warfarin dose variability results from common genetic variants. Therefore, it is very necessary to recognize warfarin sensitivity in individuals caused by genetic variants. Based on combined polymorphisms in CYP2C9 and VKORC1, we established a clinical classification for warfarin sensitivity. In the International Warfarin Pharmacogenetic Consortium (IWPC) with 5542 patients, we found that 95.1% of the Black in the IWPC cohort were normal warfarin responders, while 74.8% of the Asian were warfarin sensitive (P < 0.001). Moreover, we created a clinical algorithm to predict warfarin sensitivity in individual patients using logistic regression. Compared to a fixed-dose approach, the clinical algorithm provided significantly better performance. In addition, we validated the derived clinical algorithm using the external Easton cohort with 106 chronic warfarin users. The AUC was 0.836 vs. 0.867 for the Easton cohort and the IWPC cohort, respectively. With the use of this algorithm, it is very likely to facilitate patient care regarding warfarin therapy, thereby improving clinical outcomes.