The presented study explores the clustering of arterial oscillogram (AO) data among a sample of patients, focusing on ultra-low-frequency (ULF) indicators and their relationship with depression levels. Through dimensionality reduction using UMAP, two distinct classes emerged, categorized as lighter and more severe cases. Utilizing machine learning methods, an automated classifier was developed based on correlated ULF indicators, which led to improved classification accuracy. By incorporating ULF parameters, products of correlated parameters, and additional measured factors, the classifier achieved high reliability in estimating depression levels. Specifically, the nearest neighbors method yielded accuracies up to 0.9792. This research supports the creation of an automated diagnostic classification AI service capable of reliably estimating at least four levels of depression based on AO analysis.