Burrello J, Burrello A, Pieroni J, et al. - Using patient clinical and biochemical characteristics, researchers sought to develop prediction models for subtype diagnosis of primary aldosteronism (PA). The sample consisted of patients referred to a tertiary hypertension unit. In a training (N = 150) and in an internal validation cohort (N = 65), respectively, diagnostic algorithms were built and tested. The models have been validated in an external independent cohort (N = 118). The diagnostic models included six parameters associated with LPA diagnosis (aldosterone at screening and after confirmatory testing, lowest potassium value, presence/absence of nodules, nodule diameter, and computed tomography results). Machine learning algorithms showed high accuracy at training and internal validation (79.1%-93%), while a 20-point score reached an area under the curve of 0.896, and a sensitivity/specificity of 91.7/79.3%. Diagnostic modelling techniques can be used in patients with PA in centers where AVS is not available for subtype diagnosis and to guide surgical decision.