Machine Learning-Based Models for Prediction of Subtype Diagnosis of Primary Aldosteronism Using Blood Test

Abstract

Abstract Context: Primary aldosteronism (PA) is a common cause of secondary hypertension and is associated with an increased risk of cardiometabolic diseases, especially in unilateral subtype. Despite its high prevalence, the case detection rate of PA is limited, partly because of no clinical models available in general practice to identify patients highly suspicious of unilateral subtype of PA, who can be cured by unilateral adrenalectomy and should be referred to specialized centers. The use of machine learning has been introduced in some fields of clinical prediction. Combining machine learning with clinical data could lead to the development of new models for predicting unilateral subtype of PA. Objective: The aim of this study was to develop a predictive model of subtype diagnosis of PA based on machine learning methods using clinical data available in general practice. Design and setting: This was a retrospective cross-sectional study in referral centers. Patients: We retrospectively analyzed 91 patients with unilateral and 138 patients with bilateral subtype of PA diagnosed according to adrenal venous sampling findings, and stratified randomly split to the training (80%) and test cohorts (20%). Four supervised machine learning classifiers; logistic regression, support vector machines, random forests (RF), and gradient boosting decision trees, were used to develop prediction models from 21 clinical variables. Classifiers were trained using stratified 10 fold cross-validation of the training cohort and hyperparameters of each classifier were adjusted using grid search to optimize the accuracy in the training cohort. Main Outcome Measures: The accuracy and the area under the receiver operating characteristic curve (AUC) for subtype prediction in the test cohort were compared among the optimized classifiers. Results: Of the four classifiers, the accuracy and AUC were highest in RF, with 95.7% and 0.990, respectively. Serum potassium levels, plasma aldosterone concentrations, and serum sodium levels were highlighted as important variables in this model. For feature-selected RF with the three variables, the accuracy and AUC were 89.1% and 0.950, respectively. With an independent external PA cohort, we confirmed a similar accuracy for feature-selected RF. Conclusions: Machine learning models developed using blood test can help predict subtype diagnosis of PA in general practice.