Reliability of a Bayesian network to predict an elevated aldosterone-to-renin ratio

Prediction Tool to Estimate Potassium Diet in Chronic Kidney Disease Patients Developed Using a Machine Learning Tool: The UniverSel Study

There is a need for a reliable and validated method to estimate dietary potassium intake in chronic kidney disease (CKD) patients to improve prevention of cardiovascular complications. This study aimed to develop a clinical tool to estimate potassium intake using 24-h urinary potassium excretion as a surrogate of dietary potassium intake in this high-risk population. Data of 375 adult CKD-patients routinely collecting their 24-h urine were included to develop a prediction tool to estimate potassium diet. The prediction tool was built from a random sample of 80% of patients and validated on the remaining 20%. The accuracy of the prediction tool to classify potassium diet in the three classes of potassium excretion was 74%. Surprisingly, the variables related to potassium consumption were more related to clinical characteristics and renal pathology than to the potassium content of the ingested food. Artificial intelligence allowed to develop an easy-to-use tool for estimating patients' diets in clinical practice. After external validation, this tool could be extended to all CKD-patients for a better clinical and therapeutic management for the prevention of cardiovascular complications.

Prediction of Autonomy Loss in Alzheimer’s Disease

The evolution of functional autonomy loss leads to institutionalization of people affected by Alzheimer’s disease (AD), to an alteration of their quality of life and that of their caregivers. To predict loss of functional autonomy could optimize prevention strategies, aids and cost of care. The aim of this study was to develop and to cross-validate a model to predict loss of functional autonomy as assessed by Instrumental Activities of Daily Living (IADL) score. Outpatients with probable AD and with 2 or more visits to the Clinical and Research Memory Centre of the University Hospital were included. Four Tree-Augmented Naïve bayesian networks (6, 12, 18 and 24 months of follow-up) were built. Variables included in the model were demographic data, IADL score, MMSE score, comorbidities, drug prescription (psychotropics and AD-specific drugs). A 10-fold cross-validation was conducted to evaluate robustness of models. The study initially included 485 patients in the prospective cohort. The best performance after 10-fold cross-validation was obtained with the model able to predict loss of functional autonomy at 18 months (area under the curve of the receiving operator characteristic curve = 0.741, 27% of patients misclassified, positive predictive value = 77% and negative predictive value = 73%). The 13 variables used explain 41.6% of the evolution of functional autonomy at 18 months. A high-performing predictive model of AD evolution of functional autonomy was obtained. An external validation is needed to use the model in clinical routine so as to optimize the patient care.

A comprehensive scoping review of Bayesian networks in healthcare: Past, present and future

No comprehensive review of Bayesian networks (BNs) in healthcare has been published in the past, making it difficult to organize the research contributions in the present and identify challenges and neglected areas that need to be addressed in the future. This unique and novel scoping review of BNs in healthcare provides an analytical framework for comprehensively characterizing the domain and its current state. A literature search of health and health informatics literature databases using relevant keywords found 3810 articles that were reduced to 123. This was after screening out those presenting Bayesian statistics, meta-analysis or neural networks, as opposed to BNs and those describing the predictive performance of multiple machine learning algorithms, of which BNs were simply one type. Using the novel analytical framework, we show that: (1) BNs in healthcare are not used to their full potential; (2) a generic BN development process is lacking; (3) limitations exist in the way BNs in healthcare are presented in the literature, which impacts understanding, consensus towards systematic methodologies, practice and adoption; and (4) a gap exists between having an accurate BN and a useful BN that impacts clinical practice. This review highlights several neglected issues, such as restricted aims of BNs, ad hoc BN development methods, and the lack of BN adoption in practice and reveals to researchers and clinicians the need to address these problems. To map the way forward, the paper proposes future research directions and makes recommendations regarding BN development methods and adoption in practice.

A Clinical Decision Support Tool to Find the Best Initial Intravenous Cyclosporine Regimen in Pediatric Hematopoietic Stem Cell Transplantation

To optimize cyclosporine A (CsA) dosing regimen in pediatric patients undergoing hematopoietic stem cell transplantation (HSCT), we aimed to provide clinicians with a validated decision support tool for determining the most suitable first dose of intravenous CsA. We used a 10-year monocentric data set of pediatric patients undergoing HSCT. Discretization of all variables was performed according to literature or thanks to algorithms using Shannon entropy (from information theory) or equal width intervals. The first 8 years were used to build the Bayesian network model. This model underwent a 10-fold cross-validation, and then a prospective validation with data of the last 2 years. There were 3.3% and 4.1% of missing values in the training and the validation data set, respectively. After prospective validation, the Tree-Augmented Naïve Bayesian network shows interesting prediction performances with an average area under the receiver operating characteristic curve of 0.804, 32.8% of misclassified patients, a true-positive rate of 0.672, and a false-positive rate of 0.285. This validated model allows good predictions to propose an optimized and personalized initial CsA dose for pediatric patients undergoing HSCT. The clinical impact of its use should be further evaluated.

Bayesian networks in healthcare: What is preventing their adoption?

There has been much research effort expended toward the use of Bayesian networks (BNs) in medical decision-making. However, because of the gap between developing an accurate BN and demonstrating its clinical usefulness, this has not resulted in any widespread BN adoption in clinical practice. This paper investigates this problem with the aim of finding an explanation and ways to address the problem through a comprehensive literature review of articles describing BNs in healthcare. Based on the literature collection that has been systematically narrowed down from 3810 to 116 most relevant articles, this paper analyses the benefits, barriers and facilitating factors (BBF) for implementing BN-based systems in healthcare using the ITPOSMO-BBF framework. A key finding is that works in the literature rarely consider barriers and even when these were identified they were not connected to facilitating factors. The main finding is that the barriers can be grouped into: (1) data inadequacies; (2) clinicians' resistance to new technologies; (3) lack of clinical credibility; (4) failure to demonstrate clinical impact; (5) absence of an acceptable predictive performance; and (6) absence of evidence for model's generalisability. The facilitating factors can be grouped into: (1) data collection improvements; (2) software and technological improvements; (3) having interpretable and easy to use BN-based systems; (4) clinical involvement in the development or review of the model; (5) investigation of model's clinical impact; (6) internal validation of the model's performance; and (7) external validation of the model. These groupings form a strong basis for a generic framework that could be used for formulating strategies for ensuring BN-based clinical decision-support system adoption in frontline care settings. The output of this review is expected to enhance the dialogue among researchers by providing a deeper understanding for the neglected issue of BN adoption in practice and promoting efforts for implementing BN-based systems.

A maximum likelihood approach to electronic health record phenotyping using positive and unlabeled patients

OBJECTIVE

Phenotyping patients using electronic health record (EHR) data conventionally requires labeled cases and controls. Assigning labels requires manual medical chart review and therefore is labor intensive. For some phenotypes, identifying gold-standard controls is prohibitive. We developed an accurate EHR phenotyping approach that does not require labeled controls.

MATERIALS AND METHODS

Our framework relies on a random subset of cases, which can be specified using an anchor variable that has excellent positive predictive value and sensitivity independent of predictors. We proposed a maximum likelihood approach that efficiently leverages data from the specified cases and unlabeled patients to develop logistic regression phenotyping models, and compare model performance with existing algorithms.

RESULTS

Our method outperformed the existing algorithms on predictive accuracy in Monte Carlo simulation studies, application to identify hypertension patients with hypokalemia requiring oral supplementation using a simulated anchor, and application to identify primary aldosteronism patients using real-world cases and anchor variables. Our method additionally generated consistent estimates of 2 important parameters, phenotype prevalence and the proportion of true cases that are labeled.

DISCUSSION

Upon identification of an anchor variable that is scalable and transferable to different practices, our approach should facilitate development of scalable, transferable, and practice-specific phenotyping models.

CONCLUSIONS

Our proposed approach enables accurate semiautomated EHR phenotyping with minimal manual labeling and therefore should greatly facilitate EHR clinical decision support and research.

Development of a clinical decision tool to reduce diagnostic testing for primary aldosteronism in patients with difficult-to-control hypertension

BACKGROUND

Satisfactory tools to preclude low-risk patients from intensive diagnostic testing for primary aldosteronism (PA) are lacking. Therefore, we aimed to develop a decision tool to determine which patients with difficult-to-control hypertension have a low probability of PA, thereby limiting the exposure to invasive testing while at the same time increasing the efficiency of testing in the remaining patients.

METHODS

Data from consecutive patients with difficult-to-control hypertension, analysed through a standardized diagnostic protocol between January 2010 and October 2017 (n = 824), were included in this cross-sectional study. PA was diagnosed by a combined approach: 1) elevated aldosterone-to-renin ratio (> 5.0 pmol/fmol/s), confirmed with 2) non-suppressible aldosterone after standardized saline infusion (≥280 pmol/L). Multivariable logistic regression analyses including seven pre-specified clinical variables (age, systolic blood pressure, serum potassium, potassium supplementation, serum sodium, eGFR and HbA1c) was performed. After correction for optimism, test reliability, discriminative performance and test characteristics were determined.

RESULTS

PA was diagnosed in 40 (4.9%) of 824 patients. Predicted probabilities of PA agreed well with observed frequencies and the c-statistic was 0.77 (95% confidence interval (95%CI) 0.70-0.83). Predicted probability cut-off values of 1.0-2.5% prevented unnecessary testing in 8-32% of the patients with difficult-to-control hypertension, carrying sensitivities of 0.98 (95%CI 0.96-0.99) and 0.92 (0.83-0.97), and negative predictive values of 0.99 (0.98-1.00) and 0.99 (0.97-0.99).

CONCLUSIONS

With a decision tool, based on seven easy-to-measure clinical variables, patients with a low probability of PA can be reliably selected and a considerable proportion of patients with difficult-to-control hypertension can be spared intensive diagnostic testing.

Variables linked to hepatitis B vaccination success in non-dialyzed chronic kidney disease patients: Use of a bayesian model

BACKGROUND

Hepatitis B vaccination is recommended for chronic kidney disease (CKD) patients before starting dialysis. We performed an analyis aimed to describe the clinical and biological parameters related to the success of vaccination in CKD patients before starting dialysis.

METHODS

We extracted data of 170 non-dialyzed patients who were offered hepatitis B vaccination from a register. They received a first vaccination of 40μg followed by boosters after one, two and six months. Patients were considered protected if their hepatitis B antibody level was >10IU/L, three months apart. A logistic regression and a Bayesian model were used to describe the relationships between variables and the success of vaccination.

RESULTS

Vaccination protected 50.6% of the patients. Model adjustment to the data was higher using the Bayesian model compared to the logistic regression (with area under the ROC curve of 0.955±0.007 vs 0.775±0.066 respectively). The Bayesian model's robustness studied using a 10 fold cross validation showed a percentage of misclassified subjects of 12.4±1.8%, a sensitivity of 87.7±0.3%, a specificity of 87.5±0.3%, a positive predictive value of 87.8±0.3% and negative predictive value of 87.4±0.2%. As classified by the Bayesian model, the variables most related to successful vaccination were, in descending order: age, eGFR, protidemia, albuminemia, cause of renal failure, gender, previous vaccination and weight.

CONCLUSION

The Bayesian network confirmed that both kidney function and nutritional status of patients are important factors to explain the success of vaccination against hepatitis B in CKD patients before dialysis. For research purposes, before an external validation, the network can be used online at www.hed.cc/?s=Bhepatitis&n=ReseauhepatiteBsup10.neta.